berklee college - music production & engineering

RECORDING JULY 1999

The roads of Boston are famous for their random wa n-d e r i n g . Few streets intersect at right angles. I n d iv i d u a ls t reets change names, become one way, or dead end with-out wa r n i n g . The native s , not just the rent-a-car equippedt o u r i s t s , admit that it’s easy to get lost. A running jokes ays that somew h e re in this city you’ll re a ch an inters e c-tion while driving down a one way street and innocentlyencounter all three signs of doom at once: no left turn,no right turn, and do not enter.

M aybe there is a flashing red light to warn you area p p ro a ching this dreaded inters e c t i o n . And there is pro b-ab ly also a sign admonishing you to yield to pedestrians,as if your ability to make progress we re n ’t alre a dy limit-ed enough. N a t u ra l ly there are no signs telling you whats t reet you are on or what street you have re a ch e d .

The cars , most of them taxis, just line up. Your bloodp re s s u re rises. Your appointments ex p i re .You scream toyo u rs e l f, “ Why am I driving in this town any way ? ”

Without some fundamental understanding of how astudio is connected, you’ll eve n t u a l ly find yo u rself at theaudio equivalent of this intersection: feedback loopss c ream through the monitors , no fader seems capable ofturning down the vo c a l , drums rattle away in the head-phones but are n ’t getting to tape... I could go on. B e l i evem e . I could go on.

At the center of this mess is the mixing console (a.k.a.m i x e r, b o a rd , or desk). In the hands of a qualified engi-n e e r, it manages the flow of all audio signals, g e t t i n gthem to their appropriate destination safely and smooth-ly. The untrained user can expect to get lost, e n c o u n t e rfender benders , and eve n t u a l ly be para ly zed by gr i d l o ck .

The role of the mixerThe ultimate function of the console is to contro l ,

m a n i p u l a t e , and route all the various audio signals ra c-ing in and out of the different pieces of equipment in thestudio or synth ra ck—it provides the appropriate signalpath for the re c o rding task at hand.

Consider mixdow n . The signal flow goal of mixing is tocombine seve ral tra cks of music that have been oh-so-

c a re f u l ly re c o rded on a mu l t i t ra ck into two tra cks ofmusic that our friends, the radio stations, and the re c o rdbuying public can enjoy. Th ey all have stere o s , so we‘ c o nvert’ the mu l t i t ra ck re c o rding into stereo: 24 tra ck si n , t wo tra cks out. The mixer is the device that does this.

N a t u ra l ly, t h e re ’s a lot more to mixing than just com-bining the 24 tra cks into a nice sounding 2-tra ck mix. Fo rex a m p l e , we might also add reve r b . And equalization.And compre s s i o n . And a little turbo-auto-panning-flange-wah-distortion™ (patent pending. I t ’s just a little patchI’m wo rking on in the ol’ digital multi-effects box ) .

It is the mixing console’s job to provide the signal flows t r u c t u re that enables all these devices to be hooked upc o r re c t ly. It ensures that all the appropriate signals getto their destinations without running into any t h i n g . Ap r i m a ry function of the console is revealed: the mixermust be able to hook up any audio output to any audioi n p u t . See Fi g u re 1 for an example of the many possiblehookups you might expect your mixer to prov i d e .

In connecting any of these outputs to any of these inputs,the console is asked to make a nearly infinite number ofoptions possible.We mentioned mixdown as an ex a m p l eab ove , but we do more than mix. Our signal routing dev i c ehas to be able to configure the gear for re c o rding a bunchof signals to the mu l t i t ra ck re c o rder simu l t a n e o u s ly, l i kewhen we re c o rd a big band. It should also be able to makethe necessary signal flow adjustments re q u i red to permitan ove rdub on the mu l t i t ra ck .A dd i t i o n a l ly, we might needto re c o rd or broadcast live in stere o.

Fo r t u n a t e ly, all sessions fall into one of the fo l l ow i n gc a t e g o r i e s .

1. BasicsA mu l t i t ra ck re c o rding project begins with the basics

s e s s i o n . When doing the basics session, nothing is ontape ye t , lots of musicians are in the room play i n g , a n dthe engineer is ch a rged with the task of getting the firs tt ra cks onto tape.

You know how it goes. The band all plays tog e t h e r, a n dyou re c o rd them onto separate tra ck s . Of course the

Ladies and gentlemen: By now most of you know Alex Case’s writing very well (for insta n c e ,

he wrote that famous ‘The Snare—mastering the art of noise’ article in the January’issue).

W h at you may not know is that Alex is a busy engineer in the Boston area, that he teaches

music production and engineering at Berklee College of Music, and that he’s just the right

person to be starting our new beginner’s series.

Before turning the mighty pen over to Alex for this extended-mix opening column, let me just

reassure fellow fans of Mike Rivers’ ‘Oops Wrong Button’ that the series will continue in its

more advanced state. It’s simply time to start over from Square 1 so recording novices can

be brought up to speed.

And now without further ado I give you…Alex.—NB

Part 1: Consoles and ConnectionsPart 1: Consoles and Connections

RECORDING JULY 1999

singer will want to redo her part as an ove rdub later.Ditto for the guitarist. You still re c o rd eve ry t h i n g , a ssometimes the keeper take is the one that happens dur-ing basics. No pre s s u re , just sing/play along so the bandcan keep tra ck of which ve rse they are on, and we ’ l lre c o rd a more careful tra ck in a few we e k s .

S u ch freedom often leads to cre a t ivity and ch a n c e - t a k-i n g , key components of a great take . So you may one daybe glad you re c o rded the singer that day. Ditto for theg u i t a r i s t .

With the intent to do so many tra cks as ove rdubs latera ny way, the audio mission of the basics session isreduced to getting the killer drum and bass perfo r m a n c eonto the mu l t i t ra ck . And sometimes even the bass partgets deferred into an ove rd u b .

So for basics we re c o rd the entire band playing all atonce to get the drummer’s part on tape. C h e ck out theset-up sheet for a ve ry simple basics session. I t ’s just at r i o — d r u m s , b a s s , g u i t a r, and vocals—and yet we ’ ve gotat least 15 microphones going to at least ten tra ck s .

I say “at least” because it is easy to throw more mics onthese same instruments (e.g. c reate a more intere s t i n gguitar tone through the combination of seve ral differe n tkinds of mics in different locations around the guitara m p ) . And if you have enough tra ck s , it is tempting to useeven more tra cks (e.g. re c o rd the bass DI direct to themixer as a separate tra ck from the miked bass cab i n e t ) .

The console is in the center of all this, as shown inFi g u re 3. It routes all those mic signals to the mu l t i t ra ckso you can re c o rd them. It routes them to the monitors soyou can hear them. It routes those same signals to theheadphones so the band members can hear each other,the pro d u c e r, and the engineer. And it sends and re c e ive saudio to and from any number of signal pro c e s s o rs (moreis better): compre s s o rs , e q u a l i ze rs , reve r b s , e t c .

2. OverdubbingFor the ove rdubs there are often fewer musicians play-

i n g , f ewer microphones in action, and possibly fewer bandm e m b e rs aro u n d . It is often a mu ch calmer ex p e r i e n c e .

During basics there is the unspoken but stro n g lyimplied pre s s u re that no one can mess up or the wholet a ke will have to be stopped. The crowd in the studio isove r w h e l m i n g . The crowd in the control room is wa t ch-i n g . The lights, m e t e rs , mics and cables all over the placecomplete that “in the lab , under a micro s c o p e ” f e e l i n g .Pe r formance anxiety fills the studio of a basics session.

O ve rd u b s , on the other hand, a re as uncomplicated asa singer, a micro p h o n e , a pro d u c e r, and an engineer. D i mthe lights. R e l a x . Do a few practice runs. A ny mu s i c a lm i s t a kes tonight are just between us. No one else willhear them. We’ll erase them. If you don’t like it, just stopand we’ll try again.

M e a n t i m e , the console routes the mics to the mu l t i-t ra ck tape. The console creates the rough mix of the micsand the tra cks alre a dy on tape and sends them to them o n i t o rs . S i mu l t a n e o u s ly, it creates a separate mix fo rthe headphones. And we never miss an opportunity top a t ch in a compressor and some effects. Fi g u re 4 lays outthe console in ove rdub mode.

3. MixdownFor mixdow n , the engineer and producer use their

musical and technical abilities to the max, coaxing themost satisfying loudspeaker performance out of eve ry-thing the band re c o rd e d . Th e re is no limit to what mightbe attempted. Th e re is no limit to the amount of gearthat might be needed.

In case yo u ’ ve never seen what goes on in a big budgetpop mix, let me reveal an important fact: nearly eve ryt ra ck (and there are at least 24, p ro b ab ly many more )gets equalized and compressed and pro b ab ly gets a doseof reverb and/or some additional effects as we l l . A fewh u n d red patch cables are used. Perhaps seve ral tens,p ro b ab ly hundreds of thousands of dollars worth of out-b o a rd signal processing is used.

Automation is re q u i re d . And an enormous console isd e s i re d . During earlier re c o rding and ove rd u bbing ses-sions you might have thought, “ This is sounding like ah i t .” I t ’s not until mixdown when you’ll re a l ly feel it. I t ’snot until the gear- i n t e n s e , t ra ck by tra ck assembly of thetune that you’ll think, “ This sounds like a re c o rd ! ”

As Fi g u re 5 illustra t e s , the signal flow associated withm i x d own is actually quite stra i g h t fo r wa rd . Gone is theneed to handle microphone signals. Gone is the need toc reate a headphone mix. Nothing needs to be sent to themu l t i t ra ck . The mission is to route mu l t i t ra ck music pluseffects to the monitors . The only addition is the master 2-t ra ck mach i n e . The point, after all, is to create a DAT,c a s s e t t e , or CD master of the mix.

4. Live to 2For many gigs we bypass the mu l t i t ra ck entire ly,

re c o rding a live performance of any number of mu s i c i a n ss t raight to the 2-tra ck master machine or sending it liveto a stereo broadcast or the house monitors .

A Live to 2 session is the rather intimidating combina-tion of all elements of a Basics and a Mixdown session.Pe r formance anxiety fills the perfo r m e rs , the pro d u c e r,and the engineer.

RECORDING JULY 1999

But for the console itself, the gig isa c t u a l ly quite stra i g h t fo r wa rd: micro-phones in, s t e reo mix out. Of cours ewe want to patch in any number ofsignal pro c e s s o rs . Then the re s u l t i n gs t e reo feed goes to the studio moni-t o rs , the house monitors , the head-p h o n e s , the 2-tra ck master re c o rd e r,and/or the tra n s m i t t e r.

Board of confusionThese four types of sessions define

the full range of signal flow re q u i re-ments of the most capable mixer. Ye tdespite having distilled the possibili-ties into these key categories, t h econsole demands to be appro a ch e dwith some org a n i z a t i o n . B ro a d ly, wecan expect to be frustrated by twoi n h e rent features of the device: com-

p l exity of flow (where is the signalsupposed to be going?) and quantityof controls (look at all these pots!).

C o m p l exity is built into the consolebecause it can provide the signalf l ow structure for any kind of re c o rd-ing session one might encounter. Th epush of any button on the consolemight ra d i c a l ly change the signalf l ow configuration of the dev i c e .

In this studio full of equipment,that little button changes what’sh o o ked up to what. A fader that usedto control the snare micro p h o n egoing to tra ck 16 might instantly besw i t ched into controlling the bari-tone sax level in the mix. It getsmessy fa s t .

The sheer quantity of controls onthe wo rk surface of the mixer is an

i n ev i t able headache because the con-sole is capable of routing so manyd i f f e rent kinds of outputs to so manyd i f f e rent kinds of inputs. 24 tra cks isthe norm for mu l t i t ra ck pro j e c t s .Most of us exceed this. Number ofm i c rophones and signal pro c e s s o rs ?We l l , l e t ’s just say that more is better.

The result is consoles that fill theroom—or a pair of 17" computerm o n i t o rs—with knobs, fa d e rs , a n dsw i t ch e s . The control room starts tolook like the cockpit of the spaces h u t t l e , with a mind-numbing collec-tion of contro l s , l i g h t s , and meters .

These two fa c t o rs , c o m p l exity andq u a n t i t y, c o n s p i re to make the con-sole a confusing and intimidatingd evice to use. It needn’t be.

Flexibility: friend or foe?In the end, a mixer is not doing

a nything especially tricky. The mixerjust creates the signal flow necessaryto get the outputs associated witht o d ay ’s session to the appro p r i a t ei n p u t s .

The console becomes confusingand intimidating when the signalrouting flexibility of the consolet a kes over and the engineer losesc o n t rol over what the console isd o i n g . I t ’s frustrating to do an ove r-dub when the console is in a Live to2 configura t i o n . The thing wo n ’t per-mit you to monitor what’s on themu l t i t ra ck tape.

Or if the console is expecting am i x d ow n , but the session wants tore c o rd basic tra ck s , you ex p e r i e n c ethat helpless feeling of not beingable to hear a single micro p h o n et h a t ’s been set up. The band ke e p sp l ay i n g , but the control ro o mremains silent.

It doesn’t take too many of theseexperiences befo re console-phobiasets in. A loss of confidence matur-ing into an outright fear of using cer-tain consoles is a natural re a c t i o n .Th rough total knowledge of signalf l ow, this can be ove rc o m e .

The key to understanding the sig-nal flow of all consoles is to bre a kthe mu l t i t ra ck re c o rding pro c e s s —whether mixing, ove rd u bb i n g , or any-thing else—into two distinct signalf l ow stages.

Fi rst is the Channel path. A l s ocalled the Record path, it is thepart of the console used to get am i c rophone signal (or synth output)

to the mu l t i t ra ck tape machine and,you know, record it.It usually has am i c rophone preamp at its input,and some nu m b e red tape busses atits output. In between you find afader and maybe some equalization,c o m p re s s i o n , effects sends, c u es e n d s , and other handy feature s

Along the way, the Monitor Path hasa fader and possibly another collec-tion of signal processing circ u i t ry likee q u a l i z a t i o n , c o m p re s s i o n , and more .

Keeping these two signal flowpaths separate in your mind wille n able you to make sense of thep l e t h o ra of controls sitting in fro n tof you on the console. Try to hang onto these two distinct signal pathsc o n c e p t u a l ly, as this will help yo uu n d e rstand how the signal flows t r u c t u re changes when going fro mbasics to mixdow n . Try to break upthe console real estate into ch a n n e lsections and monitor sections so thatyou know which fader is a ch a n n e lfader and which is a monitor fa d e r.

Split consolesConsole manu fa c t u re rs offer us

t wo channel/monitor layo u t s . O n eway to arrange the Channel pathsand Monitor paths is to separa t ethem phy s i c a l ly from each other. P u tall the Channel paths on, s ay, the leftside of the mixer and the Monitor

associated with getting a gre a tsound to tape.

The second distinct audio path isthe Monitor path. It is the part of theconsole you use to actually hear t h esounds you are re c o rd i n g . It typicallybegins with the mu l t i t ra ck tapereturns and ends at the mix bus.

RECORDING JULY 1999

paths on the right as in Fi g u re 8A.This is a split configura t i o n .

Wo rking on this type of console isfa i rly stra i g h t fo r wa rd . See the snareove rload on the mu l t i t ra ck? This is are c o rding pro b l e m . Head to the leftside of the board and grab theChannel fader governing the snarem i c . L evels to tape look good, b u tthe guitar is drowning out the vo c a l ?This is a monitoring pro b l e m . R e a chover to the right side of the consoleand fix it with the Monitor fa d e rs .

Sitting in front of 48 fa d e rs is lessconfusing if you know the 24 on theleft are controlling microphone lev-els to tape (channel fa d e rs) and the24 on the right are controlling mixl evels to the loudspeake rs (monitorfa d e rs ) . So it’s not too confusing thatt h e re are two fa d e rs lab e l e d ,“ L e a dvo c a l .” The one on the left is the micyo u ’ re re c o rding; the one on theright is the tra ck yo u ’ re listening to.

In-line consolesA clever but often confusing

enhancement to the console is thein-line configura t i o n . H e re the ch a n-nel and monitor paths are no longers e p a rated into separate modules ons e p a rate sides of the mixer. In fa c t ,t h ey are combined into a single mod-ule set; see Fi g u re 8B.

Experience tells us that our fo c u s ,and there fo re our signal pro c e s s i n g ,tends to be oriented towa rd eitherthe channel path or the monitorp a t h , but not both. During tra ck i n gthe engineer is dedicating ears ,b ra i n s , h e a r t , and equipment to there c o rd path, t rying to get the bestsounds on tape as possible.

S u re the monitoring part of theconsole is being used. The mu s i cbeing re c o rded couldn’t be heardo t h e r w i s e . But the monitor section isjust creating a ‘ rough mix,’ giving thee n g i n e e r, p roducer and musicians anhonest aural image of what is beingre c o rd e d .

The real wo rk is happening on thechannel side of things, and the moni-tor path should only report theresults of that wo rk accura t e ly.A dding elab o rate signal pro c e s s i n gon the monitor path only adds confu-sion at best, and misleading lies atwo rs t . For example adding a “ s m i l eyfa c e ” equalization curve — b o o s t i n gthe lows and the highs so that agraphic eq would seem to smile—onthe monitor path of the vocal could

hide the fact that a box y, t h i n ,a n dmuffled signal is what’s actuallybeing re c o rded onto the mu l t i t ra ck .

It turns out that for tra ck i n g , ove r-d u bb i n g , m i x i n g , and live to 2 ses-s i o n s , we only re a l ly need signal pro-cessing once, in the channel or themonitor path. We ’ ve just seen thechannel path focus of tra ck i n g .Mixing and Live to 2 sessions arealmost entire ly focused on the finals t e reo mix that we hear, so the engi-

neer and the equipment becomem o re monitor path oriented.

H e rein lies an opportunity toi m p rove the console. If the normalc o u rse of a session ra re ly re q u i re ssignal processing on both the moni-tor path and the channel path, t h e nw hy not cut out half the signalp ro c e s s o rs? If half the equalize rs ,f i l t e rs , c o m p re s s o rs , aux sends, e t c .a re re m ove d , the manu fa c t u rer canoffer the console at a lower price,

RECORDING JULY 1999

or spend the freed re s o u rces on ahigher quality ve rsion of the signalp ro c e s s o rs that re m a i n , or little bitof both.

And as an added bonus the consolegets a little smaller and a lot ofthose knobs and sw i t ches disappear,reducing costs and confusion furthers t i l l . This motivates the creation ofthe in-line console.

On an in-line console, the ch a n n e lpath and the monitor path are com-bined into a single module so theycan share some equipment. S w i t ch e slie next to most pieces of the con-s o l e , letting the engineer decide,piece by piece, whether ag iven feature is needed in thechannel path or the monitorp a t h . The equalize r, for ex a m-p l e , can be sw i t ched into there c o rd path during an ove rd u band then into the monitorpath during mixdow n . D i t t ofor any other signal pro c e s s-i n g .

Of cours e , some equipmentis re q u i red for both the ch a n-nel path and the monitorp a t h — l i ke fa d e rs . So there isa lways a channel fader and as e p a rate monitor fader (lessex p e n s ive mixers often usemonitor pots). The in-lineconsole is a clever collectionof only the equipment need-e d , when it’s needed, w h e rei t ’s needed.

Channel surfingAn unavo i d able result of stre a m-

lining the console into an in-line con-f i g u ration is the fo l l owing kind ofc o n f u s i o n . A single module, w h i chn ow consists of two distinct signalp a t h s , might have two ve ry differe n taudio sounds within it.

Consider a simple vocal ove rd u b . Ag iven module might easily have avocal microphone on its ch a n n e lfader but some other signal, l i ke ap rev i o u s ly re c o rded guitar tra ck , o nits monitor fa d e r. The live vocal tra ckis actually being monitored on someother module and there is no ch a n-

nel for the guitar, as it wa sove rd u bbed ye s t e rd ay.

L evels to tape look good,but the guitar is drowning outthe vocal? This is a monitoringp ro b l e m . The solution is toturn down the monitor fa d e rfor the guitar. But where is it?

U n l i ke the split design, a nin-line console presents uswith the ability to both re c o rdand monitor signals on eve rymodule across the entire con-s o l e . E a ch module has a moni-tor path. Th e re fo re each mod-ule might have a prev i o u s lyre c o rded tra ck under the con-t rol of one of its fa d e rs . E a chmodule also has a ch a n n e lp a t h . Th e re fo re , e a ch modulemight have a live micro p h o n esignal running through it.

To use an in-line console,you must be able to answe r

the fo l l owing question in a split sec-ond: “ Wh i ch of the perhaps 100fa d e rs in front of me controls theguitar tra ck ? ” K n ow where the gui-t a r ’s monitor path is at all times, a n dd o n ’t be bothered if the ch a n n e lfader sharing that module has noth-ing to do with the guitar tra ck . Th emonitor strip may say, “ G u i t a r.”

But you know that the ch a n n e lcontains the vocal being re c o rd e d . I tis essential to know how to turnd own the guitar’s monitor fader with-out fear of accidentally pulling dow nthe level of the vocal going to themu l t i t ra ck tape.

RECORDING JULY 1999

One must maintain tra ck sheets,set-up sheets, and other session doc-u m e n t a t i o n . These pieces of papercan be as important as the tape/harddisk that stores the mu s i c . H oweve r,rather than just re lying on thesen o t e s , it helps to maintain a mentali nve n t o ry of where eve ry micro-p h o n e , t ra ck , and effects unit isp a t ched into the mixer.

M u ch to the frustration of theassistant engineer who needs towa t ch and document what’s goingon and the producer who would liketo figure out what’s going on, m a nye n g i n e e rs don’t even bother lab e l-ing the strip or any equipment fo ran ove rdub session or even a mixs e s s i o n . The entire session set-upand tra ck sheet is in theirh e a d s .

If you have enough men-tal RAM for this, t ry to doi t . It helps you get into thep ro j e c t . You are fo rced tobe as focused on the songas the musicians are .Th ey ’ ve got lines andchanges and solos andlyrics to keep tra ck of.

The engineer can beexpected to keep up withthe microphones andreverbs and tra cks on tape.This comes with pra c t i c e .And when you know thel ayout of the console thisi n t i m a t e ly, the ove rl a p p i n gof microphones and tra ck sthat appears on an in-lineconsole is not so confusing.

S u re the split console offers someg e ographic separation of mic sig-nals from tape signals, w h i ch make sit a little easier to remember what’sw h e re . But through practice you aregoing to keep up with all thedetails in a session any way. The in-line console becomes a perfectlyc o m fo r t able place to wo rk .

Getting your ducks in a rowIf yo u ’ ve dialed in the perfect

equalization and compression for thes n a re drum during a basics session,but fail to notice that you are pro-cessing its monitor path instead ofits channel path, you are in for a sur-p r i s e . When you play back the tra ckn ext week for ove rd u b s , you’ll find

that that powerful snarewas a monitoring cre a t i o no n ly and didn’t go to tape.It eva p o rated on the lastp l ay b a ck last we e k .

H o p e f u l ly you re m e m b e rand/or document the set-tings of all signal pro c e s s-ing equipment any way, b u tm o re helpful would be toh ave had the signal pro-cessing chain in front ofthe mu l t i t ra ck tapem a ch i n e , not after. No wo r-r i e s .

Th rough ex p e r i e n c e ,you’ll learn the best placefor signal processing fo ra ny given session.E q u a l i z a t i o n , c o m p re s s i o n ,reve r b , the headphones—

e a ch has a logical choice for itss o u rce: the channel path or monitorp a t h . And it varies by type of ses-s i o n . Once yo u ’ ve lived through avariety of sessions it becomesi n s t i n c t ive .

Your mission is to know how topiece together channel paths, m o n i-tor paths, and any desired signal pro-cessing for any type of session. Th e nthe signal flow flexibility of anym i x e r, split or in-line, is no longeri n t i m i d a t i n g .

By staying oriented to the ch a n n e lportion of the signal and the monitorportion of the signal, you can useeither console to accomplish thewo rk of any session. You can fo c u sinstead on music making.

RECORDING JULY 1999

What’s that switch do?I will admit that there is such a

thing as too mu ch . You may be anexcellent engineer capable of re c o rd-ing sweet tra ck s , but when Pe t e rG abriel invites you to his studio andyou sit in front of his 72 channel G-Series SSL, you will have tro u b l edoing what you know how to do( re c o rding the sweet tra cks) whiledealing with what you don’t knowh ow to do (use this enormous mixerw i t h , g u l p, m o re than 8,000 knobsand sw i t ch e s ) .

Good news: that vast control sur-face is primarily just one smallerc o n t rol group (the module) re p e a t e dove r, and ove r, and over again. K n owh ow to use a single module and yo uk n ow how to use the whole collectionof 72 modules.

I m p ress your clients. I m p ress yo u rf r i e n d s . H e ck , i m p ress yo u rs e l f.Master the many subtle aspects ofjuggling monitor and channel pathst h rough different types of sessions,and learn to sit calmly in front ofconsoles that have grown we l lb eyond 100 modules, and you’ll haved eveloped 90% of the ability to usea ny console any w h e re .

Alex Case always has a conv i n c i n g lyinnocent look on his face when he seesa traffic cop or a console. You can writeto Alex with questions or suggestionson wh at you’d like to see in ‘Nuts &Bolts’ at case@re c o r d i n g m a g . c o m .

basics (or tracking): The earlystages of a re c o rding pro j e c t —re c o rding the individual tra cks onthe mu l t i t ra ck re c o rd e r. This is doneb e fo re adding ove rd u b s , mixing tos t e re o, or mastering for final duplica-t i o n / d i s t r i b u t i o n .

bus (sometimes spelled buss): Asignal path that can accept and mixsignals from various sourc e s .

channel, channel path (or inputpath or record path): The signalcoming from your source (mic,instrument, or returning from analready-recorded track on your mul-titracker) into one of the mixer’schannels, passing through that chan-nel’s electronics, then usually get-ting split to go to several destina-tions (monitor section for listening,multitracker to be recorded, effectsends for delay/reverb etc, mastersection for stereo mix).

compression: Dynamic tre a t m e n tof a signal so that the differe n c e

b e t ween the loudest and softestmoments is re d u c e d .

D e l ay: E l e c t ro n i c a l ly created re p e-titions of a sound (ech o e s ) . S h o r t e rd e l ays are perc e ived as fl a n g i n g ,ch o-rusing or doubling.We’ll study theseeffects another time.

DI: D i rect Inject or Direct Input—bypassing an instrument amp by tak-ing the signal (usually from guitarsand bass guitars) straight to a ch a n-nel input of the board . U s u a l ly this isdone via a small device called ad i rect box , w h i ch matches levels sothe instrument’s weak signal ism a t ched to the board ’s input.

equalization or eq: Tonal tre a t-ment of a signal by attenu a t i n g( reducing) or boosting selectedranges of the total spectrum (bassand treble controls are the simplestex a m p l e s ) . Th e re are many types ofe q , w h i ch we’ll learn about later.

filter: An electronic device thatreduces certain ranges of the totals p e c t r u m . For ex a m p l e , a l o w - p a s sfilter a t t e nuates (reduces) high fre-q u e n c i e s , passes (leaves alone) lowf re q u e n c i e s . Equalization is genera l-ly done with arrays of filters .

l ive to 2: Bypassing a mu l t i t ra ckre c o rd e r, mixing any number ofinput sources all at once into stere o.

microphone preamp: An electro n i cd evice that increases the typicallyve ry weak signals produced bym i c rophones so that these signalscan join others at “ l i n e ” l evel in am i x e r.

mix bus: See bu s .m i x d own: U s u a l ly stage three in a

re c o rding project after basics andove rd u b s , this is when all prev i o u s lyre c o rded tra cks on the mu l t i t ra cke ra re routed through (returned to) theb o a rd , their levels and panning andeffects adjusted, resulting in a finals t e reo mix.

m i xer (or console, b o a r d , d e s k ) :An apparatus with many electro n i c

GlossaryWe’ll include a list of terms introduced

in each installment of the column, and

collect them on our Web site as an

ongoing reference. Here’s a starting

list of terms mentioned in this article.

RECORDING JULY 1999

c i rc u i t s , all designed to acceptaudio signals, split (duplicate)t h e m , re - route them, combine them,adjust their leve l s , tonal ch a ra c t e r-i s t i c s , and placement in the finals t e reo mix.

module: A group of electronic cir-cuits that combine to ach i eve a spe-cific task, as in a mixer’s ch a n n e lm o d u l e

monitor path: A mixer signal paththat accepts and mixes signals to bem o n i t o red (listened to).

outboard signal processing:Treatment (reve r b , d e l ay s , o t h e rs) ofsignals outside of the board (re a ch e dvia effects or auxiliary send bussesand send outputs, returned to theb o a rd via return inputs and re t u r nb u s s e s )

overdub: A dding one tra ck or sev-e ral tra cks to prev i o u s ly re c o rd e dt ra cks (e.g. a singer adds vocals afterthe instrumental tra cks have beenre c o rd e d ) .

p a t ch cable: A cord connecting twopoints to carry a signal from A to B.

pot: Short for p o t e n t i o m e n t e r, ad evice that increases or decre a s e sthe signal strength (a kind of vo l u m ec o n t rol) or tweaks eq settings, e t c .B a s i c a l ly a techie term for a knob.

return (tape or aux or effects): Atype of input into the board bringingb a ck signals other than the originals o u rces (mics or instruments), e i t h e rp rev i o u s ly re c o rded mu l t i t ra ck s , o rsignals returning from outboardp ro c e s s o rs . See s e n d .

r everb: An electro n i c a l ly cre a t e dillusion of room acoustics.

send (aux or cue): C i rcuits (busses)that lead to an output connectorf rom where signals can be sent too u t b o a rd pro c e s s o rs or to monitoring(listening) setups. See re t u r n .

stereo bus: The final two circ u i t sin a board that accept and mix sig-nals to become the Left and Rightchannels of a stereo mix.

t ape bus: A circuit that acceptsand mixes signals to or from tapere c o rd e rs .

t wo mix: See s t e reo bu s .

Excerpted from the July edition of RECORDING magazine. ©1999 Music Maker Publications, Inc. Reprinted with permission.

5412 Idylwild Trail, Suite 100, Boulder, CO 80301 Tel: (303) 516-9118 Fax: (303) 516-9119

For Subsciption Information, call: 1-800-582-8326

RECORDING AUGUST 1999

The more entertaining part of ourjob is signal pro c e s s i n g . After micro-phone selection and placementt we a k s , we genera l ly turn to signalp rocessing for fun/help, using somecombination of filters , e q u a l i ze rs ,c o m p re s s o rs , g a t e s , d e l ay s , reve r b s ,and multieffects pro c e s s o rs .

The important question is, H ow dowe hook up all this outboard gear toan alre a dy convoluted signal patht h rough the console?

a convincing emotional presence for avoice fighting its way out of a pair ofl o u d s p e a ke rs .

The distinguishing ch a ra c t e r i s t i cof this type of p a rallel signal pro c e s s-ing is that it is added to the signal—itd o e s n ’t replace the signal. The struc-t u re is illustrated in Fi g u re 1A. Th ed ry (i.e. without reverb; more gener-a l ly, without any kind of effect) sig-nal continues on its merry wayt h rough the console as if the reve r bwe re never add e d . The reverb itselfis a parallel signal path, b e g i n n i n gwith some amount of the dry vo c a l ,going through the reve r b , and re t u r n-ing elsew h e re on the console to becombined with the vocal and the re s tof the mix. (Note that in these ex a m-p l e s , signals are being routed to aL / R bus for monitoring on the speak-e rs show n , as discussed last issue.)

C o nve rs e ly, consider equalization.A typical application of equalizationis to make a mediocre sounding tra ckb e a u t i f u l . A dull acoustic guitar ismade to shimmer and spark l e ,c o u r-tesy of some boost around 10 or 12k H z . A shrill vocal gets a care f u l lyplaced dip somew h e re between 3 and7 kHz to become more listenab l e .

The idea is that this ‘fixes’ thesound; you don’t want to hear theu n p rocessed ve rsion any m o re , j u s tthe good one. A dding shimmer to aguitar isn’t so useful if the mu rkyguitar sound is still in the mix too.And the point of eq-ing the vo c a lt ra ck was to make the painful edgi-ness of the sound go away.

H e re the signal processing is placedin series with the signal flow, as show nin Fi g u re 1B. E q u a l i z i n g , c o m p re s s i o n(and other dynamics pro c e s s i n g ) ,d e -e s s i n g , wa h - wa h ,d i s t o r t i o n , and sucha re all typically done serially so thatyou only hear the effected signal andnone of the unaffected signal.

We’ll explain what all thosep rocesses are , of cours e — s t i ck withthis series.

The effects sendNot surp r i s i n g ly, p a rallel and seri-

al flow structures re q u i re differe n ta p p ro a ches on the console. For para l-lel pro c e s s i n g , some amount of ag iven tra ck is sent off to an effectsunit for pro c e s s i n g .

Th a t ’s what the effects send is fo r.Also known as an echo send or auxsend—short for auxiliary—it is a sim-ple way to tap into a signal withinthe console and send some amount of

Parallel and serial processingP h i l o s o p h i c a l ly, t h e re are two

a p p ro a ches to adding effects to am i x . Consider first the use of reve r bon a vocal tra ck . The right reve r bsupports a vocal tra ck that was pro b-ab ly re c o rded in a dead room with amic positioned close up.

I t ’s not mere ly a matter of support,h oweve r. A touch of just the rightkind of reverb can enable the vocal tosoar into pop music heave n ,c re a t i n g

Part 2 in our beginner’s series

Outboard Signal Process i n gOutboard Signal Process i n g

If you read last month’s article on the funda m e n ta l s

of signal flow through a console, you should hav e

(or be developing) a comfortable understanding of

channel paths and monitor paths. But the fun pa rt of

our job isn’t just getting a microphone signal to

tape—it’s being able to capture accurat e ly, enhance

s u b t ly, or ra d i ca l ly reshape the sounds being

recorded and mixed.


that signal to some destination. I fthat destination is, s ay, a reverb ord e l ay, the “ e f f e c t e d ” signals comeb a ck to the console’s effect re t u r n so raux re t u r n s—another set of your con-s o l e ’s inputs that usually feeds t raight into the master L/R b u s .

P ro b ab ly ava i l able on eve ry ch a n-nel module of the console, the effectssend is re a l ly just another fader ork n o b . Not a channel fader or monitorfa d e r, the effects send fader or knobdetermines the level of the signalbeing sent to the signal pro c e s s o r.R eve r b , d e l ay, and such are typicallydone as parallel effects and there fo rere ly on effects sends. C h e ck outFi g u re 2A to see how it wo rk s .

Th e re ’s more to the effects s e n dthan meets the eye , h oweve r. I t ’s notjust an ‘effects fa d e r.’ An importantbenefit of having an effects sendl evel knob on eve ry channel on theconsole is that a single effectsp rocessor can be shared by all thosech a n n e l s . Unless you have lots ofve ry high quality (i.e. ve ry ex p e n-s ive) reve r b s , for instance, it isn’tp ractical to use one on just thes n a re , or just the piano, or just thevo c a l .

Turn up the effects send level onthe piano tra ck a little to add asmall amount of reverb to the piano.Turn up the effects send level onthe vocal a lot to add a genero u samount of reverb to the vo c a l . I nfa c t , the effects send levels acro s sthe entire console can be used toc reate a separate mix of all themusic being sent to an outboardd ev i c e . I t ’s a mix the engineer does-n ’t usually listen to; it’s the mix thereverb ‘listens’ to when genera t i n gits sound.

Fading fastSo in case you thought there wa s n ’t

enough for the engineer to do duringa session, l e t ’s rev i ew the fa d e rs thata re in action: the channel fa d e rs arec o n t rolling the levels of the signalsgoing to the mu l t i t ra ck , the monitorfa d e rs are controlling the levels of allthe different tra cks being listened toin the control ro o m , and the effectsends are controlling the levels of allthe different components of mu s i cgoing to the reve r b . Th ree differe n tsets of fa d e rs have to be care f u l lyadjusted to make musical sense fo rtheir own specific purp o s e s .

Th e re are two more subtleties tobe ex p l o re d . Fi rs t , as we are ra re lysatisfied with just one kind of effectin a mu l t i t ra ck pro j e c t , we wo u l dp ro b ab ly like to employ a number ofd i f f e rent signal pro c e s s o rs all atonce on a single pro j e c t .

Each one of them might expect touse its own effects send. That is, wemight have one box with a sweetand long reverb dialed in, anotheradding a rich, thick chorus, and per-haps a third box generating aneighth note delay with two or threefading repetitions.

The lead vocal might be sent inva rying amounts to the reve r b , ch o-r u s , and delay; the piano gets just at o u ch of reverb; and the back gro u n dvocals get a heavy dose of ch o r u s ,e ch o, and a hint of reve r b . We needm o re than an effects send to dot h i s — we need three effects s e n d s .

The solution, f u n c t i o n a l ly, is thatsimple: more effects s e n d s . I t ’s an

important feature to look for on con-s o l e s , as the number of sends deter-mines the number of different para l-lel effects devices you can use atonce during a typical session.

Beyond this ability to build up sev-eral different effects sub-mixes,effects sends offer us a second, veryimportant advancement in our ses-sion work: cue mixes. (On some con-soles, there are two sets of sends—one set labelled ‘effects’ and onelabelled ‘aux’. In that case, it’s usual-ly the aux sends that fulfill the func-tions we’re about to discuss.)

Generally sent to headphones inthe studio or, in the case of livesound, fold-back monitors on the

Excerpted from the August edition of RECORDING magazine. ©1999 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119

For Subscription Information, call: 1-800-582-8326


stage, an aux send is used to createthe cue mix, a mix the musicians useto hear themselves and each other. Asthe parameters are the same as aneffects send (the ability to create amix from all the channels and moni-tors on the console), the cue mix canrely on the same feature. With oneexception: the cue mix, unlike thereturns from the effects devices, isnot returned to the console, but sentto monitors or headphones.

N ow let’s do a fader ch e ck: ch a n n e lfa d e rs control any tra cks being re c o rd-e d , monitor fa d e rs build up the control room mix, aux oreffects send number one might control the mix feeding theh e a d p h o n e s , aux or effects send number two might contro lthe levels going to a long hall reverb progra m , aux oreffects send number three might be the signals going to at h i ck chorus patch , and aux or effects send number fo u rfeeds a delay unit—six different mixes care f u l ly cre a t e dand maintained throughout the ove rd u b .

Th a t ’s a lot to do at once. O h , and by the way it’s notenough for the right signals to get to the right places;t h ey also have to make musical sense. The levels to tapeneed to be just right for the medium on which you arere c o rd i n g . The monitor mix needs to sound thrilling. Th eheadphone mix needs to sound inspiring. And the effectsneed to be appro p r i a t e ly balanced—too mu ch or notenough of any signal going to any effects unit and themix loses impact.

This is some high-resolution mu l t i t a s k i n g . And it ismu ch more manageable if the console is a comfo r t ab l eplace to wo rk . Experience through session wo rk in com-bination with studying magazines like this one make thisnot just doab l e , but fun.

Pre or postWith all these fa d e rs performing different functions on

the console, it is important to revisit the regular monitorfader to see how it fits into the signal flow.

C o m p a re the monitor mix in the control room to thecue mix in the headphones. The singer might want avo c a l - h e avy mix (also known as “ m o re of me”) to sing to,with ex t ra vocal reverb for inspiration and no distra c t i n gguitar fills.

No pro b l e m . Use the send dedicated to the headphonesto create the mix she wa n t s .

But you have different priorities—you want to hear thevocal in an appropriate musical context with the othert ra ck s . M o re ove r, ex t ra reverb on the vocal would make itdifficult to evaluate the vocal performance going to tapeas it would perhaps mask any pitch ,t i m i n g , or dictionp ro b l e m s .

So clearly the cue mix and the control room mix needto be two independent mixes. Th ey ’ re created using auxsends and monitor fa d e rs .

But other things go on in the control room during asimple vocal take. For example, you mightwant to turn up the piano and pull downthe guitar to experiment with some alter-native approaches to the arrangement. Orperhaps the vocal pitch sounds iffy. Theproblem may be the 12-string guitar, notthe singer. So the 12-string is temporarilyattenuated (its level is lowered) in the con-trol room so you can evaluate the singer’spitch relative to the piano, which is intune.

All these fader moves in the control ro o mneed to happen in a way that doesn’t affectthe mix in the headphones—an obvious dis-t raction for the perfo r m e r. Th a t ’s what thep re/post sw i t ch shown in Fi g u re 2A is fo r.

A useful feature of many aux or effectssends is that they can grab the signal befo re( i . e .p re) or after (i.e. post) the channel or mon-itor fa d e r. C l e a rly, i t ’s desirable for the head-phone mix to be sourced pre - fader so that itwill play along independently, u n changed bya ny of these control room activ i t i e s .



What is the usefulness of a post-fader send, you mightthen ask? The answer lives in the aux send’s other prima-ry function: effects sends.

L e t ’s observe a ve ry simple two - t ra ck folk music mix-d own: fader one is the vocal tra ck and fader two is the gui-tar tra ck (re q u i red by the folk standards bureau to be anacoustic guitar). The we l l - re c o rded tra cks are made to

sound even better by the oh-so-careful addition of someroom ambience to support and enhance the vocal while at o u ch of plate reverb adds fullness and width to the guitar.

After a few hours , e r, I mean five minutes of twe a k i n gthe mix, the re c o rd label re p re s e n t a t ives arrive andremind you that “ I t ’s the vocal stupid.” O o p s , the engi-neer is so in love with the rich and spark ly acoustic gui-tar sound that the vocal tra ck was a little neglected.

It must be fixed. Not too tricky—just turn up the vo c a l .H e re ’s the rub. While pushing up the vocal fader will

change the re l a t ive loudness of the vocal over the guitarand there fo re make it easier to fo l l ow the ly r i c , it alsochanges the re l a t ive balance of the vocal ve rsus its ow nreve r b . If the vo c a l ’s reverb send is pre - fa d e r, turning up thevocal leaves its reverb behind; the vocal becomes too dry,the singer is left too ex p o s e d , and the larger than life magiccombination of dry vocal plus super- sweet reverb is lost.

The solution is the post-fader effects send. If thes o u rce of the signal going to the reverb is after the fa d e r,then fader rides will also change send levels to thereve r b . The all-important re l a t ive balance between dryand processed sound will be maintained.

Effects are genera l ly sent post-fader for this re a s o n .

The engineer is re a l ly making two different decisions,determining the amount of reverb desired for this vo c a l ,and the level of the vocal appropriate for this mix.F l exibility in solving these two separate issues is main-tained through the use of the post-fader send.

The insertAs seen in Fi g u re 1B, serial processing is mu ch simpler

than the buss concept of sends and returns we ’ ve justt a l ked about for parallel pro c e s s i n g . Rather than sharingone effect over many channels and mixing it with a drys i g n a l , yo u ’ re routing an entire signal through an effectand not sharing that effect with any other signals.

You could just run your signals through your effectsb e fo re sending them to the mixer. But there ’s anotherway to do serial processing that offers some adva n t a g e sin flexibility—the channel insert, s h own in Fi g u re 2B.

An insert point has a send that goes from a mixerchannel to an effect, and a return that comes back fro mthe effect to the mixer ch a n n e l . Using this pair of con-nections “ i n s e r t s ” your outboard effect into the signalf l ow.

Why use an insert? Remember that eve ry mixer ch a n-nel can do a number of things to a signal—input gainc o n t ro l , e q , l eve l , aux sends, buss assignment and pan-n i n g . Sometimes certain signals may sound better if yo ucan wo rk on them a bit with the mixer b e f o re t h ey go tothe outboard effect. For instance, you might want to setthe input gain or eq on a sound befo re compressing it.

For those situations, a channel insert lets you take asignal from a mixer ch a n n e l , send it through an effect,and bring it back to the point from which it left. S o m em i x e rs allow you to select where the insert attaches tothe signal flow; the most common inserts come after theeq and befo re the level fader and bus/pan contro l s .

On most mixers , an insert is a single t i p - r i n g - s l e e v e(TRS) jack , w h i ch accepts a TRS plug like the ones yo usee on headphones. A TRS jack can carry two signals,

one on the tip and one on the ring, with the sleeve actingas ground re f e rence for both. For headphones, one signalis the left channel and the other is the right, but for ani n s e r t , one signal is the send and one is the re t u r n .Signals travel to and from the mixer on one cab l e , w h i choften splits into two separate cables to plug into theeffect in/out (a “ Y ” c ab l e ) . U s u a l ly the tip is the sendand the ring is the re t u r n , but ch e ck your mixer’s manu a lto be sure .

Wrench turningsAn org a n i zed appro a ch to the console and also the out-

b o a rd processing equipment will help make it easy andfun to wo rk in a room full of gear. An intuitive under-standing of when to use an aux or effects send and whento use an insert will free your mind to be cre a t ive withthe effects.

And knowing that cue mixes genera l ly use pre - fa d e rsends while most parallel effects need post-fader sendswill keep you out of tro u b l e .

Alex Case welcomes suggestions for ‘Nuts & Bolts.’ W r i t eto him at case@re c o r d i n g m a g . c o m .


A touch of reverb can enable the voca lto soar into pop music heaven,

c r e ating a convincing emotional presence for a voice fighting its way out of a pair of loudspeakers.

RECORDING SEPTEMBER 1999

H ave you ever been picked on by some bully in sch o o l ?I have . After the event I re p l ayed it over and over in mymind until I came up with the perfect comeb a ck — t h eone I wished I’d delive red instead of giving him my Spamand potato chip sandw i ch .

The desire to improve history by rewriting it is pre t t yi n s t i n c t ive . And in music-making it isn’t just a wish, i t ’s amodus opera n d i .

The tool that lets us re l ive a situation as long as itt a kes to perfect it is the mu l t i t ra ck re c o rd e r. Rather thanl iving with the “ l ive - t o - 2 ” re c o rd i n g , the mu l t i t ra ck give sus that mu ch - a p p reciated second—or third , or fifteenth—chance to get a better take .

The ins and outsH e re ’s the idea: as music makes its way from the va r i-

ous microphones to the final 2-tra ck master, we store itt e m p o ra r i ly on a mu l t i t ra ck . And that’s how it’s hooke dup: microphone out, t h rough the console re c o rd path( channel path), to the mu l t i t ra ck re c o rd e r. On play b a ck ,we send signals from the mu l t i t ra ck re c o rder out thro u g hthe console’s monitor path to the mix bus. ( Fi g u re 1s ke t ches it out in a general way: ch e ck out our prev i o u st wo installments to re f resh your understanding of thec o n s o l e ’s busses and signal paths.)

Falling in between the channel path and the monitorp a t h , the mu l t i t ra ck re c o rding device—whether a com-puter hard disk re c o rd e r, digital tape deck , or analog

tape re c o rd e r — re c e ives whatever tra cks you are curre n t-ly re c o rding at its input; it plays back whatever tra ck sa re alre a dy re c o rd e d . The mu l t i t ra ck is nothing morethan an audio storage dev i c e . It stores the drums whileyou add bass. It stores the rhythm section while you addvocals and solos.

What are good devices for audio storage? Th e re are justa couple of valid answe rs (so far): tape and disk. For analogs t o ra g e , tape is the only practical mu l t i t ra ck medium. Fo rd i g i t a l ,t h e re ’s tape media like A DAT and TA S CA M ,b u tdisks in all their formats are also possible: magneto-opticald i s k s , internal or external hard disks, re m ovable disks…

N a t u ra l ly the re c o rding device must possess highsound quality, re l i ab i l i t y, and affo rd ab i l i t y. Th ree otherf e a t u res are perhaps less obv i o u s .

Fi rs t , it must be able to be erased and then re - re c o rd e dove r, on the off chance someone makes a mistake .S e c o n d , the re c o rding must be ava i l able for immediatep l ay b a ck right after being re c o rd e d . So while 35mm filmmight be a great release format for sound, it is impra c t i-cal in the studio because it re q u i res processing in a filml ab o ra t o ry befo re it can be played back .

A final functional re q u i rement of the mu l t i t ra ckre c o rder is that it must be able to re c o rd and play backs i mu l t a n e o u s ly. Its power as a cre a t ive tool in the re c o rd-ing studio depends on its ability to ov e r d u b: re c o rd a newt ra ck while simu l t a n e o u s ly playing back prev i o u s lyre c o rded tra ck s .


M u lt i t rack Re c o r d e r sM u lt i t rack Re c o r d e r s All about the machine atthe center of every studio


H ow can a re c o rder play back onething while re c o rding another?Simple enough. E a ch tra ck of themu l t i t ra ck re c o rder assumes one oft wo states: play b a ck or re c o rd . Wh e na tra ck is in the play b a ck mode, i t saudio is sent to the mu l t i t ra ckm a ch i n e ’s output. In re c o rd mode,n ew audio for that tra ck isw r i t t e n / s t o red on the tape or disk.

OverdubsThe cool thing about a mu l t i t ra ck is

that it can enter re c o rd mode selec-t ive ly, t ra ck by tra ck , so that it re c o rd so n ly on the tra cks desire d .The othert ra cks are n ’t being re c o rded onto, s ot h ey instead stay in play b a ck mode.This accommodates the ove rd u b .

L e t ’s ch e ck out a power trio sessionconsisting of drums on tra ck one,bass on tra ck two, guitar on tra ckt h re e .Vocals are to be done as anove rdub onto tra ck fo u r. During theove rd u b , it is pretty clear what mu l t i-t ra ck outputs one, t wo, and three are .

What signal appears at outputnumber four? Seems logical that itshould be the vocal being ove r-d u bb e d . But one wo n d e rs how are c o rder can play back the samet ra ck it’s re c o rd i n g . In fa c t , it can’t !So it doesn’t .

On the tra ck actually being re c o rd-e d , the tape machine can’t play backwhat it’s laying onto tape or disk.Th e re is an inev i t able delay betwe e nwhen the signal is re c o rded andwhen it is played back . That delay islong enough to cause the mu s i c a le q u ivalent of a train wre ck .

The solution is that the mach i n ed o e s n ’t even attempt to play backthe tra ck it is re c o rd i n g . Instead theoutput for the tra ck being re c o rd e dis its own input. H o n e s t . No typot h e re . The vocal signal being re c o rd-ed is sent to mu l t i t ra ck input nu m-ber fo u r, and it is split within themu l t i t ra ck machine befo re beingre c o rd e d . The divided vocal signalgoes both to the re c o rder and simu l-t a n e o u s ly to the mu l t i t ra ck output.This is standard operating pro c e-d u re , and is shown in Fi g u re 2.

The mode that routes the input ofthe tra ck active ly being re c o rded toits own output is called input mode. I fa tra ck isn’t in input mode, its outputsignal is the audio alre a dy re c o rd e don that particular tra ck . P l ay b a ckmode (or repro mode, f rom ‘ re p ro-duce’) describes this configura t i o n ,and it is the standard signal flow fo rt ra cks not curre n t ly being re c o rd e d .

So there are two choices for whatsignal appears at each output of amu l t i t ra ck . It can play b a ck what’sa l re a dy on the tape or disk (that’s

re p ro mode), or it can play backwhat is curre n t ly being sent to thetape or disk (that’s input mode).Yo u ’ ve got two options here: re re a dthis para graph half a dozen times orsit in front of a tape machine for acouple minu t e s . I t ’s mu ch more con-fusing to say than it is to do.

The tricks and treatsO k ay, so a mu l t i t ra ck is used to

re c o rd the rather elab o rate audioa r rangement of a pop tune a fewt ra cks at a time—an arra n g e m e n tthat might use more than 24 tra ck sof re c o rded mu s i c .

N a t u ra l ly, we do more than justprint tra cks with the mu l t i t ra ck .

L e t ’s ex p l o re some of the more sub-tle production capabilities offered bythe humble mu l t i t ra ck re c o rd e r.

One handy feature is the ability tore c o rd from one tra ck to another, ap rocess called b o u n c i n g. Th e re are af ew reasons to bounce tra ck s .

The first reason is for conve n i e n c e .As a project progre s s e s , the mu l t i-t ra ck can get a little messy, w i t halternate vocal tra ck s , solo out-take s ,b a ck ground vocal harmony ideas,and that experimental (but ultimate-ly rejected) contrabassoon solo alls p read out to various locationsamong the keeper tra ck s .

I t ’s often helpful to re o rg a n i ze thet ra cks into a more logical order: all the

drums on the first few tra ck s , all thevocals on the last few tra ck s , with ther hythm section laid out in an ord e rthat is logical and comfo r t able for yo u .To move a signal from one tra ck toa n o t h e r, s i m p ly hook up the output ofone tra ck to the input of another, s e tthe source tra ck to play b a ck , the tar-get tra ck to re c o rd mode, and re c o rd .

On analog mach i n e s , this costs yo ua generation of quality, w h i ch ism o re than tolerable on some of thebetter mach i n e s . On digitalm a ch i n e s , this bouncing ability oftenexists digitally within the mach i n e .No patch i n g , and no generation loss.Needless to say, we bounce tra ck sm o re often on digital mu l t i t ra ck s .

Another variation on the bounc-ing theme is submixing. Instead ofdoing a direct transfer from onet ra ck to another, it is often handy toc reate and re c o rd a submix of acomponent of the tune. If drumswe re re c o rded across 12 tra cks ofthe mu l t i t ra ck , it can be a good idea

to mix just the drums down to twon ew tra ck s , l e aving the non-drumt ra cks unmixed for now.

This is helpful in two way s . Fi rs t , i tf rees up tra cks for other purp o s e s .Digital re c o rding softwa re can some-times offer “infinite nu m b e rs of

t ra ck s ,” but that can be limited bythe size of the storage medium andthe power of the machine doing there c o rd i n g . And analog re c o rd e rs , a n ddigital tape and MiniDisk mach i n e s ,h ave fixed nu m b e rs of tra cks ava i l-ab l e . So bouncing downlets you take


a large number of tra cks and pre - m i xthem to make room for more of yo u ro rch e s t ra t i o n .

S e c o n d , if some elements of thet u n e — l i ke the drums in this case—a re alre a dy care f u l ly pre - m i x e d , t h e nc re a t ive energ y, effects dev i c e s ,p a t ch cab l e s , and fader fingers aref ree to focus on the remaining ele-ments of the mix.

Of cours e , t h e re is a downside tos u b m i x i n g . In order to free a tra ckby submixing, you must perform therather scary act of erasing the origi-n a l . Submixing twe l ve drum tra ck sto a stereo pair of tra cks will indeedf ree up ten tra cks for vo c a l s , s o l o s ,and other musical ideas, but only ifyou are willing to erase the originals n a re tra ck , the original kick drumt ra ck , e t c .

And the submixed drums are onlyuseful in the final mix if the submixitself is, as they say, t o t a l ly killer.Without knowing ex a c t ly what thee n t i re mix will sound like , yo u ’ ve gotto create an appro p r i a t e ,c o m p l e m e n-t a ry, compelling submix of the drums.

C l e a rly, submixing some number oft ra cks down to fewer tra cks is itsown skill, requiring not only basicmixing chops but also a little bit ofex t ra - s e n s o ry perception to pre d i c tthe appropriate mix goal of a give nelement of the ove rall mix. Expect tom a ke a few mistake s . Plan to re m i xthese submixes a few times, and tryto have a backup of any tra cks yo ue ra s e . This is easier to do with digitalsystems than with analog re c o rd e rs ,since you don’t lose a generation ofsound quality when you dub off ab a ck u p.

Sometimes submixes are printedto the mu l t i t ra ck not so mu ch to fre eother tra cks but to ‘ s t o re’ a mixm ove . Printed mix mov e sa re a goodway to have manual fader rides andc razy pan pot moves in a mix withouta u t o m a t i o n . Just do the mix movem a nu a l ly, re c o rding the audio re s u l tto spare tra cks of the mu l t i t ra ck .

Yet another variation on thebouncing theme is called c o m p i n g. Acomp is hip-speak for a composite. I tre f e rs to creating a single tra ck thatis in fact a collection of pieces of anynumber of different tra cks—the bestchorus happened on take three whilethe best bridge happened on takes eve n , and the best intro was ye s t e r-d ay ’s scra t ch vo c a l .A a ck !

M i x d own with vocals all over themu l t i t ra ck coming up on fa d e rs allover the console is ve ry distra c t i n g .Th ey are comped by re c o rding allthe appropriate pieces to a separa t et ra ck . The comped tra ck thena p p e a rs in one place, on one fa d e r,and is a lot less distracting duringm i x d ow n . In fa c t , comping is nothingm o re than bouncing from many dif-f e rent source tra cks one at a time tothe same destination tra ck .

The mu l t i t ra ck can do more thanjust re c o rd instruments, other tra ck s ,and submixes. Why not re c o rd someeffects to the mu l t i t ra ck? If yo ustumble upon a truly magic effectthat you think may be difficult tore p ro d u c e , re c o rd it to its own tra ck s .

Sometimes it’s necessary to re c o rdthe effect because yo u ’ re borrow i n gthe $3500 Spastron Digital Nirva n aB ox and it has to be returned tonight.M o re like ly it’s because the totaleffect uses an elab o rate signal patht h rough 14 different effects units, a n dthough you actually own them, t h eexact settings may be difficult tore c re a t e . Printed effects are a goodh abit when you have spare tra cks andh ave created a rather dramatic effect.

Fi n a l ly, t h e re is no reason thee n t i re mix itself can’t be re c o rded onthe mu l t i t ra ck . N a t u ra l ly, the mix isre c o rded to the 2-tra ck masterm a ch i n e , be it DAT, d i s k , or analogt a p e . But if you have spare tra ck s ,print a safety ve rsion of the mix onthe mu l t i t ra ck . B eyond the comfo r tof having a backup copy, you cre a t ethe basis for a fa s t , and there fo rech e a p, re c a l l .

Yo u ’ ve no doubt experienced thetemptation to recall a mix.Yo u , t h ea r t i s t , or the label decide a week ort wo after mixdown that eve rything inthe mix was gorg e o u s . . . but it justneeds a little ex t ra reverb on thes n a re , and some slap echo on theslide guitar would be nice.

These changes genera l ly re q u i rethe entire mix to be re c a l l e d . That is,the same studio full of the same gearhas to be re s t o red to the exact samec o n f i g u ration it was in the day yo um i x e d , knob by knob, sw i t ch bysw i t ch . And all manual moves mu s tbe re - p e r formed ex a c t ly as befo re .

This ain’t triv i a l . I t ’s difficult to geta recall to truly match the originalm i x . Often the best you can do is get“close enough for ro ck and ro l l ”a n dm ove on. If the entire mix is on twot ra cks of the mu l t i t ra ck , the recall isp retty triv i a l .

Push up the two fa d e rs with theoriginal mix on them, and use thes n a re and slide guitar tra cks ass o u rces for the additional effects. M i x

Excerpted from the September edition of RECORDING magazine.©1999 Music Maker Publications, Inc. Reprinted with permission.

5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119



the new effects with the old stereo mix, and for just a fewm i nutes wo rk yo u ’ ve got a new - a n d - i m p roved mix that willplease eve ryone for at least another week or two.

Pros and consWe can see where a mu l t i t ra ck is a core part of yo u r

studio and its opera t i o n s , and the right machine will letyou do eve rything from storing and assembling passes torecalling entire mixes. So how do you choose the rightm a ch i n e ?

When evaluating which mu l t i t ra ck to buy, re n t , or bor-row, the normal priorities apply.You’ll be looking for thebalance of sound quality ve rsus price that fits your budget.

When evaluating the cost of a mu l t i t ra ck , do keep inmind the cost of the media as well as the cost of them a ch i n e .Yo u ’ ve got the one-time cost of buying the tapem a chine to take care of firs t , but the per- p roject cost of thetape (or disks) to justify to yo u rself or the clients later.

B eyond this value calculation, some other feature sshould be given due considera t i o n . The mu l t i t ra ckre c o rd e r, whether tape- or disk-based, is a mech a n i c a l lycomplicated dev i c e . U n l i ke , s ay, a digital mu l t i e f f e c t su n i t , a mu l t i t ra ck has moving parts, and lots of them. I tneeds to be well designed and well maintained.

For a tape mach i n e , look for a manu fa c t u rer you or acolleague knows to be trustwo r t hy. If new, m a ke sure ithas a wa r ranty to get you through any manu fa c t u r i n gfa u l t s . If used, t ry to assess the amount of loving care ors l o p py abuse the machine endure d .

If the machine hiccups during day one of an album yo umight lose the gig. If the machine crashes on day 231 yo umight lose all the audio for the entire pro j e c t . I t ’s toughto put a price on re l i ab i l i t y, so give it some thoughtb e fo re you tra n s a c t .

Related to quality is the ‘feel’ of the mach i n e . H owq u i ck ly does it fast fo r wa rd and rewind? Is it coopera t iveor cantanke rous as you re c o rd a ch o r u s , rewind to thebeginning of the ch o r u s , re - re c o rd the ch o r u s , rewind tothe second repeat of the wo rds “ . . . b aby, ye a h . . . ,” re -re c o rd the wo rds “ . . . b aby, ye a h . . . ,” rew i n d , re p e a t , e t c .

The process of re c o rding a song is fa i rly active andve ry non-linear, and the mu l t i t ra ck needs a tra n s p o r tthat can keep up with the cre a t ive needs of the session.Some feel like Italian sports cars , anticipating your eve ryre c o rding desire; some drive like an old school bus with aflat front left tire .

The decision to use a digital or an analog mu l t i t ra ckshould re a l ly be governed by how it meets the ab ove cri-t e r i a . Choose the machine you feel gives you appro p r i a t esound quality for the dollar while offering acceptab l ere l i ability and a comfo r t able tra n s p o r t .

If you go analog , you get the ability to edit tape. A l lyou need is a ra zor blade for cutting, some special tapefor taping the tape to another, e r, piece of tape, and anediting block to help you cut consistently. With appro p r i-ate equipment, clean and steady hands, and a willingness

to live on the edge, you can litera l ly cut and paste thetape to change the song in any way you desire . S wa pve rse one for ve rse thre e . Cut the solo to half its length.Pull out two bars of the intro. ( D o n ’t shudd e r — i t ’s howwe all did it, until ve ry re c e n t ly. )

Should you go digital, yo u ’ ve got the additional decisionof whether to be tape- or disk-based. C o m p a red to hardd i s k s , tape-based formats are genera l ly mu ch less ex p e n-s ive , a re often more portab l e , and they offer the ability tochange instantly from one project to another. C h a n g i n gtapes is a lot easier and quicker than backing up one harddisk project and restoring another (unless you sw i t ch hardd r ive s ,w h i ch can also be an option, if an ex p e n s ive one).

Splicing digital tape is possible on the (ve ry ex p e n s ive )open reel fo r m a t s , but is verboten for helical scan, c a r-tridge-based tapes like the ones used in DAT, DTRS (DA -88 series), and A DAT fo r m a t s . H a rd disks ove rcome thatd i s a dvantage by offering nondestructive softwa re - b a s e de d i t i n g .‘ N o n d e s t r u c t ive’ means that the edits don’t alterthe actual audio file—they ’ re done in softwa re “on thef ly,” so yo u ’ re free to change your mind.

Th e re are other advantages to hard disk re c o rd i n g .These include eve r- d e c reasing prices, external and re m ov-able drives to improve portab i l i t y, and random access.

You are familiar with the appeal of random access ifyou have ever tried to zip to the fifth song on an albumon both CD and cassette. The CD gets there instantly ande f fo r t l e s s ly. On cassette you fumble for the location, fa s t -fo r wa rding and rewinding until you find it.

Tape-based mu l t i t ra ck machines have the same pro b-l e m — o n ly there ’s the added benefit of a room full of peo-ple waiting for you to find the right place on tape. Ta p ec o u n t e rs , m e m o ry locations, and good notes on a takesheet can make this less of a headach e , but disk-basedre c o rd e rs offer true random access. Want to hear tra cksix? Click . H e re it comes.

Another type of mu l t i t ra ck re c o rder we have n ’t men-tioned is the DAW (digital audio wo rk s t a t i o n ) . These sys-tems are in the hard disk category, but they ’ re re a l ly awhole subject unto themselve s . Th ey combine all thea dvantages of stand-alone hard disk re c o rd e rs with ex c e l-lent editing. And tra ck bouncing in a DAW is differe n t ,because you can do it without deleting the originalt ra ck s . If this sort of power appeals to yo u , you shouldfo l l ow the ‘ DAW Diaries’ in this magazine for tips andt r i cks that have wo rked for my fellow writers .

In my re c o rding life I use both tape and hard diskm a ch i n e s . I guess I’m showing my age, but I prefer the‘vibe’ that comes from using tape-based mu l t i t ra ckm a ch i n e s . R ewinding befo re re-taking a solo gives the ses-sion a sort of pace that I find natura l . Instant access to thebeginning of the solo makes it all to easy to wo rk way toofast and lose the chance to take a breath and be cre a t ive .

On the other hand, random access locating and itsassociated nondestructive editing are clearly a powe r f u lp roduction tool. The choice is yo u rs , and a bit of pra c t i c eand fo rethought will help you pick the mu l t i t ra ck that’sright for yo u .

Alex Case welcomes suggestions for ‘ N u t s & B o l t s .’ You cancontact Alex at case@re c o r d i n g m a g . c o m .

M u lt i t rack Re c o r d e r s

The mult i t rack gives us that much-appreciated second—

or third, or fifteenth— chance to get a better ta k e .

M u lt i t rack Re c o r d e r s

RECORDING NOVEMBER 1999

w h i ch today usually has its back head re m oved so that itis completely open on the audience side. The single-head-ed kick drum provides a useful picture for seeing howacoustic sound is cre a t e d .

A component of the acoustic wave begins eve ry timethe drummer hits the kick drum. The beater strikes theh e a d . The initial motion of the drum head is towa rds thea u d i e n c e , squishing the air together immediately in fro n tof the drum head. This is an increase in pre s s u re thatradiates outwa rd towa rd the audience.

Fo l l owing the initial attack of the beater striking theh e a d , the head vibrates fre e ly back and fo r t h . Th eunique pattern and speed of this vibration spells out thatch a racteristic ‘thump’ we all know and love . Within thisp a t t e r n , e a ch motion of the drum head towa rd the audi-ence creates a tempora ry increase in pre s s u re , while therecoil of the head away from the audience creates ad e c rease in pre s s u re .

In this way a series of compressions and ra re factions isc re a t e d . The compressions re p resent a tempora ry andu s u a l ly ve ry slight increase in pre s s u re re l a t ive to thes i l e n t , u n d i s t u r b e d , ambient pre s s u re that had been inthe room befo re the music started. L i kew i s e , the ra re fa c-tion re p resents mere ly a decrease in pre s s u re re l a t ive toambient pre s s u re . It is not a total va c u u m , just a pocke tof air pre s s u re that is just slightly lower than it wo u l dh ave been in silence.

Those changes in pre s s u re push our ear drums in andpull them out so that we can hear the beat—and tap ourf e e t . When it’s wo rking well we call this mu s i c . But it getsa little messy when we take this concept to logical ex t e n-sions beyond the kick drum.

The acoustic sound of the piano is created by themotion of its soundboard in air, w h i ch is itself motiva t-ed to move by the elab o rate mach i n e ry around it (fin-g e rs , key s , h a m m e rs , s t r i n g s , and the like ) . Same goesfor the guitar and the violin. The player makes stringsv i b ra t e . The strings (through the bridge) push thesound board up and down and eve rything connected toit starts mov i n g , changing the air pre s s u re around it.

The re s u l t , s o m e h ow, is mu s i c .

It is the job of the microphone to capture this complexpattern of changes in air pressure and convert it into anelectrical property we can manipulate (amplify, equal-ize, compress, distort, delay, and so on). The microphonecreates in the electrical domain an analogy for what hadbeen happening in the air—hence the term analogaudio. The microphone maps air pressure changes intovoltage changes.

The idea is that a microphone in a silent room puts out0 vo l t s . As music play s , the positive air pre s s u re is con-verted into a positive vo l t a g e . In the subsequent ra re fa c-t i o n , w h e re the air pre s s u re is a negative (i.e. b e l owambient) air pre s s u re , the micro p h o n e ’s output is a nega-t ive vo l t a g e .

R e a l ly high pre s s u re displacements lead to higher vo l t-a g e s .E x t reme reductions in pre s s u re produce highamplitude negative vo l t a g e s . The mic cable then containsa pattern of voltage changes that are identical in shapeto the pattern of air pre s s u re changes that occurred atthe microphone capsule.

Interacting with airH ow does a mic go from air pre s s u re patterns in to

voltage patterns out? The voltage part we cove red in lastm o n t h ’s column. In the studio, we genera l ly employ am oving coil, r i bb o n , or condenser apparatus to create ourvoltage output based on the motion induced on somecapsule by the air.

Starting from there , you can ach i eve a total under-standing of how microphones wo rk by unders t a n d i n gh ow the capsule interfaces with the air. I t ’s important tou n d e rstand what is pushing the coil, m oving the ribb o n ,or flexing one side of the capacitor.

With the exception of the ribbon micro p h o n e , it is per-f e c t ly appropriate to picture the diaphragm of a micro-phone as a taut, round membrane like a drum head.( Ap o l ogies to Swe d e n ’s Pe a rl Lab s , who put re c t a n g u l a rd i a p h ragms in their mics.) It is suspended from its cir-c u m f e rence and free to move most at its center. If yo uever had the pleasure of playing on a round tra m p o l i n e ,yo u ’ ve got total, intimate knowledge of how a capsuled i a p h ragm behave s .


Microphones 2– Measuring Air Pressure and Air Ve l o c i t y(Which way is up?)

Consider the mighty kick drum,


M u ch of a micro p h o n e ’s behavior is determined by thefo l l owing simple distinction: is the diaphragm open to airon one side or both? Fi g u re 1 demonstrates this distinc-t i o n . The upper capsule shows a diaphragm that is openon one side but blocked on the other. The lower capsuleis open to the acoustic pre s s u re on both sides.

The figure also shows a particularly illustra t ive snap-shot of an ongoing acoustic wave moving across thee n t i re figure from left to right. The capsules are orientedso that they are both open on the left side; they are ‘ l o o k-ing’ towa rd the oncoming wave .

The top capsule has a compression wave immediatelyin front of it. This instant of high pre s s u re pushes thed i a p h ragm of the capsule inwa rd , to the right. S i m i l a rly,the lower capsule sees a higher pre s s u re to the left thanit does to the right, so it too is pushed to the right.

So far the two types of capsule seem to behave identi-cally. Consider Figure 2, which shows the two micro-phones rotated 90° so that they are oriented upward. Asthe acoustic wave rolls by in this instance, the uppercapsule is again pushed inward as the pressure on theopen outside of the diaphragm is greater than theenclosed inside.

The lower capsule, on the other hand, sees the samehigh pre s s u re on both sides of the diaphra g m . The inter-esting result is that this diaphragm doesn’t move at all—ito n ly moves when there is a pre s s u re difference betwe e nthe two sides. The upper capsule measures pre s s u re . Th el ower capsule measures a pre s s u re differe n c e , or to bem o re mathematically pre c i s e , a pre s s u re gra d i e n t .

N a t u ra l ly, the lower capsule is not typically called ap re s s u re difference or pre s s u re gradient mic, at least notin ro ck in ro l l . I n s t e a d , it goes by the slightly coolername: a velocity tra n s d u c e r.

It is perhaps intuitive ly obvious that whenever there isa pre s s u re difference in the air (that is, w h e n ever thereis noise), the air particles themselves move from theregion of high pre s s u re towa rd the region of lower pre s-s u re . Th ey don’t get far because the high and low pre s-s u re points are changing constantly, but they start mov-ing any way.

So it would be fair to say that wherever there is an airp re s s u re differe n c e , t h e re is also air particle motion. I nother wo rd s , it is appropriate to think of the lower cap-sule as responding to the motion of the air particles,rather than measuring the pre s s u re difference betwe e nthe two sides.

You can think of the velocity transducer as being like aflag or a sail that responds to the air blowing against it.U n l i ke flags and sails, t h o u g h , the velocity diaphra g m

flaps in the wind at audio frequencies—perhaps as slow-ly as 20 times per second and as quick ly as 20,000 timesper second.

These two types of tra n s d u c e rs , p re s s u re and ve l o c i t y,a re both perfectly capable of converting music into vo l t-a g e s . Both types are common in any studio’s mic closet.But there are differences between them.

DirectionalityThe physical orientation of the capsule itself is funda-

mental to determining its dire c t i o n a l i t y. A key differe n c eb e t ween the pre s s u re microphone and the velocity micro-phone has alre a dy been demonstrated in Fi g u res 1 and 2.The pre s s u re mic (the upper capsule in each figure )reacts to sound coming from in front or from the side.

In fa c t , it responds to pre s s u re waves no matter whattheir angle of arriva l . Being equally sensitive to soundsf rom all dire c t i o n s , it earns the moniker o m n i d i re c t i o n a l .


Figure 1: The upper capsule is open on one side only,

measuring pressure. The lower capsule is open on both sides,

measuring velocity

Figure 2: For sound from the side, the diaphram of

the upper capsule is displaced. The lower capsule rests, completely uneffected.


Microphones 2–

The velocity mic (the lower capsulein each figure) demonstrates an ab i l-ity to ‘hear’ sound arriving from thef ro n t , yet it ignores sound comingf rom the side.

The arrangement in which the cap-sule is open on both sides is mosts e n s i t ive to sound coming straight atthe diaphra g m — f rom the front orthe rear—and least sensitive tosound coming from the sides. Th em i c ’s sensitivity decreases gra d u a l lyas sound sources move off-axis fro mf ront to side.

To understand this better we needto graph it on polar coord i n a t e s . I fwe plot the sensitivity of the micro-phone as a function of the angle ofa r r ival of the sound from the sourc e ,we can make visual the dire c t i o n a ldiscrimination properties of the mic.

Fi g u re 3 shows the three polar pat-terns we most often see in the stu-d i o. And parts A (omnidire c t i o n a l )and B (bidirectional) we ’ ve just dis-c u s s e d . The omnidirectional pick u ppattern shown in 3A is equally sensi-t ive at all angles, and is a natura lresult of being a pre s s u re tra n s d u c e r.

The bidirectional pattern (alsocalled the f i g u re - e i g h tpattern) show st wo points of maximum sensitiv i t yd i re c t ly in front of and behind thec a p s u l e , diminishing sensitivity asthe angle of arrival goes towa rd thes i d e , and finally total rejection fo rsounds fully at the side.

The bidirectionality of the mic is aby p roduct of being a velocity tra n s-d u c e r. It only measures the move-ment of particles against it, i g n o r i n gparticle velocity that moves along-s i d e , p a rallel to the diaphragm itself.

But there is a little more to theb i d i rectional pattern. The front andthe back lobes of the figure - e i g h tpattern are not ex a c t ly the same.

You can achieve a

t o tal under-

s tanding of how

mics work by

u n d e r s ta n d i n g

how the ca p s u l e

i n t e r faces with

the air.

Excerpted from the November edition of RECORDING magazine. ©1999 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119



Fi g u re 4 shows a velocity micro-p h o n e ’s reaction to a given soundwave as it propagates left to right.Fi g u re 4A orients the mic facing leftinto the sound. The higher pre s s u re ,left ve rsus right, suggests the air par-ticles and the diaphragm will movet owa rd the right. This motion will cre-ate an output voltage of, s ay, one vo l t .

Consider placing the same micro-phone on the same sound wave atthe same time, but facing the oppo-site dire c t i o n . The higher pre s s u re onthe left still pushes the diaphragm tothe right. The air causes the samep hysical motion. But from the per-s p e c t ive of the micro p h o n e , t h i sidentical sound has caused thed i a p h ragm to move the opposited i re c t i o n .

Motion of the diaphragm thisd i rection will in fact create a nega-t ive vo l t a g e . An appropriate conclu-s i o n , t h e n , is that the velocity micro-

phone can be equally sensitive tosounds in front of or behind the mic,but it picks up sound from behindwith reve rse polarity. In front of them i c , a positive pre s s u re creates ap o s i t ive vo l t a g e , while behind them i c , a positive pre s s u re creates an e g ative vo l t a g e .

Look that wayIt is this reve rse polarity of fro n t

ve rsus back that enables us to cre a t ea u n i d i rectional m i c ro p h o n e . Fi g u re3C shows this type of pickup pattern,w h i ch is most sensitive in only oned i re c t i o n .

This is quite helpful in the studiowhen you wish to re c o rd seve ra linstruments at once, but each to itsown tra ck . When physical isolationi s n ’t ava i l able in the form of isola-tion booths, e n g i n e e rs ach i eve a sortof acoustic isolation by using unidi-

Microphones 2–

Figure 3: Microphone Polar patterns


rectional microphones aimed dire c t ly at their intendedi n s t r u m e n t s , rejecting/minimizing the unwanted neigh-boring instruments.

If you add up the response of an omni to that of a fig-u re - e i g h t , you end up with the cardioid response shown inFi g u re 3C. ( I t ’s called a ‘ c a rdioid’ because, to someonewho knew Latin, it looked heart shaped. But it’s a pre t t yf u n ny looking heart, not the sort of heart that would sella valentine greeting card . I guess there wa s n ’t a Latin-based way to say “Looks kind of like a pizza with oneslice missing.” ) .

Want to build a cardioid response? Grab a pre s s u ret ransducer (or any omnidirectional microphone) and avelocity transducer (or any bi-directional micro p h o n e ) .Place them as near each other as possible, facing thesame way, and mix them together onto one tra ck . If yo umonitor with the two microphones at equal amplitude,you’ll have created a cardioid pick-up pattern using a 2-mic combination.

To see how a cardioid pattern is born, look closely atsome landmark points in the response of the two compo-nent patterns of Fi g u res 3A and 3B. D i re c t ly in front ofthe microphone you get a contribution from both cap-s u l e s . Off to each side, o n ly the omnidirectional pre s s u ret ransducer picks-up sound.

Behind the microphone you have the contribution of theo m n i d i rectional piece being undone—litera l ly cancelled—by the polarity-reve rsed rear lobe of the bidirectional mic.Placing a pre s s u re capsule and a velocity capsule in thesame place and combining them gives you double the sen-s i t ivity in front of the pair and total rejection to the re a r.

I t ’s a good trick . The downside is that you get a singlemic for the price (and noise floor) of two. Th e re ’s anotherway to do it that re q u i res only a single capsule.

Since the goal of a unidirectional microphone is toreject sound coming from behind, c l ever manu fa c t u re rsh ave modified the velocity tra n s d u c e r. The sound comingf rom behind the microphone needn’t re a ch thed i a p h ragm dire c t ly.

It is possible to delay the components of sound re a ch-ing each side of the diaphragm so that for sources behindthe mic, t h ey arrive at ex a c t ly the same time. Ports intothe microphone are configured so that there is no dire c tpath from the rear of the microphone to the rear side ofthe diaphra g m .

The arriving sound must navigate the short detour ofan acoustic labyrinth on its way to the back side whiles i mu l t a n e o u s ly wrapping around to the fro n t . If the timeit takes the wave to diffract around to the front of themic is equal to the time it takes the same wave to re a chthe back of the diaphragm via the longer path, t h ed i a p h ragm will not move .

When the diaphragm is pushed from the front by thep o s i t ive portion of a cy c l e , it is simu l t a n e o u s ly pushedf rom the rear by the positive portion of the cy c l e . Th i spush/push phenomenon emulates the situation of Fi g u re2 in which sound arriving from the side presents thesame pre s s u re on both sides of the velocity diaphra g m .

Mission accomplished: acoustic manipulation of thesignal ach i eves rejection from behind. P retty darn cleve r.

But there ’s a little more to it. For this modified micro-phone to be of any use, sound arriving from sources thata re in front of the microphone must still be effective atm oving the diaphra g m .

This is ach i eved by making sure that the front ve rs u sb a ck time-of-arrival difference at the diaphragm fo rsound arriving from the front of the capsule is ex a c t ly(or nearly) equal to 180° of phase differe n c e . In this waythe wave form is presented to both sides of thed i a p h ragm in a complementary way. When thed i a p h ragm is pushed from the front by the positive por-tion of a cy c l e , it is simu l t a n e o u s ly pulled from the re a rby the negative portion of the cy c l e .

Not only does sound arriving from the front of themicrophone still move the diaphragm, but it does so in


Figure 4: Reversing the orientationof a velocity microphone reversesthe polarity of the output signal

All studio microphones are

either pressure tra n s d u c e r s ,

velocity transducers, or

some combination thereof...

...And all of the intermediat e

mic patterns can be creat e d

by mixing variable amounts of

two types of tra n s d u c e r s .


this push/pull fashion—it simultaneously pushes on oneside and pulls on the other. That translates intoincreased sensitivity.

This clever manipulation of the wave form as it re a ch e sboth sides of the velocity transducer leads to a card i o i dpattern: enhanced sensitivity in the fro n t , total re j e c t i o nat the re a r. In fa c t , all this acoustic signal pro c e s s i n gtries to make a single capsule that is half sensitive top re s s u re and half sensitive to ve l o c i t y. It is the acousticcombination of the two microphones we combined elec-t r i c a l ly ab ove . By using a single capsule, t h o u g h , i taccomplishes this at a mu ch more appealing price.

M a ny single diaphragm cardioid microphones (thefamous Shure SM57 and Electro - Voice RE20, among them a ny good examples) offer a good visual ex a m p l e . It iseasy to see the ports on the body of the microphone thata re the entry points for the sound into the back of thed i a p h ra g m .

Look this wayBy mixing differing amounts of pre s s u re and ve l o c i t y

t ra n s d u c t i o n , other polar patterns can be cre a t e d . We ’ veseen how equal parts pre s s u re and velocity produces ac a rd i o i d . M o re pre s s u re than velocity leads to a dire c t iv i-ty that is, not surp r i s i n g ly, m o re omni than card i o i d .Called a s u b c a r d i o i d, it is slightly more sensitive fro n tve rsus back; it partially but not completely re j e c t ssounds behind it (Fi g u re 3D).

C o nve rs e ly, h aving less pre s s u re than velocity tilts thebalance towa rd the bi-directional pattern. This mored i rectional pattern is usually called a hy p e r c a r d i o i d( Fi g u re 3E). It is more sharp ly focused fo r wa rd . B e c a u s eit is less pre s s u re than ve l o c i t y, h oweve r, t h e re is nolonger perfect cancellation at the re a r. The hy p e r- c a r-dioid develops a small rear lobe of sensitivity that is theresidual rear lobe of the bidirectional component.

Enhanced fo r wa rd sensitivity comes at the expense ofdiminished re a r wa rd re j e c t i o n . You’ll no doubt find spe-cific session situations where these other patterns arejust what you need.

As befo re , all these interme-diate patterns can be cre a t e dby mixing va r i able amounts oft wo types of tra n s d u c e rs , u s i n gt wo mics and a mixer. The re l a t ive levels of the two micsdetermines the degree of omni ve rsus bidirectional in thenet polar pattern. A l t e r n a t ive ly, sub- and hy p e rc a rd i o i dpatterns can be created on a single capsule by acoustical-ly mixing the two types of transduction through thec l ever design of ports re a ching the rear of the diaphra g m .

A particularly good visual case study comes via thevenerable microphone manufacturer Neumann. Theyrecently released small diaphragm omnidirectional andhypercardioid mics to complement their well known car-dioid, the KM184. Check out the photo of the completeSeries that shows the mics side by side. The only visibledifference among them is the rear ports.

Know it allAll studio microphones are either pre s s u re tra n s d u c-

e rs , velocity tra n s d u c e rs , or some combination there o f. I na dd i t i o n , most all studio mics employ one of the fo l l ow-ing design types: moving coil, r i bb o n , or condenser.

We ’ ve spent two months digging into these conceptsand found that within all of these types of micro p h o n e sl ives a knowab l e , s t ra i g h t fo r wa rd pro c e s s . Armed withthis knowledge of the physics behind the tech n o l og y, n ex tmonth we’ll discuss the basic specifications, f e a t u res andsw i t ches you might find on a micro p h o n e . You’ll find theyall stem from these microphone fundamentals.

Deciding which microphone to buy or which micro-phone to use on a specific instrument in a specific situa-tion will depend on your knowledge of this basic pro c e s sof transduction from acoustic to electric energ y, in com-bination with your feeling about what sounds best.

Alex Case wants to know wh at you want to know.R e q u e s tNuts & Bolts topics via case@re c o r d i n g m a g . c o m .


Figure 5: Cardioid pick-up is achieved throughacoustic manipulation of sound reaching each side

of the diaphragm

Want to build a cardioid response? Grab any omni mic and

any bi-directional mic. Place them as near each other as

p o ssible, facing the same way, and mix them onto one tra c k .


Microphones 3– s p e c i f i cations and controls

RECORDING DECEMBER1999

In the re l a t ive ly stra i g h t fo r wa rd case of re c o rding ap ower trio, you might have to select maybe a doze nm i c ro p h o n e s . The trio: drums, b a s s , guitar and vo c a l s .The mics: kick , s n a re , h a t , t h ree toms, t wo ove r h e a d s , t woout in the ro o m , b a s s , guitar and vo c a l .

Selecting the right mic for the job is an eve r- p resent partof the re c o rding gig.The creation of an album invo l ve smaking this small decision maybe hundreds of times.

In the last two episodes of ‘Nuts & Bolts’ we ex p l o red theinner wo rkings of micro p h o n e s . This month we tackle themeaning of the various specifications, and the function ofthe various controls that might appear on a micro p h o n e .

The goal is to convert microphone selection from a ra n-d o m , l u ck - o f - t h e - d raw process into an org a n i zed systembuilt on total knowledge of all microphone tech n o l og i e sand para m e t e rs .

Frequency responseSelecting a mic re a l ly begins with its fre q u e n cy

response; we want to know how it sounds. A fre q u e n cyresponse plot is the first view into this. This descriptionof the micro p h o n e ’s output at different fre q u e n c i e sreveals any biases for or against any particular fre q u e n cyra n g e s . Fi g u re 1 offers a few samples.

The oft-cited ‘color’ of a microphone is ve ry mu ch deter-mined by its fre q u e n cy re s p o n s e . Try to have in mind arough sense of the fre q u e n cy response of eve ry mic yo uu s e .You can store the data (in your bra i n , that is) visually,l i t e ra l ly picturing fre q u e n cy response plots in your mind.

A l t e r n a t ive ly, you can store the data in wo rds: wa r m ,b ox y, p re s e n t , e d g y, a i ry. As your experience grow s , t h e s ewo rds develop a ve ry precise meaning. As time goes byand natura l ly you acquire more mics, you’ll need to addn ew wo rds to your lexicon to be more pre c i s e . I t ’s notjust wa r m — i t ’s thick , t u bby, b i g , p h a t , p u n chy, h e av y, o rsome such . I t ’s not airy; rather it’s bre a t hy, it shimmers , i ts o a rs , it spark l e s . . . .

So the fre q u e n cy response plot is a good starting pointfor learning the ‘sound’ of the dev i c e . But your pro f e s-sional development will alway s — for the rest of yo u rlife—include refining your own internal sense of thesound of each make and model of micro p h o n e .

E a ch session will reveal ever more . A mic with a lowend hump in its fre q u e n cy response might sound punchyon congas, but mu rky on bass. This helps you ze ro in onthe exact shape and location of that low fre q u e n cyh u m p.

Wa r n i n g: it is a hazard of this job that the wo rds yo uuse to describe a sound like ly have a different meaningto someone else. Your sense of the sonic ch a racter ofsomething that is ‘ t h i ck’ could be slightly—and some-times ve ry — d i f f e rent than someone else’s internal con-ception of the sound.

Audio ear training (à la Dave Moulton’s Golden Ears)in combination with professional interaction with otherswhose work you admire will help these descriptivewords hover near a common definition. Just beware ofthe difficulty when you try to communicate withstrangers on the topic, be it fellow engineers or a micro-phone salesperson.

The handy thing is, when it comes to the selection of am i c rophone by yo u , your own internal meaning for thewo rds is correct and sufficient.Your descriptions of thesonic ch a racter of each mic are all you need to make agood guess at which mic will sound best on today ’s ove r-d u b .

So push yo u rself to develop a feeling for the fre q u e n cyresponse of eve ry mic you own or have access to. A n dc o n s t a n t ly refine your internal sense of its fre q u e n cyresponse towa rd an ever more precise meaning. Th e‘color’ of a mic is a ve ry pers o n a l , ve ry detailed concept.

Off-axis colorationN a t u ra l ly, the fre q u e n cy response plot that comes with

the microphone and that lives in our head is an ove rs i m-plification of the complex behavior of the tra n s d u c e r.One frustrating point is that the fre q u e n cy response of am i c rophone changes with the angle of the sound’s arriva lat the mic from the sourc e .

While I encourage you (and constantly remind my s e l f )to consider omnis and figure-eight patterns more often inthe pop/ro ck re c o rding studio, the fact is, c a rdioid pat-terns preva i l . But there is a potential hazard to that car-dioid pattern that needs scrutiny.



The idea of a uni-directional pattern is that the micro-phone is focused most on sounds directly in front of themic. Sounds arriving from the side are attenuated.Andsounds arriving from the rear are rejected. That’s thetheory.

The fact is, off-axis sounds are n ’t just attenu a t e d — t h eoff-axis fre q u e n cy response of a microphone is often dif-f e rent from the on-axis fre q u e n cy re s p o n s e . The result is

that sounds arriving from any w h e re but in front of themic are spectra l ly altere d . S p e c i f i c a l ly, most cardioid pat-terns are better at rejecting high frequencies off to theside and behind than low fre q u e n c i e s .

Said another way, the cardioid microphone is more of atrue cardioid at high frequencies and more omnidire c-tional at low fre q u e n c i e s .

Fi g u re 2 (see p. 5 6 )d e m o n s t rates a pretty typical fre-q u e n cy trend in card i o i d s .The response dire c t ly in front ofthe mic is consistent from low frequencies to high—a ‘ f l a t ’f re q u e n cy re s p o n s e , o n - a x i s . But beside and behind them i c , the microphone attenuates the highs more than ita t t e nuates the low s . This off-axis coloration means the micis effective ly acting like an equalizer for the sounds com-ing from all around it. I t ’s rolling off the highs, yet hanging

on to the lows and doing something inb e t ween for the middle fre q u e n c i e s .

The significance of this behavior can-not be ove rs t a t e d . When you place ad i rectional microphone near a sourc e ,you still re c o rd sounds arriving from thes i d e s . Sometimes that ‘leakage’ of othersounds into this microphone is inaudiblyl ow, but other times you can hear it.When the instruments you are re c o rd i n ga re re q u i red to be ve ry near each other,you’ll get an unavo i d able amount of leak-age from each instrument into the neigh-boring mics.

C l o s e - m i ked drum kits are the mostcommon situation where this occurs . Th ehigh-hat is always near the snare . Th es n a re is often ve ry near the top ra ck tom,

and so on. M o re ove r, these instrumentstend to be quite loud, fo rcing themselve son eve ry microphone in the zip code.Leakage ab o u n d s .

But it’s not just drums that re q u i re us toconsider this acoustic leakage issue. Wh e nwe wo rk with loud instruments (e.g. h o r n s ,p e rc u s s i o n , and the obligatory electric gui-tar) or in tight quarters (small booths, a n dmost home or project studio re c o rding sit-u a t i o n s ) , we get significant off-axis soundinto our directional micro p h o n e s .

In all cases, if that off-axis sound isdull or mu rky it will drag down the soundof the mix.

And another thing (this one seemsm o re obvious but is all too often neglect-ed): we often re c o rd off-axis sound on

p u rp o s e . D r u m s , electric guitars , sections (of horns,s t r i n g s , vo i c e s , etc.) and many other tra cks welcome theplacement of some distant mics for re c o rding the ambi-ent sound in the ro o m .

In these situations, choose a mic that welcomes off-axissound and doesn’t impose an unappealing coloration ontothe sound.These ambient room mics are supposed to be fa r

Which microphoneshall we try ?

This question willsometimes fill yo u

with dread andpanic. (“I’ve never

recorded ac o n t ra b a ss o o n . . . ” )



easy to hear and easy to pre d i c t .P re s s u re tra n s d u c e rs do not ex h i b i tthis fre q u e n cy response-altering phe-nomenon at all; velocity tra n s d u c e rsd o. Th e re fo re , the apparent fre q u e n cyresponse of omnidirectional mics isunaffected by the closeness of thes o u rce to the mic.

P roximity effect is a property ofvelocity tra n s d u c e rs alone. Th a tmeans you can expect bi-dire c t i o n a lmics to add an amplitude boost inthe bottom fre q u e n cy range whenev-er they are placed ve ry near aninstrument (within about one tot h ree feet). C a rdioid mics, being halfp re s s u re sensitive and half ve l o c i t ys e n s i t ive , will also exhibit prox i m i t ye f f e c t , but with about half the bassboost of a pure velocity tra n s d u c e r.

H ow do you get rid of this bassboost? We l l ,f i rst decide if you wa n tto get rid of it at all.

Used with a little re s t ra i n t , t h eenhanced low end can make a vo i c e

sound larger than life. This is helpfulwhen advertising monster trucks h ow s , announcing sports, or justplain ol’ talking. DJs sound morei m p re s s ive (and taller) when theya re in close on their directional mics.Lead vocals in ro ck and roll and popre ly on this as we l l . Hit songs need tosound better than the original instru-m e n t s , better than re a l i t y.

Modern studio production tech-niques leve rage proximity effects e l e c t ive ly for many tra ck s .

Roll-offBut bumping up the low end isn’t

a lways a good thing. Getting in closeto a snare , a piano, or an acousticguitar can lead to an ove rly boomy,a n n oy i n g ly thumpy sound. When yo uhear this sort of pro b l e m , you arehearing an unwanted prox i m i t ye f f e c t . K n ow that backing the micaway from the instrument might beall it takes to solve the pro b l e m .

f rom the source and are genera l ly sup-posed to be picking up room re f l e c-tions coming from all dire c t i o n s .

Choosing an omnidirectional mic isone solution. But it is perfectly accept-able to want a directional mic toa ch i eve some re j e c t i o n . Keep in mindthe off-axis coloration the micro p h o n emight add . Choose one whose off-axisresponse enhances the ambient soundyou are trying to capture .

A subtle part of microphone ch o i c ethen has to do with the degree of off-axis coloration the mic imparts. As ifyou didn’t have enough to memorizeabout a micro p h o n e , n ow yo u ’ ve got

to learn its fre q u e n cy response as afunction of angle... D o n ’t sweat it,though—it will come over time.

Proximity effectO k ay, so yo u ’ ve got the fre q u e n cy

response of a microphone thoro u g h lyi n t e r n a l i ze d , both on- and off-axis.What next? Proximity effect: the lowf re q u e n cy accentuation that occurswhen a sound source is ve ry close toa directional (i.e. n o n - o m n i d i re c t i o n-al) micro p h o n e .

P roximity effect re p resents anothera l t e ration to the fre q u e n cy re s p o n s eof a micro p h o n e . Fo r t u n a t e ly, it is

In the relat i v e lys t ra i g h t f o r w a r dcase of recording

a power trio,you might

h ave to selectm aybe a dozenm i c r o p h o n e s .


When the close mic location is justright except for some unwanted lowend due to proximity effect, it ishelpful to kick in a high pass filter.A l l owing the highs thro u g h , the highpass filter attenuates only the low s .

Studio speak calls this ‘ ro l l - o f f,’ andm a ny microphones have a built-insw i t ch that does ex a c t ly this.Engaging the roll-off circuit re m ove sor diminishes a problematic prox i m i t ye f f e c t .

A dd i t i o n a l ly, it may be used simplyto get rid of unwanted low fre q u e n cysounds that sneak into a studio. A i rconditioning and traffic noise fro mh i g h ways or train tra cks are typicall ow fre q u e n cy headach e s .

The roll-off sw i t ch is a good solu-tion to these pro b l e m s . And whenthe mic doesn’t have a roll-off filterbuilt in, you can often find one onthe console or in an outboard micp re or equalizer designed for thesame purp o s e .

Listen care f u l ly when you engagea filter, because poorly designed fil-t e rs can affect the higher fre q u e n-cies audibly, even though they passt h rough the filter. If yo u ’ ve got thet i m e , the gear, and the ear tra i n i n g ,c o m p a re the highpass filter on yo u rm i c rophone to the highpass filter onyour console to any other highpassf i l t e rs your studio may have . Yo um ay find that all too common tre n dh e re: the more ex p e n s ive filterssounds better.

SensitivityI t ’s not just a New A g e , p o l i t i c a l ly

c o r rect term—microphone sensitiv i t ydescribes how mu ch output them i c rophone creates electrically for ag iven acoustic input. That is, if theassistant engineer screams at ex a c t ly90 dB SPL into a micro p h o n e , w h a tvoltage will come out? When theassistant screams at ex a c t ly 90 dBSPL into another micro p h o n e , w h a tvoltage comes out?

The more sensitive micro p h o n eg e n e rates a higher amplitude vo l t-age for the same sound pre s s u rel evel input. The hotter outputre q u i res less amplification at the micp re a m p, w h i ch can mean a lowe rnoise floor will be re c o rd e d . This is agood thing. On the other hand, p l a c-ing a ve ry sensitive microphone neara ve ry loud sound source can ove r-load the electro n i c s , causing distor-t i o n . This is (usually) a bad thing.

Think of sensitivity as a specifica-tion that re a l ly only needs to be wo r-ried about at its ex t re m e s . That is, i fyou know you must re c o rd a ve ry

quiet instrument (have you eve rg a t h e red sound effects like fo o tsteps in sand or water dripping?),seek out a sensitive micro p h o n e . I fyou know the instrument is ra g i n g lyloud (trombone comes painfully tom i n d ) , perhaps look for a less sensi-t ive tra n s d u c e r.

O t h e r w i s e , mic selection is more afunction of polar pattern, f re q u e n cyre s p o n s e , off-axis colora t i o n , e t c .

PadSometimes the pairing of a loud

sound with a sensitive micro p h o n eleads to distortion. Should the

It’s not justwarm—it’s thick,t u b by, big, phat ,

punchy, or heav y .It’s not airy ;rather it’s

b r e athy, it soars,it spa r k l e s . . . .

s p e c i f i cations and controls


acoustic energy hitting a micro p h o n eove rload the mic’s internal electro n-ics or the microphone pre a m p l i f i e r, apad can be engaged.

The pad offers a fixed amount ofa t t e nu a t i o n , s ay 10 or 15 dB, to thesignal just leaving the tra n s d u c e r.The lower voltage coming out of thet ransducer after the pad will (hope-f u l ly) no longer ove rload the elec-t ro n i c s , e n abling the microphone tobe used even on a louder sound.

M a ny microphones have the ab i l i t yto sound gorgeous on, for ex a m p l e ,both acoustic guitar and snare drum.A mic close to a snare drum mightencounter sounds well ab ove 130 dBS P L . A subtle nylon string acoustic

guitar might be a mere 40 dB SPL orless at the desired mic position. Th epad enables the same mic to be usedon instruments of such ra d i c a l ly dif-f e rent loudnesses. The pad is turnedon when re c o rding the snare andturned off when re c o rding the guitar.

Sometimes a pad isn’t enough;acoustic signals can become too loudfor the micro p h o n e . If the acoustics t i mulation of the transducer fo rc e sthe diaphragm into the ex t reme lim-its of its physical motion, it maybecome non-linear. That is, when thesound is too loud for the capsule, t h emotion of the diaphragm no longerfo l l ows the sound smoothly, rather itslams into the limits of its fre e d o mto move—it distorts mech a n i c a l ly.

When re c o rding ve ry loud soundsl i ke trumpets and space shuttles,look for a microphone with a ve ryhigh rated maximum sound pre s s u rel eve l . This indicates the point beyo n dw h i ch the microphone cannot tra n s-duce without distortion.

M o re ove r, putting the mic in asoundfield greater than this ra t i n gcould possibly damage the mic. Th em a x i mum SPL rating sets the upperamplitude limit for the dev i c e .

Session variablesWh i ch microphone shall we try ?

This question will sometimes fill yo uwith dread and panic. ( “ I ’ ve neve rre c o rded a contrab a s s o o n . . .”) Other

t i m e s , this question will fillyou with anticipation. ( “ Th i sn ew mic sounded great onA my ’s guitar. I can’t wait tot ry it on yo u rs .” )

It is possible to break themic selection process dow ninto smaller decisions.Yo u ’ ve got to get the bestsound possible. E x p e r i e n c eand ear training will helpyou match sound sources toc o m p l e m e n t a ry micro-p h o n e s .

Session re q u i re m e n t smight narrow your options,fo rcing you into a give npolar pattern. For ex a m p l e ,you might need to use a micwith a cardioid pattern ifyou have to put the sax play-er next to the piano.

G iven placements and pairingsm ay or may not re q u i re you to sw i t chin a roll-off or a pad, but your know l-edge of what these sw i t ches do incombination with your reaction towhat yo u ’ re hearing make thesedecisions pretty stra i g h t fo r wa rd .

The simplicity of the micro p h o n et e ch n o l ogy re q u i res that we masterjust a few concepts; the subtlety ofre c o rding acoustic sounds demandsthat we then proceed care f u l ly, w i t hour ears wide open.

Alex Case wants to know wh at youwant to know. Request Nuts & Boltstopics via case@re c o r d i n g m a g . c o m .Th a n k s .

Microphones 3–

The oft-cited ‘color’ of a mic is very muchdetermined by its frequency response.

Excerpted from the December edition of RECORDING magazine.©1999 Music Maker Publications, Inc. Reprinted with permission.



RECORDING OCTOBER 1999

Selecting the right microphone is aconstant part of the job. B a n d :

The Has Beens. Song #5: The Hair IUsed to Have . O ve rdub #16: ukulele.Wh i ch microphone should be used?

Depending on the studio, the engi-neer has to choose among maybe ad o zen or maybe even a hundre dm i c ro p h o n e s . Th ey come from count-less manu fa c t u re rs , offering seve ra lmodel nu m b e rs .

What will a given micro p h o n esound like on a particular instru-ment in a specific style of music inthis unique re c o rding space? A a a rg h !Th e re is no end to the possibilities.

One develops insight and intuitionabout which mic to try for a give nsituation through ex p e r i e n c e . But wecan help our experience along bylearning how they wo rk .

I t ’s helpful to break down the va s trange of microphone possibilitiesinto some subgro u p s . In the re c o rd-ing studio, the re c o rding engineert y p i c a l ly chooses among three typesof microphone designs: moving coil,r i bb o n , or condenser.

How do they work?U nusual in our wo rld of complicat-

ed gear (ever open up a digital 8-t ra ck ? ) , the microphone is an ele-g a n t ly simple, c o m p l e t e ly knowab l et e ch n o l og y. And knowing how thething wo rks gives us some insightinto how to use it.

A fascinating parallel betwe e nelectricity and magnetism exists andseems tailor-made for audio.Wh e n ever an electrical conductor—l i ke a wire — m oves through a mag-netic field, an electrical current isinduced onto it.

L eve raging this principle, c a l l e de l e c t romagnetic induction, you cang e n e rate your own electricity if yo uwa n t . Just persuade someone to hoponto a bicycle modified so that therear tire is a coil of wire . Set it up sothat the wire rotates through the gapof a magnet when he or she pedals.If he or she pedals hard enough andif the coil and magnet are bige n o u g h , you could power all yo u rfavorite equipment free (assumingyou don’t pay this pers o n ) . We don’tk n ow people willing to do that, s oinstead we have power companies.

Power companies use giant steam-powered turbines to spin generatorsthat rely on this same fundamentalphysical property. And not only doesa magnetic field induce a current ona wire that moves through it, butalso a changing current on a wirecreates a magnetic field around it.That is, electromagnetic inductionalso works in reverse.

Using electricity to create a mag-netic field is a basic necessity when

re c o rding music on magnetic tape orp l aying music back through a loud-s p e a ke r. M o re on all that in futureepisodes of Nuts & Bolts; for nowl e t ’s apply electromagnetic inductionto micro p h o n e s .

M i c rophones that re ly on electro-magnetic properties to convert anacoustic event into an electrical sig-nal are called e l e c t r o dynamic ( m o rec o m m o n ly ‘ dynamic’) mics. Th e re aret wo types of dynamic micro p h o n e sused in the studio: moving coil andr i bb o n . And they both are appealing-ly stra i g h t fo r wa rd dev i c e s .

The moving coil dynamic micro-phone converts sound into electricitywith essentially three components: ad i a p h ragm that moves with the air, acoil that is moved by the diaphra g m ,and a magnet that induces electricalc u r rent onto the coil when it move s .

This type of mic takes adva n t a g eof the motion of air particles duringan acoustic sound to move a coil ofw i re through the magnetic field of apermanent magnet. The coil move-ment creates an electrical signalwhose voltage changes as a dire c tresult of the acoustic eve n t . I t ’s a sat-i s f y i n g ly simple pro c e s s .

The ribbon microphone take sa dvantage of the same electro dy n a m-ic principle we ’ ve discussed. As am a chine that converts acoustic ener-gy into electrical energ y, it is eve nsimpler than the moving coil system.The ribbon microphone cleve rly com-bines the diaphragm and the coilab ove into a single device: a ribb o n .That is, the thing that moves in theair is also the conductor of electricity.

The ribbon is a piece of metal sus-pended between the poles of magnet.


M i c r o p h o n e s 1 —Transducer Designs

We can gain insight

about which mic to

t ry for a given

s i t uation by learning

how they work.


When a musical instrument play s , a i rmolecules move .The air moleculesnear the ribbon fo rce it to move; themotion of the ribbon through themagnetic field induces electrical cur-rent onto the ribbon itself.Vo l t a g echanges that are a perfect analogy tothe acoustic event are cre a t e d .

A third microphone tra n s d u c e rt e ch n o l ogy employed in the studiod o e s n ’t re ly on electro m a g n e t i cinduction at all. The condenser micro-phone relies on the electrical pro p e r-ty of capacitance instead. We knowthat if we hook up a voltage sourc e( e . g . a battery) across a wire , e l e c t r i-cal current will flow. If we cut thatw i re , the current stops.

It turns out that there is somethingin between a closed circuit (the wire )and an open circuit (the seve re dw i re ) . Imagine that after cutting thew i re we bring the two ends of thew i re re a l ly close to each other with-out touch i n g . I t ’s easy to imaginet h a t , without current actually flow-ing across the gap we ’ ve made in thew i re , the two ends would influencee a ch other electrically.

A capacitor is a component thatdoes this on purp o s e . Wh e re the wirewas bro ke n , plates of metal area t t a ch e d . And these two plates areb rought up ve ry close to each other,again without touch i n g . The result isthat an electrical ch a rge builds upon the plates, pulled by the influ-ence of the voltage source across thegap in between the plates.

The ability to store a ch a rg e , o rcapacitance (hence the name capaci-t o r ) , is a function of the vo l t a g ea c ross the plates, the size of thep l a t e s , and the distance between thet wo plates. As the plates separa t e ,t h ey become more like a fully bro-ken circuit and the ch a rge dissipates.As the plates conve rg e , t h ey have as t ronger and stronger influence one a ch other and try to appro a ch theb e h avior of a completed circ u i t — t h ech a rge on the plates then incre a s e s .

Once upon a time this type of elec-trical component was called a c o n-d e n s e r. While the component is todayg e n e ra l ly called a capacitor instead,the microphone built around thist e ch n o l ogy hangs on to the namec o n d e n s e r.

A condenser microphone is noth-ing more than a va r i able capacitord r iven by acoustic sound wave s . O n eplate of the capacitor is thed i a p h ragm whose motion is a re s u l tof the changing sound pre s s u rea round it. As the diaphragm move s ,the capacitance ch a n g e s . The electri-cal output of the microphone is apattern of voltage changes derive df rom this change in capacitance.

Mission accomplished: acousticmusic in, electrical signal out.

Which one do I use?Knowing the type of transducer

technology a microphone employsgives the engineer some insight intohow it might sound and what appli-cations it is best suited for. But letme preface this discussion withsome very important, really goodnews: we’re lucky to be in the audiobiz in 1999.

The quality of the design, m a t e r i-a l s , and manu facturing tech n i q u e sused today is enabling all micro p h o n et e ch n o l ogies to conve rge towa rd ac o n s i s t e n t ,h i g h - q u a l i t y, h i g h - d u rab i l i-ty pro d u c t .B e l ow I discuss some gen-e ral properties of microphones basedon the type of transducer used.This isa good starting point for decidingw h i ch mic to use in a given situation.And it’s certainly helpful when usingthe eve r-popular older micro p h o n e s .

Ta ke note, h oweve r, that some newm i c rophones have add ressed many ofthe historic design weaknesses clev-e rly, c reating mics that are oftena p p ropriate in a broad range ofre c o rding situations.

So with the caveat that these gen-e ralities don’t apply to all mics, c o n-sider the fo l l ow i n g .


DurabilityM oving coil microphones are often

c o n s i d e red to be the heartiest of theb u n ch . As a result they are often thet ransducer of choice for live sounda p p l i c a t i o n s , w h i ch are ve ry toughon delicate equipment.

At the other end of the durab i l i t ychain is the ribbon mic.The ribb o nitself is pretty fra g i l e — e s p e c i a l ly onthe vintage (i.e. ex p e n s ive) ribb o nm i c rophones ava i l able at some studios.

R e m e m b e r, the job of the ribbon isto react instantly to any change inthe air pre s s u re around it. And ift h e re is, s ay, a 10 kHz component tothe music you are re c o rd i n g , t h e nthe ribbon has to be able to moveb a ck and forth ten thousand times as e c o n d . P hysics asks it, t h e re fo re , t oh ave as little weight as possible.

U n fo r t u n a t e ly, as the ribbon losesmass it necessarily loses stre n g t h .Some ribbon microphones are stillm a nu fa c t u red today, and the ribb o nwithin those mics is certainlytougher than the ribbons in Gra n ny ’sm i c ro p h o n e s .

But nobody dares stick a ribb o nm i c rophone in the high amplitudewo rld of a ro ck and roll kick drum.Some new ribbons are designed to betough enough for screaming vo c a l sand thundrous electric guitar. B u tt h ey all want a chance at the horns(not too close, thank yo u ) , the piano,and the acoustic guitar, among others .

In the durability category, c o n-d e n s e rs genera l ly fall somew h e re inb e t ween moving coil and ribb o nd e s i g n s . As a re s u l t , you’ll certainlyfind some of them performing onstage or placed near the ve ry loudinstruments such as kick drum, t r u m-p e t , t ro m b o n e , and so fo r t h .

Sound qualityThough microphones of all types

seem to be improving in capab i l i t y, i tis worth making some genera l i z a t i o n sabout how a microphone of a give nt ransducer tech n o l ogy might sound.

M oving coil dynamic micro p h o n e sa re the largest mechanism used fo rc o nverting acoustic waves into elec-trical ones. Not surp r i s i n g ly, t h e n ,t h ey genera l ly have a natural highf re q u e n cy roll-off as the ability ofthe device to transduce diminishesat higher fre q u e n c i e s .

Consider the fo l l owing hy p o t h e t i-cal ex p e d i t i o n . B e fo re the sessionb e g i n s , you go food shopping for theband you are wo rking with, and thedrummer helps. The shopping listconsists solely of potato chips andb e e r, but in enough quantity to getthe band through a two - week session.

You and the drummer go to theneighborhood Chomp ‘n Gulp, get twoshopping carts, and you fill one cartwith chips while the drummer fillsthe other with beer. ‘ This has nothingto do with micro p h o n e s ,’ you—and myeditor—think to yo u rs e l ve s .

But consider these questions:Wh i ch cart is easier to drive? Wh i chcart is easier to stop and start? Th echip cart and the beer cart can gop retty mu ch the same speed, but thebeer cart needs a stronger shove toget going. C l ever drummers startemptying (i.e. drinking) some cansfor this ve ry re a s o n .

For our micro p h o n e s , the mov i n gcoil is more like the beer cart. Q u i t es i m p ly, the diaphragm/coil assemblyis too big to react quick ly, a sre q u i red by ve ry high fre q u e n c i e s .

The ribbon microphone is morel i ke the chip cart. Consisting of a sin-gle moving part (the ribbon) it is alighter mech a n i s m . As a result ther i bbon transducer is typically moreagile than a moving coil, a ch i ev i n gm o re sensitivity at the high fre q u e n-cies as a re s u l t .

The condenser microphone is gen-e ra l ly lightest of all, b e h aving morel i ke an empty shopping cart in thea n a l ogy ab ove . The only moving part,the diaphra g m , can be an ex t re m e lythin plastic membrane with just thelightest coating of a metal to make it

U n u s ual in our world of complicated gear

(ever open up a digital 8-tra c k ? ) ,

the microphone is an elega n t ly simple,

c o m p l e t e ly knowable technology.


conduct electricity. As a result thecondenser offers the best opportuni-ty to capture the detail of a tra n-s i e n t , or the ve ry high fre q u e n cy por-tion of a high hat.

The apparent transient responseweakness in the moving coil designcan in fact be quite a handy engi-neering tool. By reacting slowly to asudden increase in amplitude, it actsmechanically as a compressor mightact electrically. It reduces the ampli-tude of the peaks of a transientsound.

This is helpful for two major re a-s o n s . Fi rs t , this reduction of peakscan help prevent the sort of distor-tion that comes from ove rl o a d i n gyour electro n i c s . The true spike ofamplitude that comes off a congamight easily distort the micro p h o n ep reamplifier or ove rload the tape yo ua re re c o rding onto. The use of adynamic mic might be just the rightsolution to capture the sound withoutd i s t o r t i o n .

B eyond this issue of audio fidelityand the prevention of distortion,dynamic microphones with their nat-u ral lethargy are often used for cre-a t ive re a s o n s . The sound of a clave ,s n a re , k i ck , d u m b e k , and many otherinstruments is often mu ch more com-

pelling after the subtle reshaping ofthe transient that a moving coilm i c rophone intro d u c e s .

As soon as a session permits, t ryusing a moving coil and a condenser

on the same instrument, placed asnear to the same location as possi-b l e . Then listen critically to the dif-f e rent coloration of each mic. M o s ta p p a rent will be the fre q u e n cyresponse differe n c e s , with the con-denser sounding a little brighter atthe high end while the moving coilo f f e rs perhaps a presence peak inthe upper mid-ra n g e .

But listen beyond this, to the ch a r-acter of the attack of the instrument.Depending on the application, yo uwill often find that the moving coildynamic microphone squashes thet ransients into a more ex c i t i n g , m o reintense sound.

By the time the tra ck you arere c o rding gets combined with all theother tra cks in the mu l t i t ra ck pro j e c t ,and after the signal makes its way

f rom mic to tape, t h rough the consoleand various effects pro c e s s o rs to thel o u d s p e a ke rs , the re c o rded soundf rom a moving coil microphone oftenjust seems to wo rk better.

Meantime the ribbon micro p h o n e ,offering more high fre q u e n cy content

The apparent transient response weakness

in the moving coil design can in fact be

quite a handy engineering tool.


than the typical moving coil micro-phone but less high fre q u e n cy re a chthan most condensers , still finds itsplace in the re c o rding studio.

M a ny instruments have a ra t h e rpainful amount of high end. C l o s e -miking them, as we so often must doin the studio, o n ly makes this wo rs e .The natural high fre q u e n cy attenu a-

tion of a ribbon is often just theright touch to make a trumpet, acy m b a l , a tambourine, a triangle,and others become beautiful, a i ry,and spark l i n g , without being shrill,p i e rc i n g , t h i n , or edgy.

In our next Nuts & Bolts episodewe’ll look at other microphone pro p e r-ties like directionality and prox i m i t yeffect so that we can make more senseout of the vast range of options micro-phones offer. We can look fo r wa rd to ac a re e r-long ex p l o ration of the beauty

of applying different microphones ind i f f e rent musical situations.

Let the other engineers wo rk theirway through the microphone closet,

ra n d o m ly trying different micro-phones in different applications. Wecan org a n i ze our experiences basedon what we know about how them i c rophone wo rk s .

Try to ra t i o n a l i ze what you actual-ly hear with how you think it shoulds o u n d , and you’ll bring some order toan otherwise chaotic part of there c o rding gig.

Alex Case (case@re c o r d i n g m a g . c o m )is the director of Fe r m ata wh e re herecords and produces music he lov e s .He hopes you have a similar job.

Many instruments have a rather pa i n f u l

amount of high end. Close-miking them,

as we so often must do in the studio,

o n ly makes this worse.

Excerpted from the October edition of RECORDING magazine.©1999 Music Maker Publications, Inc. Reprinted with permission.



RECORDING JANUARY 2000

The Equalize r. You may well wonder: what sounds in astudio need to be made equal? Equal to what? A mored e s c r i p t ive term for equalizer would be spectral modifi-e r, or fre q u e n cy-specific amplitude adjuster. Then again,sometimes a ve ry simple term will do—even somethingas mundane as tone contro l s ,l i ke “ B a s s ” and “ Treb l e ” o r“ L ow ” and “ H i g h .”

The audio job of an equalizer is to change the fre q u e n-cy content of an audio signal. If the audio signal is dull,l a cking high fre q u e n cy spark l e , the equalizer is the toolused to fix this—provided that there is some high fre-q u e n cy content in the signal in the first place that thee q u a l i zer can bring out. If the sound is painfully bright,h a rs h ly assaulting our ears with too mu ch high fre q u e n cys i z z l e , the equalizer again offers the solution, this timeby reducing the offending portion of the sound’s high fre-q u e n cy content.

You’ll see that common sense and your ears have atleast as mu ch to do with good use of equalize rs as thet h e o ry behind them. Yet it pays to know that theory ;you’ll do better wo rk knowing the theory where commonsense would only get you so fa r.

E n g i n e e rs use equalize rs to adjust the amplitude of asignal within specific and contro l l able fre q u e n cy ra n g e s .The master fader on your console adjusts the amplitudeof the entire audio signal. Think of an equalizer as a fre-q u e n cy-specific fader; it increases or decreases theamplitude of a signal at certain frequencies only.

Looking high, looking lowThe fre q u e n cy response of a device describes its ab i l i-

ty to create output signals that are consistent across thee n t i re audio fre q u e n cy ra n g e . Fi g u re 1 shows a typicalex a m p l e , in this case a device with an output thate m p h a s i zes the low frequencies and de-emphasizes thehigh fre q u e n c i e s . It crosses 0 dB amplitude change (unityg a i n , neither more or less amplitude) at about 1 kHz.

M u s i c a l ly speaking 1 kHz is a rather high pitch , a l m o s tt wo octaves ab ove middle C. But keep in mind that manyinstruments playing lower notes will have some harmoniccontent at this fre q u e n cy, w h i ch we may need to alter.

Imagine routing a couple of sine waves into thisd ev i c e . One is set to a 1 kHz fre q u e n cy, and the otherwe’ll move up and down to various frequencies in com-parison to our 1 kHz re f e re n c e . We’ll use meters toa s s u re that both sine waves are kept at the same ampli-tude (to our ears , higher pitched sine waves sound mu chlouder than equal-amplitude waves at lower fre q u e n-c i e s ) .

When you measure the output of the device that is setup to boost low frequencies and reduce high fre q u e n c i e s(as per Fi g . 1 ) , a 100 Hz tone (a bit more than an octaveb e l ow middle C) will measure louder than a 1 kHz toneinput at equal amplitude. And a ve ry high fre q u e n cy sinewave (say 10 kHz) will measure softer.

If yo u ’ ve ever had to listen to sine waves for ve ry long—as in the experiment ab ove or when aligning analog mag-netic tape re c o rd e rs — yo u ’ ve learned that they can cre a t e

The latest installment in our series for the novice

is all about tone shaping and how to deal with it.

E q ua l i z ation, Pa rt 1BY ALEX CASE

PART 7


a rather unpleasant, d i s t i n c t ly non-musical listening ex p e-r i e n c e , becoming more and more annoying the highertheir frequencies get. Th a t ’s because sine waves have noove r t o n e s , w h i ch makes them useful for testing and cali-b rating purposes but not usually for making mu s i c .

So let’s take another look at the meaning of the fre-q u e n cy response plot in Fi g u re 1. Consider an input thatis not just a simple sine wave , but is instead an entiremix—a killer mix. The mix is a careful blend of instru-ments and effects that fills the audio spectrum ex a c t ly toyour liking, with a gorg e o u s , p resent midra n g e , an airy,detailed high end, and a rich , warm low end.

Sent through the device in Fi g u re 1, that spectral bal-ance is altere d . The mix you found oh-so-perfect becomestoo heavy in the low frequencies and loses detail up high.

The fre q u e n cy response plot quantifies ex a c t ly the sortof changes in fre q u e n cy content you can expect when asignal is run through the dev i c e . Yo u ’ ve pro b ab ly alre a dyabsorbed the idea that a “ f l a t ” f re q u e n cy response isoften desirab l e , at least during audio pro d u c t i o n . We ’dl i ke devices like microphone cables and mixing consolesto treat the amplitude of all signals the same way at allf re q u e n c i e s . We hope these sorts of devices don’t ch a n g ethe fre q u e n cy ch a racter of the mix “behind our back ”unless we choose to make such ch a n g e s .

And when we want to make such changes away from aflat fre q u e n cy response we resort to using the equalize r.If you feel your vocal tra ck or your entire mix needs a lit-tle more low end and a little less high end, you might runit through an equalizer with a fre q u e n cy response likethat in Fi g u re 1.

To understand equalization you need only unders t a n dthis: you are changing the fre q u e n cy content of a signalby running it through a device whose fre q u e n cy re s p o n s eis distinctly non-flat—on purp o s e .

The trick , and we’ll discuss this in more detail later,is to alter the fre q u e n cy response in ways that aret a s t e f u l , mu s i c a l , and appropriate to the sound. I t ’seasy to get it wro n g . Dialing up just the right eq‘ c u rve’ for a given situation will re q u i re ex p e r i e n c e ,good ears , a good monitoring env i ro n m e n t , and goodj u d g m e n t .

How many knobs?If you consider the fre q u e n cy response like that in

Fi g u re 1 to be adjustable from flat to the specific con-tour show n , you discover that configuring a device thata c t u a l ly controls these sorts of changes isn’t obv i o u s . Tosee how this is done we’ll take a tour of the equalize rsyou are like ly to find in a studio (leaving out equalize rsthat exist in softwa re for now ) .

We begin with the most flexible type of all: the p a ra-metric equalizer.No one got a Nobel Prize for naming thist h i n g . It is a parametric eq because it offers you the mostp a ra m e t e rs for changing the spectral shaping. Th a t ’s it.In fact it’s got all of three para m e t e rs for your knobt weaking pleasure .

U n d e rstanding the three para m e t e rs here make su n d e rstanding all types of equalize rs a bre e ze . All othere q u a l i ze rs will have one or two of these three para m e t e rsava i l able for adjusting on the front of the box . When yo ulearn how to use a parametric equalize r, you are learningh ow to use all types of equalize rs .

Perhaps the most obvious parameter needed is theone that selects the frequency you wish to alter. The cen-ter frequency of the spectral region you are altering isdialed up on a knob labeled Frequency. In our search for

bass, we might have decided that our signal needs addi-tional low frequency content in the area around 100 Hz.Or is it closer to 80 Hz? These decisions are made at thefrequency select control.

N a t u ra l ly, we then decide how mu ch to alter the fre-q u e n cy we ’ ve selected. The addition (or subtraction) ofbass happens via adjustment of the second para m e t e r :C u t / B o o s t .It indicates the amount of decrease or incre a s ein amplitude at the center fre q u e n cy you dialed in onp a rameter number one ab ove .

To take the shrill edge off of a horn, select a highf re q u e n cy (around 8 kHz maybe) and cut a smalla m o u n t — m aybe about 3 d B . To add a lot of bass, b o o s t9 to 12 decibels at the low fre q u e n cy that sounds best,s o m ew h e re between 40 and 120 Hz perhaps. As yo ucan see, these two para m e t e rs alone, f re q u e n cy selectand cut/boost, g ive you a terrific amount of spectra lf l ex i b i l i t y.

Dialing up just the right eq ‘curve’ for a givens i t uation will require experience, good ears, a good

monitoring environment, and good judgment.

Bandwidth and QConsider a boost of 8 dB at 100 Hz.

This could be just the trick to makea guitar sound powerful in the lowe rand fatter notes. You can almosttaste the Gra m my Awa rd after decid-ing on this eq move . You can hearthe re s u l t . But befo re you know whatyou re a l ly did to alter the fre q u e n cyre s p o n s e , you need to consider at h i rd para m e t e r. I t ’s a bit more sub-tle than the first two, and many lessex p e n s ive equalize rs (which we ’ l lc over later) do without it.

We know where we boost (at 100Hz in the above example) and howhigh we boost (by adding 8 dB)—butwe don’t just boost the narrow andexclusive frequency of 100 Hz, eventhough that’s the one we dialed up.Instead we affect a range of fre-

Fi g u re 2 demonstrates two possibleresults from the same center fre-q u e n cy and boost settings. C h e ckthem out and you’ll see what wemeant by saying that selecting a cen-ter fre q u e n cy to boost affects not

quencies both below and above that100 Hz frequency. Remember—that100 Hz is called the center frequen-cy. Just how wide is the boostedregion to either side of that centergoing to be?

just that single fre q u e n cy but theneighboring frequencies as we l l .

The degree to which we also boostother frequencies nearby is definedby the third para m e t e r, Q . The Qdescribes the width of the cut orboost re g i o n .

L e t ’s define first the b a n dwidth o fan equalization ch a n g e . B a n dwidth isc l o s e ly related to but not the sameas Q. B a n dwidth is considered to bethe fre q u e n cy region on either sideof the center fre q u e n cy that is with-in three decibels of the center fre-q u e n cy ’s cut or boost.

Starting at the center fre q u e n cyand wo rking our way out both ab oveit and below it in fre q u e n cy, we canfind the points on the curves inFi g u re 2 where the signal is thre edecibels down from the amplitude atthe center fre q u e n cy. The bandw i d t hof a cut or boost at a specific fre q u e n-cy describes the fre q u e n cy ra n g ebounded by these ‘ 3 dB down’ points.

In our example of an 8 dB boost at100 Hz, the bandwidth is based onthe frequencies that are boosted by 5dB (8 – 3 = 5) or more . Fi g u re 2s h ows two such possible boosts. Th ewide boost has ‘ 3 dB down’ points at75 Hz and 125 Hz. The bandw i d t hthen is 50 Hz (the spectral distance

f rom 75 Hz to 125 Hz). The narrowboost is 3 dB down at 95 Hz and 105H z , g iving a smaller bandwidth ofjust 10 Hz.

N ow ex p ressing values in actualHertz is ra re ly ve ry useful in the stu-d i o. We humans don’t process mu s i cthat way. When you are writing ahorn ch a r t , you don’t decide to add aflute part 440 Hertz ab ove the tenor

s a x . Instead you describe it mu s i c a l ly,s aying that the flute should be per-haps one octave ab ove the tenor sax.

For music we think in terms ofmusical ratios or interva l s , the mostfamous of which is the octave . Th eo c t ave re p resents nothing more thana mathematical doubling of fre q u e n-cy, w h a t ever the fre q u e n cy may be—440 Hertz (“tuning” A ab ove midd l e

C) is one octave ab ove 220 Hz, 1 0 0 0Hz is one octave ab ove 500 Hz, e t c .Because this is how we hear, we stickto this way of describing spectra lp roperties on the equalize r.

Using a ra t i o, we compare theb a n dwidth to the center fre q u e n cyand ex p ress them in re l a t ive terms—in octaves rather than Hz. For ex a m-ple a 50 Hz bandwidth around a 100Hz center fre q u e n cy re p resents ab a n dwidth that is half an octavewide; the bandwidth is half the va l u eof the center fre q u e n cy.

With a fixed bandwidth of “ ex a c t-ly half an octave ,” sweeping the cen-ter fre q u e n cy down from 100 Hz to50 Hz would be accompanied by ab a n dwidth that decreases automati-c a l ly from 50 Hz to 25 Hz. This nar-rowing of bandwidth as measured inHertz ensures that the equalizationch a racter you hear doesn’t ch a n g eas you ze ro in on the desired centerf re q u e n cy.

B a n dwidth ex p ressed in octaves ism o re mu s i c a l ly useful to our earsthan bandwidth ex p ressed in Hertz.If the bandwidth during the prev i o u sm ove (from center fre q u e n cy 100 Hzto down to 50 Hz) had remained at ab a n dwidth of “ ex a c t ly 50 Hz” i twould have sounded like a wider,

an ext ra $20to make an eqs w e e pable willbump up the

price of a 32-channel mixerby over $600.


less precise equalization adjustment at lower fre q u e n-c i e s . Th a t ’s because a bandwidth of 50 Hz around a cen-ter fre q u e n cy of 50 Hz is—you guessed it—a full octave .

Th a t ’s the idea of bandw i d t h . And that’s almost theend of the math in this article. But there is one moreidea to take in here befo re we ’ re done. Bring on Q.

And Q makes threeWhile ex p ressing the bandwidth of an equalizer boost

or cut in octaves makes good sense, the tradition is to flipthe ratio over mathematically (the fa n cy term for this isto take the re c i p r o c a l— i m p ress your clients!). We considercenter fre q u e n cy divided by bandwidth instead of band-width divided by center fre q u e n cy. The spectral ‘ w i d t h ’described this way (still in octaves) is the Q para m e t e r.

The wide boost discussed ab ove and shown in Fi g u re 2is 50 Hz wide at a center fre q u e n cy of 100 Hz. The Qt h e re fo re is 2 (center fre q u e n cy of 100 Hz divided by theb a n dwidth of 50 Hz). The narrow boost has a Q of 10 (100Hz divided by the narrow 10 Hz bandw i d t h ) .S t u d i o - s p e a kincludes phrases like “ l ow Q” and “high Q” to describewide (low Q) and narrow (high Q) boosts and cuts.

It then fo l l ows that the higher the Q, the more surg i c a lyour interve n t i o n . If you have a particular note or tone orhum or buzz that you need to pull out, of course you gofor the narrowest bandwidth around the offending centerf re q u e n cy, with the steepest cut your equalizer can pro-v i d e .S u ch a move is called n o t ching or n o t ch filtering.

So for a full complement of equalization parametersyou have Frequency Select, Cut/Boost, and Q as thethree controls needed to achieve any kind of alterationto a frequency response, from broad and subtle enhance-ments to aggressive and surgical notches. Parametricequalizers give you these three controls for every bandof equalization.

Band of equalization? Th a t ’s right. These three contro l soften appear in sets. A 4-band parametric eq has 12 con-t rols on it (3 controls x 4 bands = 12 controls in all)! Ito f f e rs the three para m e t e rs four different times so thatyou can select four different spectral targets and shapee a ch of them with their own amount of boost or cut, a n de a ch with a unique bandw i d t h .

The re s u l t , if your ears can fo l l ow it all, is the ability toeffect a tremendous amount of change on the spectral con-tent of a signal. Fi g u re 3 shows a possible result of 4-band

p a rametric equalization. The terrific amount of sonic shap-ing power that four bands of parametric equalization offerm a kes it a popular piece of gear in any studio.

But other options ex i s t .

Take away the QSome equalize rs fix the bandwidth internally, p rov i d i n g

access only to the Fre q u e n cy Select and Cut/Boost para m e-t e rs . Because of the dow n grade from three para m e t e rs tot wo this type of eq is sometimes called a s e m i - p a ra m e t r i c(or demi-parametric or even quasi-parametric) e q u a l i z e r.

These devices suffer from having an even less imagina-t ive name than parametric equalize rs . I t ’s pro b ab ly bestto call them s w e epable eq to emphasize that you canadjust the fre q u e n cy that you are cutting or boosting.When you see such a term in a pro d u c t ’s specs it’simplied that you cannot adjust the bandw i d t h . B e l i evem e , if the bandwidth we re adjustable the bro ch u re wo u l db rag that the device is fully para m e t r i c !

This configuration in which only two para m e t e rs( Fre q u e n cy and Cut/Boost) are adjustable is common; itis easy for the re c o rdist to use, easier for the manu fa c t u r-er to design than a fully para m e t r i c , and still ve ry usefulin music pro d u c t i o n .

Take away the frequencyD own one more step, sometimes we only have contro l

over the amount of cut or boost and can adjust neitherthe fre q u e n cy nor the Q of the equalization shape.G e n e ra l ly called p r o g ram eq, this is the sort of equalize rfound on most home stereo equipment (those “ Treb l e ”and “ B a s s ” k n o b s , re m e m b e r ? ) .

You also see this type of eq on many consoles, v i n t a g eand new. It appears most often in a 2- or 3-band fo r m :t h ree knobs labeled High, M i d , and Low that are fixed inf re q u e n cy and Q and offer you only the choice of howmu ch cutting or boosting yo u ’ re going to apply.

In the case of consoles, remember that there may bethe same equalizer repeated over and over on eve rychannel of the console. If it costs an ex t ra 20 bucks tom a ke the equalizer swe e p ab l e , that translates into abump in price of more than $600 on a 32-channel mixer.If it costs 50 bucks to make them fully para m e t r i c , a n di t ’s a 64-channel console...we l l , you do the math.

E x p r e ssing va lues inH e rtz is ra r e ly useful inthe studio. When writinga horn chart, you don’tdecide to add a flu t epa rt 440 Hertz above

the tenor sax .


The good news is that even we l ldesigned program equalization cansound ab s o l u t e ly gorg e o u s . And oftenthe preset center frequencies areclose enough to the ideal spectra llocation to get the job done on manyt ra ck s . Sometimes you don’t eve nmiss the fre q u e n cy select para m e t e r.

Take away the knobsA variation on the equalize r

described so far is the g raphic equal-i z e r. L i ke program eq, this device hasfixed Q and center fre q u e n c i e s ,offering the engineer only thecut/boost decision.

On a graphic eq, the various fre-q u e n cy bands are presented not asknobs but as sliders — l i ke fa d e rs on ac o n s o l e . The result of such a hard-wa re design is that the fa d e rs prov i d ea decent visual description of the fre-q u e n cy response modification that isbeing applied—hence the name‘ gra p h i c .’ (To be ex a c t , the actual eqc u rve outline looks more like a seriesof sharp bumps ab ove and below as t raight line than than the smoothc o n t i nuous curve one might expect.)

H a n dy also is the fact that thefa d e rs can be made quite compact. I tis not unusual to have dual 31-bandgraphic equalize rs that fit into oneor two ra ck spaces.

G raphic eq is an ex t re m e ly intuitiveand comfo r t able way to wo rk . B e i n gable to see an outline of what yo uhear will make it easier and quicke rto set up the sound you are lookingfo r. Turning knobs on a 4-band para-metric equalizer is more of ana c q u i red taste than moving sliders .

Th e re are times in the course of ap roject when one must reshape theharmonic content with great careusing a parametric eq. In otherinstances there is no time for suchc a reful tweaking and a graphic eq isthe perfect, efficient solution. P l a nto master both.

Some knobs are switchesE a rly in my audio career while

attending the AES show in New Yo rkC i t y, I admired a rather impre s s iveBritish eq. It was a super high quali-ty equalizer intended for mastering

piece of the circ u i t , it was phy s i c a l lychanging the circ u i t . That made fre-q u e n cy selection surg i c a l ly pre c i s eand ab s o l u t e ly re p e a t ab l e , a must fo rmastering applications.

I was instantly humbled, a n dlearned a lesson. In choosing whichtype of equalizer to use, you have tot rade off sound quality ve rsus priceand processing flexibility ve rsus easeof use.

This company has such high stan-d a rds for sound quality that theytook away a little bit of user flex i b i l i-ty to get a better and more re p e a t-able sound. C o nve rs e ly, if you find ane q u a l i zer that is fully para m e t r i cand swe e p able across four bands ye tcosts $39.99, you would be wise towonder how it is that they made theeq so infinitely adjustable and howmu ch sound quality was sacrificed inthe name of this flex i b i l i t y.

h o u s e s . That didn’t stop me fro mthinking it could be useful for tra ck-ing a vo c a l , ra d i c a l ly reshaping a gui-tar tone, and other silliness.

I accidentally let slip my disap-pointment that despite a 5-figureprice tag, the fre q u e n cy select knobs‘ c l i cke d .’ Fre q u e n cy select wa s n ’tc o n t i nu o u s ly swe e p able fro m , s ay125 Hz to 250 Hz; the knob clicke df rom 125 Hz to 250 Hz. If you wa n t e d

your equalization contour to be cen-t e red on ex a c t ly a fre q u e n cyb e t ween clicks you we re out of luck .

H ow could this be? I was politelyi n formed that for this particulard ev i c e , selecting a different fre q u e n-cy by clicking a knob on the fa c e p l a t eselected different electronic compo-nents inside the dev i c e .The equalize rwas phy s i c a l ly using different partsfor different fre q u e n cy selections! Itwa s n ’t just adjusting some va r i ab l e

In choosing which type of eq touse, you have to trade off soundq uality vs. price and process i n g

flexibility vs. ease of use.


Two other alternatives ex i s t . Th eshelving e q u a l i zer offers thepeak/dip response on one side of theselected center fre q u e n cy and a flatcut or boost region on the other.Fi g u re 4 demonstra t e s .

A broad equalization desire mightbe to brighten up the sound in gen-e ra l . A high fre q u e n cy shelving eqbumped up 6 dB at 8 kHz will ra i s ethe output at 8 kHz and ab ove . I ti s n ’t limited to a center fre q u e n cyand its associated bandw i d t h . Th eresulting alteration in the fre q u e n cyresponse is flat (like a shelf) beyo n dthe selected fre q u e n cy.

As Fi g u re 4 show s , the concept of ashelving eq applies to low fre q u e n-cies as well as high, and cuts as we l las boosts. In all cases there is a flatregion beyond (ab ove or below) theselected center fre q u e n cy that isboosted or attenu a t e d . A helpfulimage comes by way of beer: theshelving eq shape provides a goodflat region to set a beer on withoutrisk of spilling. Th e re is now h e rewithin the eq move to set a beerwhen using a peak/dip eq contour.

An important final option exists fo rreshaping the fre q u e n cy response ofa signal: the f i l t e r. E n g i n e e rs speakg e n e ra l ly about filtering a signalw h e n ever they change its fre q u e n cy

D o n ’t value an equalizer based onthe number of controls it has. A sim-ple program eq that allows you onlyto adjust the amount of cut or boostmight contain ex t re m e ly high qualitycomponents inside.

Knobs, switches, filtersSo far the most complicated

e q u a l i zer we can build, the one withthe most fa n cy knobs on the fa c e-p l a t e , is a parametric equalize r. I fwe allow four bands of eq we are upto 12 knobs. N a t u ra l ly, it would lookcooler if we added some sw i t ch e s .H e re ’s how.

We ’ ve talked about equalizationchanges that offer a region ofemphasis when we boost or de-emphasis when we cut. This shape iscalled a peak/dip because of the visu-al change it makes in the fre q u e n cyre s p o n s e . R o u g h ly shaped like a bellc u rve , it offers a bump up or down inthe fre q u e n cy re s p o n s e .

response in any way. Under this loosed e f i n i t i o n , all of the equalize rs we ’ vediscussed so far are made up of audiof i l t e rs . But to be more pre c i s e , astand-alone filter must have one ofthe two shapes shown in Fi g u re 5.

A highpass filter (Fi g u re 5A)a l l ows high frequencies through buta t t e nuates low s . A lowpass filter( Fi g u re 5B) does the opposite,a l l owing low frequencies to passt h rough the device without achange in amplitude, but attenu a t-ing high fre q u e n c i e s .

Because the sonic result can berather similar to shelving equalize rscutting out ex t reme high or low fre-q u e n c i e s , t h e re is some confusionb e t ween them. Fi l t e rs distinguisht h e m s e l ves from shelving equalize rsin two key way s .

N ow the faceplate of our equalize ris pretty complicated. The 4-bandp a rametric (12 knobs) gets a low passand high pass filter at each end, a swell as sw i t ches that toggle each bandb e t ween a peak/dip or shelf shape.

But such an equalizer contains ar i ch amount of capability with whichyou can fre e ly alter the spectral con-tent of any signal in your studio.These knobs and sw i t ches enable yo uto bend and shape the fre q u e n cy

response of the equalizer into almosta ny contour imaginab l e .Your stro n gc re a t ive drive to push the limits of asound must be balanced by yo u rmusical and technical knowledge ofyour sound and equipment. L i s t e nc l o s e ly, and have fun.

Alex Case encourages you to insertthe words “cup of water” wh e rever theword “beer” appears abov e .R e q u e s tNuts & Bolts topics via case@re c o r d-i n g m a g . c o m .Th a n k s .

Fi rs t , f i l t e rs are cut-only dev i c e s ;t h ey never boost at any fre q u e n cy( except in the case of resonant filterson synthesize rs , w h i ch we wo n ’t gointo now ) . Shelf eq can cut or boost.

S e c o n d , and this is important, f i l-t e rs offer an eve r- i n c reasing amountof attenuation beyond the selectedf re q u e n cy. Th ey do not flatten outl i ke the shelf; there is now h e re to setthe beer. Th ey just keep cutting, a n dc u t t i n g , all the way down to silence.

If there is some unwanted low fre-q u e n cy air conditioner rumble on at ra ck that you neve r, ever want toh e a r, a filter can essentially re m oveit entire ly. A shelf equalizer willh ave a limit to the amount of attenu-ation it can ach i eve , perhaps only 12or 16 dB dow n . The weakness ofusing a shelving equalizer in thiscase is easily revealed on eve ry quietpassage whenever that tra ck is beingp l aye d , as you’ll still heae the airconditioner rumbling on fa i n t ly inthe back gro u n d .

If you find a 4-band pa rametric eqt h at costs $39.99, you’d be wiseto wonder what was sacrificed inthe name of all that flexibility.

Excerpted from the January edition of RECORDING magazine. ©2000 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119


RECORDING FEBRUARY 2000

Last month we discussed the operation and theory ofe q . N ow let’s roll up our sleeves and look at the nuts andbolts of using it.

The techniqueThe number one appro a ch to dialing in an eq setting is

quite intuitive: boost, s e a rch , and then set the equalize r.Boost by a clearly audible amount, m aybe 12 dB or more .S e a rch by sweeping for the fre q u e n cy select knob untilyou find the sound you are looking fo r. And finally setthe eq to the desired sound—either cutting the fre q u e n-cy if you don’t like it or finding just the right amount ofboost (and bandwidth) if you do.

I t ’s that simple. O ver time, t h rough experience ande a r t ra i n i n g , you can skip the boost and search steps andinstead re a ch immediately for the fre q u e n cy range you wishto manipulate. But until then there ’s nothing wrong withthis appro a ch . And even the fa m o u s , ex p e n s ive engineersresort to the boost, s e a rch and set appro a ch on occasion.

So when do we boost, s e a rch , and set? What are we lis-tening for? Why and when do we equalize? Eq is simplein concept but not necessarily in application.

To EQ, or Not to EQ, That Is the Question

E Q UA L I Z ATION, PA RT 2BY ALEX CASE

PART 8

“I hear they used a Spasmatron 2000 equalizer on that kazoo track.”“No waaaay.”

“Yup, and that album went triple platinum.”“No waaaay.”

“And I saw on a web site that they raised it 4 dB in the lower highs.”“Wait a second. Get me a pencil...What was that killer EQ move again?”


But befo re giving in to despair,re a l i ze that all engineers have a lotto learn about eq. Ap p re n t i c e s , h o b-by i s t s , ve t e ra n s , and Gra m my win-n e rs . . . all are still exploring the sonicvariety and musical capability ofe q u a l i z a t i o n . Eq offers a huge ra n g eof possibilities and options. C r i t i c a llistening skills are developed over alifetime and re q u i re careful concen-t ra t i o n , good equipment, and a goodmonitoring env i ro n m e n t . No onelearned the difference between 1kHz and 1.2 kHz ove r n i g h t .

Interfering with this ch a l l e n g i n glearning process is the temptation toimitate others or repeat equalizationm oves that wo rked for us on the lasts o n g .“ M a g i c ” settings that makeeve ry mix sound great simply don’tex i s t . If you got the chance to writed own the equalizer settings used on,s ay, Jimi Hendrix’s guitar tra ck on‘ The Wind Cries Mary,’ it might betempting to apply it to some otherguitar tra ck , thinking that the equal-i zer goes a long way towa rd improv-ing the tone.

But the fact is, the tone of Jimi’sguitar is a result of countless fa c t o rs :the play i n g , the tuning, the type ofs t r i n g s , the kind of guitar, the amp,the amp settings, the placement ofthe amp within the ro o m , the ro o m ,the microphones used, the micro-phone placement ch o s e n , et cetera etc e t e ra . The equalizer alone doesn’tc reate the tone. In fa c t , it plays a re l a-t ive ly minor role in the deve l o p m e n tof the tone in the scheme of things.

Beware the urge

to imitate others

or repeat eq

moves that

worked on the

last song...


The way to get ahead of this infi-n i t e ly va r i ab l e , difficult to hearthing called eq is to develop ap rocess that helps you stra t e g i ze onwhen and how to equalize a sound.Armed with this org a n i zed appro a ch ,you can pursue a more completeu n d e rstanding of eq.

The audio needs and desires thatm o t ivate an engineer to re a ch fo rsome equalization fall into four cate-gories: The Fi x , The Fe a t u re , The Fi t ,and The Special Effect.

The FixA big motivation for engaging an

e q u a l i zer is to clean things up andget rid of problems that lie withinspecific fre q u e n cy ra n g e s . For ex a m-p l e , o u t b o a rd equalize rs , c o n s o l e s ,m i c rophone pre a m p l i f i e rs , and eve nm i c rophones themselves often havel ow fre q u e n cy roll-off filters . Why isthis kind of eq on all these dev i c e sand what is it used fo r ?

These devices re m ove low fre q u e n-cy energy less for cre a t ive “ t h i s ’ l lsound awe s o m e ” reasons and moreto fix the common problems of rum-b l e , h u m , b u z z , p o p s , and ex c e s s ivep roximity effect.

In many re c o rding situations, wefind the microphone picks up a ve ryl ow fre q u e n cy (40 Hz and below )r u m b l e . This low-end energy comesf rom such culprits as the building’st e m p e ra t u re control system or thev i b ration of the traffic on nearbyh i g h ways and train tra cks (note toself: don’t build studio next door toA m t rak and Interstate 10).

This is re a l ly low stuff that singersand most musical instruments arei n c a p able of cre a t i n g . Since ve ry lit-tle music happens at such low fre-q u e n c i e s , it is often appropriate toinsert a highpass (i.e. l ow cut) filterthat re m oves all the super low low se n t i re ly.

...the fact is,

“Magic” sett i n g s

t h at make every

mix sound great

s i m p ly don’t

e x i s t .


f re q u e n cy just ab ove 60 Hertz orperhaps an octave ab ove , 120 Hertz.

This is high enough in fre q u e n cythat it can audibly affect the mu s i c a lquality of the sound. E x e rcise careand listen care f u l ly when filteringout hum. M a ny instruments (e.g.some vo c a l s , most saxophones, a lotof perc u s s i o n , to name a few) are n ’tchanged mu ch sonically by such a fil-t e r. But low fre q u e n cy-based instru-ments (e.g. k i ck drum, bass guitar)a re n ’t gonna tolerate this kind ofe q u a l i z a t i o n .

Fo r t u n a t e ly, the hum might be lessn o t i c e able on these instruments any-way as their music can mask a lowl evel hum. Buzz is more ch a l l e n g i n g .The additional harmonics of buzzm a ke re m oving it only more mu s i c a l-ly destructive . D r ive care f u l ly.

Other low fre q u e n cy pro b l e m sfixed by a highpass filter are thewo o f e r- s t raining pops of a breath ofair hitting the mic whenever thesinger hits a “ P ” or a “ B ” in a wo rd .Or if you are wo rking outside (doingl ive sound or collecting natura lsounds in the field), yo u ’ ve no doubtd i s c ove red that any bre e ze acro s sthe mic leads to low-end garbage. I fyou can’t keep the wind off them i c ro p h o n e , then filter the low fre-quencies out.

When the instrument you arere c o rding is ve ry close to a dire c t i o n-al micro p h o n e , p roximity effecta p p e a rs . Sometimes this bassy effectthat increases with proximity to ad i rectional mic is good. Radio DJsl ove it—makes them sound larg e rthan life. Sometimes proximity effectis bad. Po o rly miked acoustic guitarsh ave a pulsing low fre q u e n cy soundthat masks the rest of the tone of theinstrument with each strum of theg u i t a r. Roll off the low end to lose it.

E q u a l i ze rs are employed to fixother sounds. E ver had a snare withan annoying ring? Find the fre q u e n-cy range (boost, s e a rch....) mostresponsible for the ring and trya t t e nuating it at a narrow band-w i d t h . O f t e n , turning down that ringreveals an exciting snare soundu n d e r n e a t h .

E ver tra ck a singer with a cold?I t ’s difficult to get a great soundingp e r formance out of a congestedc ro o n e r, but such a problem might bef i x ab l e . Find the dominant mu ddy i n gf re q u e n cy (pro b ab ly somew h e reb e t ween 200 and 500 Hertz) and cutit a bit. Compensate with some help-ful midrange boost and you mightfind a vocal sound that you and thesinger didn’t think was there .

Th a t ’s rumble. A slightly differe n tp roblem is hum. Hum is the interfer-ence from our power lines and powe rsupplies that is based on 60 HertzAC power (50 Hertz for many of ourfriends in other countries).

The alternating current in thep ower provided by the utility compa-

ny often leaks into our audio thro u g hd a m a g e d ,p o o rly designed, or fa i l i n gp ower supplies. It can also beinduced into our audio through prox-imity to electromagnetic radiation ofother power lines, t ra n s fo r m e rs , e l e c-tric motors , light dimmers , and such .

As more harmonics appear—120H z , 180 Hz, and 240 Hz—the humblossoms into a full-grown buzz. B u z zfinds its way into almost eve ry oldguitar amp, helped out a fair amountby florescent lighting and single coilguitar pick u p s .

A g a i n , a low pass filter helps. Tore m ove hum, we need to roll-off at a

E ver tra ck a guitar with oldstrings? Dull and lifeless. This isu n l i ke ly to be fixable (because eqc a n ’t generate missing fre q u e n c i e s ) ,but don’t rule it out until yo u ’ vetried a bit of a boost somew h e re upb e t ween 6 kHz and 12 kHz.

Sometimes a gorgeous spectral ele-ment of a sound is hidden by anoth-

e r, mu ch less appealing fre q u e n cyc o m p o n e n t . A good example of thiscan be found in drums.

Does it never sound right whenyou go searching for the right fre-q u e n cy to boost for that punchy bigbudget drum sound? The low fre-q u e n cy stuff that makes a drumsound punchy often lives just a fewHertz lower than some rather mu ddyj u n k . And boosting the lows inva r i-ab ly boosts some of the mu d .

S e a rch at narrow bandw i d t h

instead for the ugliest, mu dd i e s tcomponent of the drum sound( b e t ween about 180 and maybe 400Hertz) and cut it. As you cut thisp roblematic fre q u e n cy, listen to thel ow end. Often this appro a ch reve a l splenty of low end punchiness thatjust wa s n ’t audible befo re the we l l -placed cut was applied.

The FeatureA natural application of equaliza-

tion is to enhance a particular partof a sound, to bring out componentsof the sound you like . H e re are a fewideas and starting points.

The voice: It might be fair to thinkof voice as sustained vowels and tra n-sient consonants. The vowels happenat lower mid frequencies (200 to 1000Hz) and the consonants happen atthe upper mids (2 kHz on up).

Want a richer tone to the vo i c e ?Manipulate the vowel ra n g e . H av i n gt rouble understanding the wo rd s ?Manipulate the consonant ra n g e .Wa t ch out for ove rly sizzling “ S ”s o u n d s , but don’t be afraid to empha-s i ze some of the human ex p re s s ive-ness of the singer taking a big bre a t hright befo re a screaming ch o r u s .

The snare: I t ’s a burst of noise. Th i sone is tough to eq, as it reacts toalmost any spectral ch a n g e . O n ea p p ro a ch is to divide the sound intot wo parts. One is the low fre q u e n cye n e rgy coming from the drum itself.Second is the mid-to-high fre q u e n cye n e rgy up to 10 kHz and beyond dueto the rattling snares underneath.N a r row the possibilities; look fo rp ower in the drum-based low s , a n dexciting raucous emotion in thenoisy snare s .

Realize that all engineers have a lot to learn about eq.

Apprentices, hobbyists, veterans, and Grammy winners...

all are still exploring the musical ca pability of equa l i z at i o n .


The kick drum: L i ke the snare , consider reducing thisinstrument to two components. Th e re is the click of thebeater hitting the drum fo l l owed by the low fre q u e n cypulse of the ringing drum. The attack lives up in the3 kHz range and beyo n d . The tone is down around 50Hertz and below. These are two good targets for tailoringa kick sound.

The acoustic guitar: Try separating it into its mu s i c a ltone and its mechanical sounds. Listen care f u l ly to thetone as you seek frequencies to highlight. Fr u s t ra t i n g ly,this cove rs quite a range from lows (100 Hertz) to highs(10 kHz).

In para l l e l , consider the guitar’s more peculiar noisesthat may need emphasis or suppression: finger squeaks,f ret buzz, p i ck noise, and the perc u s s ive sound of the boxof the instrument itself, w h i ch resonates with eve rya g gre s s ive strum. Look for these fre q u e n cy landmarks ineve ry acoustic guitar you re c o rd and mix. Eq is a powe r-ful way to gain control of the various elements of thischallenging instrument.

For the instruments you play and often re c o rd , you oweit to yo u rself to spend some time examining their soundswith an equalize r. Look for defining ch a racteristics of theinstrument and their fre q u e n cy ra n g e . Also look for theless desirable noises some instruments make and filethose away on a ‘ wa t ch-out’ list.

These mental summaries of the spectral qualities ofsome key instruments will save you time in the heat of asession when you want more punch in the snare (aiml ow) and more breathiness in the vocal (aim high).

The FitA key reason to equalize tra cks in mu l t i t ra ck pro d u c-

tion is to help us fit all these different tra cks tog e t h e r.One of the simplest ways to bring clarity to a componentof a crowded mix is to get eve rything else out of theway — s p e c t ra l ly.

That is, if you want to hear the acoustic guitar whilethe string pad is sustaining, find a satisfyingly pre s e n tm i d range boost for the guitar and perform a complemen-t a ry cut in the mids of the pad. This eq cut on the stringpad keeps the sound from competing with or drow n i n gout the acoustic guitar.

The trick is to find a spectral range that highlights thegood qualities of the guitar without doing significantdamage to the tone of the synth patch . It’ll take sometrial and error to get it just right, but you’ll find thisa p p ro a ch allows you to layer in seve ral details into a mix.

Expect to apply this thinking in a few critical areas ofthe mix. A round the bass guitar, we encounter low fre-q u e n cy competition that needs add re s s i n g . If you playguitar or piano and do solo gigs as well as band sessions,yo u ’ ve perhaps discove red this alre a dy.

S o l o, yo u ’ ve got low fre q u e n cy responsibilities as yo uc over the bass line and pin down the harmony. In theband setting, on the other hand, you are free to purs u eother ch o rd vo i c i n g s . You don’t want to compete with thebass player mu s i c a l ly, and the same is true spectra l ly.

As an engineer, this means that you might be able topull out a fair amount of low end from an acoustic guitars o u n d . A l o n e , it might sound too thin, but with the bassguitar playing all is we l l . Th e re is spectral room for thel ow frequencies of the bass because the acoustic guitarno longer competes here . But the acoustic guitar still hasthe illusion of being a full and rich sound because thebass guitar is playing along, p roviding uncluttere d , f u l lbass for the song—and for the mix.

In the highs, competition appears among the obv i o u shigh fre q u e n cy culprits like the cymbals and hand per-cussion as well as the not-so-obvious: distorted sounds. I tis always tempting in ro ck music to add distortion to gui-t a rs , vo c a l s , and anything that move s .

Spectrally speaking, this kind of distortion occursthrough the addition of some upper harmonic energy.And this distortion will overlap with the cymbals andany other distorted tracks. Make them fit with the samecomplementary eq moves. Maybe the cymbals get thehighs above 10 kHz, the lead guitar has emphasized dis-tortion around 8 kHz, and the rhythm guitar hangs outat 6 kHz. Mirror image cuts on the other tracks will helpensure all these high frequency instruments are clearlyaudible in the mix.

The mid frequencies are definitely the most difficultregion to equalize . It is ve ry competitive space spectra l ly,as almost all instruments have something to say in them i d s . And it is the most difficult place to hear accura t e-ly. We tend to gravitate towa rd the more obvious low andhigh frequencies areas when we re a ch for the equalize r.

On the road to earning golden ears , plan to focus onthe middle frequencies as a key challenge and learn tohear the subtle differences that live between 500 and6,000 Hz.

There are often technica l

c o n s i d e rations behind eq

decisions, it’s true.

But music wouldn’t be music

if we didn’t selectively

abandon those approaches.

The needs and desires that

m o t i vate an engineer to

reach for equa l i z ation fa l l

into four categories: The

Fix, The Fe ature, The Fit,

and The Special Effect.


The Special EffectIf you have the sense from the dis-

cussion above that there are techni-cal considerations behind equaliza-tion decisions, that’s true. But musicwouldn’t be music if we didn’t selec-tively abandon those approaches. Afinal reason to eq is to create specialeffects. This is where we are leastanalytical and most creative. Yourimagination is the limit, but hereare some starting points.

“ Wa h - wa h ” is nothing more thanva r i able eq. If yo u ’ ve a para m e t r i ce q u a l i zer handy, p a t ch it in to a gui-tar tra ck alre a dy re c o rd e d . Dial in ap retty sharp midrange boost (high-Q,1 kHz, + 1 2 d B ) . As the tra ck play s ,sweep the fre q u e n cy knob for funand pro f i t .

On automated equalize rs you canp rogram this sort of eq cra z i n e s s .Without automation, you just printthe wa h - wah ve rsion to a spare tra ck .Your cre a t ive challenge: ex p l o re notjust middle fre q u e n c i e s , but low andhigh fre q u e n cy ve rsions; try cuts aswell as boosts; and apply it to anyt ra ck (acoustic guitar, p i a n o, t a m-b o u r i n e , a ny t h i n g ) .

Another special effect is actuallyused to improve re a l i s m . As soundwaves travel through space, t h ef i rst thing to go are the high fre-q u e n c i e s . The farther a sound hast rave l e d , the less high fre q u e n cycontent it has.

Consider the addition of a re p e a t-ing echo on a vocal line. For ex a m-p l e , the lead singer sings, “My baby ’sgonna get some Gouda Cheese.” A n dthe back ground singers sing,“ G o u d a ! ” N a t u ra l ly the mix engineerfeeds the back ground line into a dig-ital delay that repeats at the rate ofa quarter note triplet: “ G o u d a . . .G o u d a . . . G o u d a .”

For maximum effect, it is tra d i t i o n-al to equalize the signal as it is fedb a ck to the delay for each re p e t i t i o n .The first “ G O U DA ! ” is simply ad e l ay. It then goes through a low p a s sfilter for some re m oval of high fre-q u e n cy energy and is fed backt h rough the delay.

It is delayed again: “ G o u d a ! ” O n c em o re through the same lowpass filterfor still more high fre q u e n cy attenu-ation and back through the samed e l ay: “ g o u d a .” The result is (with atriplet feel): “ G O U DA ! . . . G o u d a !. . . g o u d a .” The echoes seem to growm o re distant, c reating a more engag-ing effect.


O bv i o u s ly, this eq appro a ch appliesto signals other than ech o e s , and iteven wo rks on non-dairy pro d u c t s . I ncomposing the stereo or surro u n dimage of your mix, you not only panthings into their horizontal position,but you push them back , away fro mthe listener by adding a touch morereverb (obvious) and re m oving a bitof high end (not so obv i o u s ) . This eqm ove is the sort of subtle detail thathelps make the stere o / s u r ro u n dimage that mu ch more compelling.

Speaking of stere o, a boring oldmonophonic tra ck can be made morei n t e resting and more stere o - l i ket h rough the use of equalization. Wh a tis a stereo signal after all? It is diffi-cult to answer such an intere s t i n g

question without writing a book, o rleast an entire article dedicated tothe topic. But the one sentencea n swer is: a stereo sound is the re s u l tof sending different but related sig-nals to each loudspeake r.

Placing two microphones on apiano and sending one mic left andthe other right is a clear example ofs t e re o. The sounds coming out of thel o u d s p e a ke rs are similar in that theya re each re c o rdings of the same per-formance on the same piano happen-ing at the same time.

But there are subtle (and sometimesradical) differences between thesounds at each mic due to their partic-ular location, o r i e n t a t i o n , and type ofm i c ro p h o n e .The result is an audioimage of a piano that is more intere s t-i n g , and hopefully more mu s i c a l ,t h a nthe monophonic single micro p h o n ea p p ro a ch would have been.

If you begin with a single micre c o rding of a piano and wish to cre-ate a wider, m o re re a l i s t i c , or justplain we i rd piano sound in you mix,one tool at you disposal is equaliza-t i o n . Send the single tra ck to two dif-

f e rent channels on your mixer andeq them differe n t ly. If the signal onthe left is made brighter than thesame signal sent right, then theimage will seem to come from thel e f t , brighter side (remember dis-tance re m oves high fre q u e n c i e s ) .

Consider eq differences betwe e nleft and right that are more elab o-rate and invo l ve seve ral differe n tsets of cuts and boosts so that nei-ther side is ex a c t ly brighter than theo t h e r, just differe n t . Then the imagewill widen without shifting one wayor the other. The piano becomesm o re unusual (re m e m b e r, this sec-tion of the article is called SpecialE f f e c t s , so anything goes....); itsimage is more liquid, less pre c i s e .

A dd some delay s , reve r b s , a n dother processing (topics of futureNuts & Bolts pieces) and a one-micmonophonic image takes on a rich ,s t e reophonic life.

The EndThe challenging and subtle art of

equalization needn’t be surro u n d e din my s t e ry. Wh e n ever you have at ra ck with a problem to be re m ove dor a feature to be emphasize d , t ry tograb it with eq. If a mix is gettingc rowded with too many instrumentsfighting for too little space, c a rve outd i f f e rent spectral regions for thecompeting instruments using eq.

And sometimes we just want tot a ke a sound out and make it morei n t e re s t i n g .A g a i n , just boost, s e a rch ,and set the equalizer so that you likewhat you hear.

Alex Case has cornered the marketon mu r k y, dull sounding mixes.W h at ’syour eq specialty gonna be? SuggestNuts & Bolts topics via case@re c o r d-i n g m a g . c o m .

C r i t i cal listening skills are

developed over a lifetime. No one

learned the difference between

1 kHz and 1.2 kHz overnight.

Excerpted from the Febru a ry edition of RECORDING magazine.©2000 Music Maker Publications, Inc. Reprinted with permission.



RECORDING MARCH 2000

Music signals are ra re ly consistent in leve l . E ve ryc ra ck of the snare , s y l l able of the vo c a l , and strum

of the guitar causes the signal to surge up and re c e d ed own in amplitude.

The top of Fi g u re 1 (on Page 52) shows the amplitudeof about a bar of mu s i c . Signals like this one must fitt h rough our entire audio chain without distortion: them i c ro p h o n e , the microphone pre a m p, the console, t h eo u t b o a rd gear, the mu l t i t ra ck re c o rd e r, the 2-tra ck masterre c o rd e r, the power amp, and the loudspeake rs .

The highest peak must get through these devices with-out clipping, while the detail of the lowe s t , n e a rly silentbits of music must pass through without being swa m p e dby noise. When we aim for 0VU on the meters , all we ’ redoing is trying to avoid distortion at the high end ofthings and noise down on the bottom.

To help us fit ex t re m e ly dynamic signals within theamplitude limits imposed by our studio, we re a ch for ac o m p re s s o r. Its task? Quite simply, when a signal gets tool o u d , the compressor turns it dow n .

What counts as too loud? The Th reshold setting on thec o m p ressor sets the level at which compression is tob e g i n . When the amplitude of the signal is below thist h reshold the device passes the audio thro u g hu n ch a n g e d . When the signal exceeds the threshold thec o m p ressor begins to turn the signal dow n .

Taking controlH ow does it turn it down? This question breaks in two.

H ow mu ch? And how fa s t ?

The amount of compression is determined by the R at i os e t t i n g . M a t h e m a t i c a l ly, the ratio compares the amountof the input signal ab ove threshold to the amount of thea t t e nuated output ab ove thre s h o l d .

For ex a m p l e , a 4:1 (four to one) ratio describes a situa-tion in which the input was four times higher than the out-put ab ove the threshold—4 dB ab ove threshold in becomes1 dB ab ove threshold out, 8 dB ab ove threshold inbecomes 2 dB ab ove threshold out. A ratio of X:1 sets thec o m p ressor so that the input must exceed the threshold byX dB for the output to go just one dB ab ove thre s h o l d .

H ow fast the signal is attenuated is controlled by theA t t a ck s e t t i n g .A t t a ck describes how quick ly the compre s-sor can fully kick in after the threshold has been ex c e e d e d .

Fast attack times will enable the compressor to re a c tve ry quick ly, while slow attack times are more letharg i c .Sometimes compre s s o rs change the gain so quick ly thatit becomes audible—and unmusical (although the effectcan be useful as an effect). It becomes desirable to slowthe attack time down and let the compression sneak intoa c t i o n . I t ’s a tra d e - o f f, t h o u g h , because if the purpose ofc o m p ression is to control the dynamic range of a signalto prevent distortion, then it must act quick ly.

Th re s h o l d , ra t i o, a t t a ck…then what? When the ampli-tude of the music returns to a level below thre s h o l d , t h ec o m p ressor must stop compre s s i n g . The amount of timeit takes the compressor to return to ze ro gain ch a n g eafter the signal falls below threshold is set by adjustingthe compre s s o r ’s R e l e a s e. Setting this control pro p e rlyhelps avoid introducing artifacts to your sound.

BY ALEX CASE

PART 9

What happens when you ignore w h at they were originally designed to do?

C O M P R E SS O R SWhat happens when you ignore

w h at they were originally designed to do?

C O M P R E SS O R S


Welcome to the wo rld of compre s s i o n . Sometimes it’stoo fast; other times it’s too slow. Sometimes we knowwhen it’s just right. Other times we seek to set it so thatwe can’t even hear it wo rk i n g . Tweaking a device until itsounds so good that you can’t even hear it isn’t easy.

This brings us to an important issue with compre s s i o n :it is often hard to hear. We discuss many applications fo rc o m p ression here in this month’s episode of ‘ N u t s & B o l t s .’E a ch application sounds differe n t . And most of them,until yo u ’ ve had some experience and audio ear tra i n i n g ,a re frustrating to hear accura t e ly. C o m p re s s i o n , l i ke somu ch of what we do as engineers , leads to:

- A few mistake s . O ve rc o m p ressing is a common pro b-l e m . Sometimes you can’t tell that it’s ove rc o m p re s s e duntil the next day. The affect of compression is at timesquite subtle and at other times quite obv i o u s . S p e n d i n gall day mixing one song with your ears wide open canm a ke it hard to remain objective .

- Audio hype and attitude. People might rave ab o u th ow great the compression sounds—and you don’t hearwhat on earth they ’ re talking ab o u t .A g a i n , some com-p ression is hard to hear and re q u i res ex p e r i e n c e .Perhaps they ’ ve had the chance to hear this kind of com-p ression befo re . All you need is time between the speak-e rs immersed in compression of all kinds and you’ll pickit up. On the other hand, sometimes people are just fullof bull pucky.

Beyond the controlsThese four para m e t e rs — t h re s h o l d , ra t i o, a t t a ck and

re l e a s e — e n able the compressor to care f u l ly monitor andm a ke fine adjustments to the amplitude of a signal auto-m a t i c a l ly. The engineer is then freed to concentrate moreon other things (Is the guitar in tune? Is the coffees t rong enough?)

If it we re invented today, it would have some hy p e d - u p,one wo rd with two capital letters sort of name likePowe r Fader—and it would have a Web s i t e . The humblec o m p ressor offers a handy way to control pre c i s e ly andmanipulate the dynamics of the signals we re c o rd .

While these four para m e t e rs are always at wo rk , t h eya re not always on the faceplate of the dev i c e . That is,t h ey are not always user- a d j u s t ab l e . Th e re are compre s-s o rs at all price points that leave off some of these con-t rols; it’s part of their sound. Other compre s s o rs offer fullc o n t rol over all the para m e t e rs yet also offer pre s e t s .

The presets reflect someone else’s careful tweaking toget the sound in the right place. Sometimes the pre s e t ss i mulate the attack and release ch a racteristics of other,v i n t a g e , c o l l e c t i b l e , famous sorts of compre s s o rs .

I t ’s a good idea for beginners to spend some time withthe fully adjustable type for ex p l o ration and ear tra i n-i n g . But I don’t hesitate to re a ch for those compre s s o rswith only a few knobs on the box during a session. Th eycan often get the job done more quick ly and with bettersonic re s u l t s .

Easily compressedWhen the singer re a l ly gets confident and excited he

or she sings the choruses re a l ly loud—louder than duringall the other takes in re h e a rs a l . G reat perfo r m a n c e .U nu s able tra ck .

Without some amplitude protection a killer take is lostto distortion. Be re a dy for this with some gentle (aro u n d4:1 or less) compression across the vo c a l . Then yo u raudio path can withstand the adre n a l i n e - i n d u c e di n c rease in amplitude that comes from musicians whent h ey are ‘in the zo n e .’

When the guitarist gets nervous he or she starts mov-ing around on the stool, l e aving you to mike a mov i n gt a rg e t . Compelling perfo r m e r. N e rvous in the studio.Without the constant gain-riding of a compre s s o r, yo ucan hear the guitarist moving on- and off-mic. A g a i n , alittle gentle compression might just coax a usable re c o rd-ing out of an inexperienced studio perfo r m e r.

When the bass player pulls out that wonderful old, c o l-l e c t i b l e , va l u ab l e , sweet sounding, could sure use a littlecleaning up, a re n ’t those the original strings, c o u l d n ’ts t ay in tune for eight bars if you paid it…gorgeous beastof an instrument, you can be sure that—even in the

Without amplitude protection a killer take could be lost.

Be ready with some gentle compression for the voca l .


hands of a master—the A string is consistently a littlequieter than the E string.

Of course the solution is compre s s i o n . Without thec a re f u l , p recision adjustments made to the amplitude ofthe signal, the ve ry foundation of the song (according tothe bass playe r, a ny way) becomes shaky. All too oftenyou need the careful level adjustments of gentle com-p ression (as shown in Fi g u re 1).

Th e re ’s more to it than fixing a problematic tra ck . Wealso patch gentle compression across perfectly finet ra cks to make them, e r, b e t t e r. We l l , louder any way.

A handy side effect of compre s s i n g — reducing the ove r-all dynamic range of the signal—is that now it can beturned up. While this may seem counterintuitive ,t h e re ’sroom to make the tra ck louder as a whole when the pointsof highest amplitude have been lowe red by the compre s-s o r. Fi g u re 2 demonstrates this sort of gentle compre s s i o n .

This is often taken to radical ex t remes where mixesa re ab s o l u t e ly crushed (i.e. re a l ly compre s s e d , see alsos q u a s h e d , s mu s h e d , et al.) by compression so that thea p p a rent loudness of the song exceeds the loudness ofall the other songs on the radio dial. Selling re c o rds isa competitive business. Loudness does seem to helpsell re c o rd s .

And so it goes. Often the music suffers in this commit-ment to loudness and hope for sales. A r t i s t , p ro d u c e r,and engineer must make this trade-off care f u l ly. B u teven in small measure s , a little bit of gentle compre s s i o nbuys you a little bit of loudness if you want it.

Take it to the limitAnother use of the compressor is to attenuate the

s h a rp amplitude spikes within the audio that would ove r-load a device and cause (unwanted) distortion.

During the course of a song, some snare hits are hard e rthan others . The slamming that goes on during the ch o-rus might be substantially louder than the delicate,ghost-note-filled snare wo rk of the bridge.

A limiter will attenuate the ex t reme peaks and preve n tnasty distortion. And a limiter is nothing more than ac o m p ressor taken out to rather ex t reme settings.Th reshold is high so that it only affects the peaks, l e av-ing the rest of the music untouch e d . Ratio is high,greater than 10:1, so that any signal that breaks ab ovet h reshold is seve re ly attenu a t e d .A t t a ck is ve ry fast sothat nothing gets through without limiting.

Called peak limiting,this sort of processing is used top revent distortion and protect equipment. Fi g u re 3 give san ex a m p l e .

Fitting a signal on tape without ove rl o a d i n g , or bro a d-casting a signal without overmodulating (getting tool o u d ,s i m p ly put) re q u i res that the signal never exceed acertain amplitude. L i m i t e rs are inserted to ensure theseamplitude limits are honore d .

In live sound applications, exceeding the amplitudec a p ability of the sound re i n fo rcement system can lead tof e e d b a ck , damage loudspeake rs , and turn happy crow d sinto hostile ones. L i m i t e rs offer the solution again. Th eyg u a rd the equipment and listeners dow n s t ream by stop-ping the signal from getting too loud.

Ulterior motivesWhen the answering machine was inve n t e d , its intended

p u rpose was to answer the phone and take messages whenyou we re away. But the day after the first one was sold, t h ea n swering machine took on a new, m o re important ro l e :call scre e n i n g . The most common message on thesed evices is something like ,“ I t ’s me. P i ck up. P i ck up!”

The use of a device in ways not originally intendedo c c u rs all too often, and the compressor offers a case inp o i n t . While dynamic range reduction and peak limitinga re effective , intended use for the dev i c e , we use themfor other, less obv i o u s , m o re cre a t ive reasons as we l l .

The envelope pleaseThe envelope describes the ‘shape’ of the sound, h ow

gra d u a l ly or ab r u p t ly the sound begins and ends, a n dwhat happens in betwe e n .D r u m s , for ex a m p l e , h ave as h a rp attack and nearly instant decay. That is, the enve-lope resembles a spike or impulse. Synth pads might oozein and out of the mix, a gentle envelope on both the attackand decay side. Piano offers a combination of the two. I t sunique envelope begins with a distinct, s h a rp attack andrings through a gently ch a n g i n g ,s l ow ly decaying sustain.

All instruments offer their own unique enve l o p e .Consider the sonic differences among seve ral instru-ments playing the same pitch: piano, t r u m p e t , vo i c e , g u i-t a r, v i o l i n , and didgeridoo. Th e re are obvious differe n c e s

A handy side effect of compressing—reducing the overa l l

dynamic range of the signal—is that now it can be turned up.


Find a tra ck or sample to pro c e s s .Pa t ch in a compressor and sharp e nthe attack . Be sure your attack isn’ttoo fast or you might re m ove thes h a rpness of the snare entire ly. S e tthe ratio to at least 4:1, and gra d u a l-ly pull the threshold dow n .

This type of compression has theeffect of morphing a spike onto thef ront of the snare sound. M u s i c a ljudgement is re q u i red to make surethe click of the sharper attack fitswith the remaining ring of the snare .Trading off a low threshold with ahigh ratio offers the engineer pre c i s ec o n t rol over the shape of the morea g gre s s ive attack .

And this isn’t just for snare s .A nything goes, but do try similarp rocessing on piano and acoustic gui-t a r. Done we l l , you’ll create a moreexciting sound that finds it place ina crowded mix more easily.

Another unusual effect can be cre-ated using the release of a compre s-s o r. A fast release pulls up the ampli-tude of the sound even as it decay s .

This is also shown in the snareexample of Fi g u re 4. Notice theraised amplitude and incre a s e dlength in the decay portion of thewave fo r m . Dial in a fast enoughrelease time, and the compressor canraise the volume of the sound almostas quick ly as it decay s — i t ’s almost“ u n c o m p re s s i n g ” i t .

Applied to piano, g u i t a r, and cy m-b a l s , this setting develops a nearlyinfinite sustain, making these instru-ments bell or ch i m e - l i ke in ch a ra c t e r,while still retaining the unmistakab l esound of the original instrument. Fi l ethis under ‘Special Effects,’ but don’tfo rget about it. An unnatural effectl i ke this can be just what a pop tuneneeds to get noticed.

Another interesting thing happenswhen you apply some ex t reme com-p ression with a fast release time. I fthe compressor has pulled down thepeaks of the wave form and thenq u i ck ly releases the signal after it

in the spectral content of theseinstruments; they have a differe n tt o n e . But at least as important, e a chof these instruments begins and endsthe note with its own ch a ra c t e r i s t i ce nvelope—its signature .

The compressor is the tool we useto modify the envelope of a sound. Al ow thre s h o l d , medium attack , h i g hratio setting can be used to sharp e nthe attack . The sound begins, at anamplitude ab ove threshold (set low ) .An instant later (medium attack ) ,the compressor leaps into action andyanks the amplitude of the signald own (high ra t i o ) . S u ch compre s s i o na u d i b ly alters the shape of thebeginning of the sound, g iving itm o re a more pronounced attack .

This appro a ch can of course beapplied to most any tra ck . A goodstarting point for this sort of wo rk isa snare drum sound. I t ’s demonstra t-ed in Fi g u re 4, seen on Page 58.

Pop music

pushes us to

h ave bright, airy ,

in your fa c e ,

e xciting voca l

t ra c k s . . .


has fallen below thre s h o l d , you startto hear parts of the sound that we rep rev i o u s ly inaudible.

Fast release compression enab l e syou to turn up the sound and hearm o re of the decay of a snare , t h eex p re s s ive breaths between thewo rds of a vo c a l , the ambience of theroom in between drum hits, the deli-cate detail at the end of a sax note,and so on. Once again, h e re is a useof compression to make certain partsof the signal louder.

The flip side is that you might notwa n t , s ay, the pick noise to becomeove rly accentuated.

That hurtSSSPop music standards push us to have

b r i g h t ,a i ry, in your fa c e , exciting vo c a lt ra ck s . And this convincing vo c a lsound must rise ab ove a wall of distort-ed guitars ,t o r t u red cy m b a l s ,s h i m m e r-ing reve r b , and sizzling synth patch e s .

Needless to say, we push vo c a l swith a high dose of high fre q u e n cyhype (ava i l able on your trusty equal-i ze r ) .A dd some fast release compre s-sion to this bright equalization con-t o u r, and you re a l ly start to hear theb re a t h i n g , ra s p i n g , swe a t i n g , a n dd rooling of the singer; that’s where agood deal of the emotion live s .

We can get away with this aggre s-s ive equalization move eve ry w h e reexcept where the vocal was alre a dybright to begin with: hard consonantsl i ke S and F (and even Z, X , T, D, K ) .These sounds are natura l ly rich inhigh fre q u e n cy content.

Run them through the equalize rthat adds still more high end, a n dyo u ’ ve got the sort of vocal that zingsthe ears with pain on eve ry S.Yo uc a n ’t miss it: eve ryone in the ro o mblinks eve ry time the singer hits an S.

C l ever compression will solve thisp ro b l e m . In our discussion of com-p ression so far we have been apply i n gour settings of thre s h o l d , ra t i o, a t t a ckand release to the signal being com-p re s s e d . But what if we compre s s e done signal while ‘looking at’ another?

S p e c i f i c a l ly, l e t ’s compress thelead vo c a l . But instead of compre s s-ing it based on the vocal tra ck itself,l e t ’s use a different signal to gove r nthe compre s s i o n .

We feed a modified vocal signalinto this alternative input (called as i d e ch a i n) . The vocal itself is whatgets compre s s e d , but the behavior ofthe compre s s o r — w h e n , h ow mu ch ,h ow fast and how long to compre s s —is governed by the sidechain signal.

To get rid of esses, we feed a sig-nal into the sidechain that hasenhanced esses. That is, the sidechain input is the vocal tra ck equal-i zed so as to bring out the esses, a n dd e - e m p h a s i ze the re s t .

We never hear this tra ck — o n ly thec o m p ressor does. But when thesinger sings an S, it goes into thec o m p ressor loud and clear, b re a k i n gt h reshold and sending the compre s-sor into action.

The sidechain signal is the vo c a lwith a high fre q u e n cy boost (may b e12 dB somew h e re around 4 kHz to 8k H z , w h e rever the particularlypainful consonant lives for thatsinger); you can filter out the rest ofthe side chain vo c a l . The compre s s o ris set with a mid to high ra t i o, fa s ta t t a ck , and fast re l e a s e .

The threshold is adjusted so thatthe compressor operates during theloud esses only. In between esses the

...above a wall of

g u i tars, tort u r e d

cymbals, reverb,

and sizzling

synth patc h e s .

c o m p ressor doesn’t touch the vo c a l .This vocal can be made edgy andbright without fear.

More is betterSometimes a strong dose of com-

p ression is applied—to an indiv i d u a lt ra ck or the entire mix—just for theeffect of, we l l ,c o m p re s s i o n . That is,t h e re is something about the soundof ex t reme compression that make sthe music more ex c i t i n g .

The distortion typically dialed inon most electric guitar amps add san unmistakab l e , i n s t i n c t ive ly stim-ulating effect. By modifying theamplitude of the wave fo r m , c o m-p ression is also a kind of distortion.And it seems to communicate ani n t e n s e , on the edge, pushing thelimits sort of feeling.

flying in. Jakob Dylan’svoice is certainly raised,but it’s well short of ascream.

M o s t ly the whole mixjust gets squashed bigt i m e . I almost think, a n a-ly t i c a l ly, that the songgets a little quieter ate a ch ch o r u s , with the 2-

mix compression pushing hard . B u tmu s i c a l ly, the chorus soars .

Th a t ’s the sort of compression thatsells re c o rd s .

Mercedes makes a car with the wordKo m p ressor on it. Alex Case wants one.Request Nuts & Bolts topics viac a s e @ re c o r d i n g m a g . c o m .

A pro fo u n d ly effective example ofthis is Tom Lord - A l g e ’s mix of “ O n eH e a d l i g h t ” by the Wa l l f l owe rs .A te a ch chorus there is a compellingamount of energ y. It feels right.

But if you listen analytically, notemotionally, you hear that there isno big change in the arrangement:the drummer doesn’t just start bang-ing every cymbal in sight, a wall ofextra distorted guitars doesn’t come

Excerpted from the March edition of RECORDING magazine. ©2000 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119


RECORDING APRIL 2000

Wanna know how to mix? Me too. And I’ve beendoing it (trying at least) for about ten ye a rs . Th a t ’s

part of the pleasure of our cra f t . We keep learning, c re a t-i n g , and ex p l o r i n g . Th e re is no single right way to do it.

H aving said that, t h e re is something of a standarda p p ro a ch to mixing pop music that is worth rev i ew i n g . I ti s n ’t the only way. In some cases, it isn’t even the rightway. But it is a fra m ewo rk for study, a starting point fro mw h i ch you can take off in your own dire c t i o n .

Consider a pop/ro ck tune with the fo l l owing somew h a ttypical arrangement: drums, b a s s , r hythm guitar (dou-b l e d ) , lead guitar, c l av i n e t , lead vo c a l , and back gro u n dvo c a l s . Wh e re do we begin?

Fi rst let’s lay out the console. Whether physical orDAW- b a s e d , it helps to preset the signal flow with asmu ch as we think we’ll need.

Mixing fo rces us to be cre a t ive in how we shape andcombine the various tra cks and effects, yet we have tohook eve rything up corre c t ly too. The latter tends toi n t e r f e re with the fo r m e r, so it helps to do a chunk ofthe tedious and technical thinking ahead of time sothat it doesn’t interfere with your flow of inspira t i o nwhile mixing.

Global effects

We don’t yet know yet all the effects we may want fo rthis mix, but some standards do ex i s t . We’ll pro b ab lywant a long reverb (hall-type program with a reverb ove rt wo seconds), a short reverb (plate or small to mediumroom with a reverb time around one second), a ‘ s p re a d e r ’(see the sideb a r ) , and some delays (eighth note, q u a r t e rn o t e , or quarter note triplet in time).

L a u n ch the appropriate plug-ins or patch in the appro-priate hardwa re . These are effects we ’d like to have atour fingertips so that we can instantly send a bit ofvo c a l , s n a re , and lead guitar to the same effect. The wayto have all these effects handy is to use aux sends (seeNuts & Bolts #2, 8 / 9 9 ) .

Kick-starting

With the console laid out, we can start mixing.Wh e re do we start? We l l , the vocal is almost always themost important single piece of eve ry pop song. S omost engineers start with... the d r u m s. Starting withthe vocal makes good sense, because eve ry tra ckshould support it. But easily 99% of all pop mixesstart with the drums.

Why? Because the drums are often the most difficultthing to get under contro l . “ The drum part” is a partwith at least eight separate instruments playing all atonce in close proximity to each other (kick , s n a re , h i - h a t ,t wo or three ra ck toms, a floor tom, a crash cy m b a l , aride cy m b a l , and all the other various add-ons the drum-mer has managed).

I t ’s hard to hear the problems and tweak the sounds ofthe drums without listening to them in isolation. So wetend to start with the drums so that they are out thereall alone. Once the vocals and the rest of the rhythm sec-tion are going, i t ’s hard to dial in just the right amount ofc o m p ression on the ra ck toms.

What do we do with the drums? The kick and snare arethe source of punch , p owe r, and tempo for the entiret u n e . Th ey ’ ve got to sound awe s o m e , so it’s natural tostart with these tra ck s .

BY ALEX CASE

PART 10

As this month’s issue of Recording focuses on mixdown, Alex talks us through a mix

Mixing By The NumbersAs this month’s issue of Recording

focuses on mixdown, Alex talks us through a mix

Mixing By The Numbers

Excerpted from the April edition of RECORDING magazine. ©2000 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119 For Subscription Information, call: 1-800-582-8326

Step one: keep them dead centerin the mix. The kick , s n a re , b a s s , a n dvocal are all so important to the mixthat they almost always take centers t a g e . The kick needs both a clear,crisp attack and a solid low fre q u e n-cy punch . Eq and compression areyour best tools for making the mostof what was re c o rd e d .

The obvious: eq boost at around 3kHz for more attack and eq boost atabout 60 Hz for more punch . Not soo bvious: eq cut with a narrow band-width around 200 Hz to get rid ofsome mu ddiness and reveal the lowf requencies beneath (see Parts 7 and8 of this series, 1 & 2 / 0 0 ) .

C o m p ression does two things fo rthe kick . Fi rs t , it controls the re l a-t ive loudness of the kick s , m a k i n gthe we a ker kicks sound almost as

Getting en-snared

The snare is next. It likely gets asimilar treatment: eq and compres-sion. The buzz of the snares is broad-band, from 2 kHz on up. Pick arange you like: 8 kHz might soundtoo edgy or splashy, but 12 kHzstarts to sound to delicate and hi-fi.You make the call.

A low fre q u e n cy boost for punch i-ness is also cool for snare . Look high-er in fre q u e n cy than you did on thek i ck — m aybe 100 Hz or so. Also lookfor some unpleasant sound to cut.S o m ew h e re between 500 and 1000Hz lives a cluttere d , b oxy sound thatd o e s n ’t help the snare tone and iso n ly going to fight with the vo c a land guitars any way. Try to find a nar-row band to cut and the rest of themix will go more smoothly.

The snare definitely benefits fro mthe addition of a little ambience.Plan to send it to the short reve r band/or hope to find some natura lambience in the other drum tra ck s .The overhead microphones are agood source of ex t ra snare sound.And any re c o rded ambience or ro o mt ra cks should be listened to now.

With the kick and snare punchyand nicely equalize d , i t ’s time toraise the overheads and hear the kit

s t rong as the powerful ones. Th esecond goal of compression is tomanipulate the attack of the kickso that it sounds punchy and cutst h rough the rest of the mix. See lastm o n t h ’s column for a description ofthe sort of low thre s h o l d , m e d i u ma t t a ck , high ratio compression thats h a rpens the amplitude envelope ofthe sound.

Placing the compressor after thee q u a l i zer lets you tweak in somec l ever way s . The notch around 200

Hz keeps the compressor from re a c t-ing to that unwanted mu rk i n e s s .A n das you push up your low fre q u e n cyboost on the eq you can hear thec o m p ressor re a c t . An aggre s s iveboost of lows fo rces the compre s s o rto yank down the signal hard .Yo u ’ l lfind plenty of punchiness using thisa p p ro a ch .

Why not distort

the voca l ?

Or flange the

reverb?...

fall into a single, p owerful whole.The overheads have the best ‘ v i ew ’of the kit and the snare often soundsphenomenal there . Combine themwith the kick and snare tra cks tom a ke the song re a l ly move .

I t ’s tempting to add a gentle highf re q u e n cy boost across the ove r-heads to keep the kit crisp. If thet ra cks are alre a dy bright as re c o rd-e d , d o n ’t feel obligated to add morehigh end. In fact a gentle and widep resence boost between 1 and 5 kHzcan often be the magic dust thatm a kes the drummer happy.

If you’ve got the toms on separatetracks, reach for your tried and trueeq and compression. Eq in a littlebottom, and maybe some crispattack around 6 kHz. Try to eq outsome 200 Hz muddiness, as with thekick. Compress for attack andpunch, and you’ve completed yourdrum mix—for now.

Get down

Moving on to bass, we find similarissues. We need to compress to bal-ance the bass line. Some notes arelouder than others, and some stringson the bass are quieter than others.Gentle compression (4:1 ratio orless) can even out these problems. A

slow attack time adds punch to thebass in exactly the same way we didit on the drums.

Release is tricky on bass guitar.M a ny compre s s o rs can release so fa s tthat they fo l l ow the sound as itcycles through its low fre q u e n cyo s c i l l a t i o n s . That is, a low note at,s ay, 40 Hz cycles so slow ly (onceeve ry 25 milliseconds) that the com-p ressor can actually release duringe a ch individual cy c l e .S l ow therelease down so that it doesn’t dis-

tort the wave form in this way andrides the sound from note to note,not cycle to cy c l e .

The obvious eq move is to add lowe n d . But be care f u l , as the tra ck maya l re a dy have a lot of low end. Th et r i ck is to get a good balance of lowf requencies from 30 through 300 Hz.Listen for a bump or dip in the

response—either too mu ch or too lit-tle in a single low fre q u e n cy are a ,and equalize in a corre c t i o n .

Glance back at your kick drum too.If your kick sound is defined in thel ow end, s ay at about 65 Hz, t h e nm a ke room for it in the bass guitarwith a complementary but gentlec u t . Find eq settings on both the kickand the bass so that the kick ’s punchand power don’t disappear when thebass fader is brought up.

We often add a touch of chorus tothe bass. This is most effective if thechorus effect doesn’t touch the lowf re q u e n c i e s . The bass prov i d e simportant sonic and harmonic stab i l-ity in the low frequencies; a ch o r u swith its associated motion and pitchbending would undermine this.

Simple solution: place a filter onthe send to the chorus and re m oveeve rything below about 250 Hz. Th echorus effect wo rks on the ove r t o n e sof the bass sound, a dding that desir-able richness without we a kening thes o n g ’s foundation at the low end.

Chugging on

I t ’s a ro ck and roll cliché to tra ckthe same rhythm guitar twice. Th et wo tra cks might be identical ineve ry way except that the perfo r-

...Or distort the

flanged reverb?

Anything goes.

Travel sa f e .


mance is oh so slightly, h u m a n ly differe n t . This results ina rich , w i d e , e a r-tingling wall of sound.

The effect is better still as the subtle differe n c e sb e t ween the two tra cks are stre t ched slightly. Pe r h a p sthe second tra ck is re c o rded with a different guitar, a dif-f e rent amp, d i f f e rent mics, d i f f e rent microphone place-m e n t , or some other slightly different sonic appro a ch .

In mixdown you make the most of this doubling bypanning them to opposite ex t remes: one goes hard left,the other hard right. Balance their levels so that the netresult stays centered between the two speake rs .

A touch of compression might be necessary to contro lthe loudness of the perfo r m a n c e , but often electric gui-t a rs are re c o rded with the amp cra n ked to its phy s i c a ll i m i t s , g iving it amplitude compression effects alre a dy.C o m p l e m e n t a ry equalization contours (boost one wherethe other is cut and vice ve rsa) can add to the effect ofthe doubled, s p read sound.

Key in

The clavinet completes our rhythm section. It pro b-ab ly wants compression to enhance its attack in mu chthe same way the kick , s n a re , and bass guitar we ret re a t e d .

Giving it a unique sound through eq and effects willensure that it gets noticed. Consider adding some flangeor distortion (using a guitar foot pedal or an amp simu-lation plug-in—or re-recording it through an actual amp)to make it a buzzy source of musical energy.

Panning it midway off to one side is a good use of thes t e reo soundstage. Pan it opposite the toms and sologuitar to keep the spatial counterpoint most ex c i t i n g .A dd a short delay panned to the opposite side for am o re live ly feeling.

With drums, b a s s , g u i t a r, and clav going in the mix,we ’ ve completed the rhythm section. Time to add the funparts: vocal and lead guitar.

Speak up

The vocal gets a good deal of our attention now. Th evoice must be pre s e n t , i n t e l l i g i b l e , s t ro n g , and ex c i t i n g .

P resence and intelligibility live in the upper midd l ef re q u e n c i e s . Use equalization to make sure the conso-nants of eve ry wo rd cut through that rich wall of rhy t h mg u i t a rs yo u ’ ve cre a t e d . S e a rch care f u l ly from 1 to may b e5 kHz for a region to boost the vocal that raises it out ofthe guitars and cy m b a l s .

You might have to go back and modify the drum andguitar eq settings to get this just right. Mixing re q u i re sthis sort of itera t ive appro a ch . The vocal highlights ap roblem in the guitars , so you go back and fix it. Tra d i n goff among the competing tra ck s , you’ll find a balance ofc rystal clear lyrics and perfectly crunchy guitars .

S t rength in the vocal will come from panning it to thec e n t e r, a dding compre s s i o n , and maybe boosting theupper lows (around 250 Hz). C o m p ress to control thedynamics of the vocal performance so that it fits in thec row d e d , hyped-up mix yo u ’ ve got screaming out of thel o u d s p e a ke rs .

This compression and equalization tra ck by tra ck hasso maximized the energy of the song that it wo n ’t fo rg ivea weak vo c a l . N a t u ral singing dynamics and ex p re s s i o na re often too ex t reme to wo rk—either the quiet bits aretoo quiet or the loud screams are too loud, or both.

C o m p ress the dynamic range of the tra ck so that it canall be turned up loud enough to be clear and audible. Th esoft wo rds become more audible. But the loud wo rds arepulled back by the compressor so that they don’t ove rdo it.

The vo c a l , a tiny point in the center, risks seeming alittle small re l a t ive to the drums and guitars . Th e‘ s p reader’ to the rescue (again, see the sideb a r ) . S e n dsome vocal to the spreader so that the vocal starts tot a ke on that mu ch desired larg e r-than-life sound.

As with a lot of mix move s , you may find it helpfulto turn the effect up until you know it’s too mu ch andthen back off until it’s just audible. Too mu ch spre a d-er is a common mistake , we a kening the vocal with a

ch o r u s e d - l i ke sound. The goal is to make the vo c a lm o re conv i n c i n g , a dding a bit of width and support ina way that the untrained listener wo u l d n ’t notice asan effect.

A dditional strength and excitement comes from may b ea high fre q u e n cy eq boost (10 or 12 kHz or higher!) andsome slick reve r b . The high fre q u e n cy emphasis willhighlight the breaths the singer take s , revealing more ofthe emotion in the perfo r m a n c e . It is not unusual to addshort reverb to the vocal to enhance the stereo-ness ofthe voice still further and to add a long reverb to givethe vocal added depth and rich n e s s .

Where do we sta rt? Well, the vocal is almost alw ay s

the most important single piece of every pop song.

So most engineers sta rt with... the drums.

S t r e n gth in the vocal will come from panning it

to the center, adding compression, and may b e

boosting the upper lows (around 250 Hz).


Sending the vocal to an additional delay or two isanother common mix move . The delay should be tuned tothe song by setting it to a mu s i c a l ly re l evant delay time( m aybe a quarter note). It is mixed in so as to be subtlys u p p o r t ive but not ex a c t ly audible.

A dd some feedback on the delay so that it gra c e f u l lyrepeats and fa d e s . Send the delay return to the longreverb too, and now eve ry wo rd sung is fo l l owed by awash of sweet reve r b e rant energy that pulses in timewith the mu s i c .

E q , c o m p re s s i o n , d e l ay s ,p i t ch shifting, and two kindsof reverb re p re s e n t , b e l i eve it or not, a normal amount ofvocal pro c e s s i n g . I t ’s going to re q u i re some ex p e r i m e n t a-t i o n , going back and forth among eve ry piece of the longp rocessing ch a i n .

And that’s just a basic patch . Why not add a bit of dis-tortion to the vocal? Or flange the reverb? Or distort theflanged reverb? A nything goes. Travel safe.

The back ground vocals might get a similar tre a t m e n t ,but the various parts are typically panned out away fro mcenter and the various effects can be pushed a littlem o re . Hit the spreader and the long reverb a little hard-er with back ground vocals to help give them more of thatmagic pop sound.

Going solo

The lead guitar can be thought of as replacing the leadvocal during the solo. It doesn’t have to compete with thelead vocal for attention, so your mix challenge is to get itto soar ab ove the rhythm section.

An eq contour like that of the lead vocal is a goods t rategy: presence and low end stre n g t h . C o m p re s s i o nshould be used with re s t raint if at all; electric guitarsa re natura l ly compressed alre a dy. A dditional reverb isalso unusual for guitars . The tone of the guitar is re a l lyset by the guitarist, and that includes the reverb builtinto the amp.

Solo guitar might get sent to the spre a d e r, and it mightfeed a short slapback delay. The slap delay might bes o m ew h e re between about 100 to 200 milliseconds long.It adds excitement to the sound, a dding a just perc e p t i-ble echo reminiscent of live concerts and the sound ofthe music bouncing back off the rear wa l l .

I t ’s good to pan the solo about halfway off to one sideand the slap a little to the other. If the singer is the gui-t a r i s t , it might make more sense to keep the solopanned to center.

Of cours e , you can add a touch of phaser, f l a n g e r — s o m e-thing in your digital multieffects unit that yo u ’ ve beendying to try, and you can even add additional distortion.

Overall

The entire stereo mix might get a touch of eq and com-p re s s i o n . As this can be done in mastering, I re c o m m e n dresisting this at firs t . But as your mixing chops are deve l-o p e d , you should feel free to put a re s t rained amount ofs t e reo effects across the entire mix.You are trying tom a ke it sound the best it possibly can, after all.

For equalization, u s u a l ly a little push at the low sa round 80 Hz and the highs around or ab ove 10 kHz isthe right sort of polish. Soft compression with a ratio of2:1 or less, s l ow attack and slow release can help makethe mix sound even more pro f e s s i o n a l .

As the entire mix is going through this equipment, m a kes u re you are using good sounding, l ow noise, l ow distortioneffects dev i c e s . And don’t fo rget to ch e ck your final mix inmono to make sure it’ll surv ive radio airp l ay.

That sums up the components of one appro a ch to onem i x . It is meant to demonstrate a way of thinking ab o u tthe mix, not the step by step rules for mixing. I hope iti n s p i res you to form your own variation on this appro a ch .

A l e x ’s mixes often feat u re didgeridoo panned dead centerand doubled kazoos panned hard left and right.C o m p l a i nabout this to case@re c o r d i n g m a g . c o m .

What’s a spreader?I t ’s often desirable to take a mono signal and make it a little more stereo-like. A standarde ffect in pop music is to spread a single track out by sending it through two short delays.Each is set to a diff e rent value somewhere between about 15 and 50 milliseconds. Not toos h o rt or it starts to flange/comb filter; not too long or it pokes out as an audible echo.One delay re t u rn is panned left and the other panned right. The idea is that these quickdelays add a kick of supportive energy to the mono track being processed, sort of like theearly sound reflections that we hear from the left and right when we play in a real ro o m .The extra trick is to pitch shift them ever so slightly, if you have the gear that can do it.That is, take each delay and detune it by a nearly imperceptible amount, maybe 5 to 15cents. Again, we want a stereo sort of effect, so it is nice if the ‘spreader’ has slightly dif-f e rent processing on the left and right sides. Just as we dialed in a slightly diff e rent delaytime for each side, dial in a slightly diff e rent pitch shift as well—maybe the left sidegoes up 9 cents while the right side goes down 9 cents.Now we are taking advantage of our signal processing equipment to create a widenedsound that only exists in loudspeaker music; it isn’t possible in the physical world. Thiss o rt of thinking is a real source of creative power in pop music mixing: consider a phys-ical effect and then manipulate it into something that is better than reality (good luck,and listen carefully). We are going to add this effect to the lead vocal, among others. And the lead vocal isgoing to be panned straight up the middle. In order for the spreading effect to keep thevocal centered, it helps to do the following.Consider the delay portion of the spreader only. If you listen to the two panned short delays(and I definitely recommend trying this) you find the stereo image pulls toward the short-er delay. Now listen to just the pitch side of the spreading equation. The higher pitch tendsto dominate the image. Arrange it so that the two components balance each other out (e.g.,delay pulls right while pitch pulls left). This way the main track stays centere d .Experiment with diff e rent amounts of delay and pitch change. Each offers a unique sig-n a t u re to your mix. Overused, the vocal will sound too digital, too pro c e s s e d .C o n s e rvatively applied, the voice becomes bigger and more compelling.

If your kick sound is defined in the low end, say at about

65 Hz, then make room for it in the bass guita r

with a complementa ry but gentle cut.

BY ALEX CASE

PART 11

Preproduction and live recordingThe Session

Preproduction and live recordingThe Session

RECORDING MAY 2000

Su re ly part of the pleasure of music re c o rding is that itis such a fre e , l i q u i d , no-rules sort of endeavo r. In this

episode of ‘Nuts & Bolts’ we look at the actual process ofre c o rd i n g .( Ye s , I know we pro b ab ly should’ve done thisb e f o re looking at a mix the way we did last month, but itwas Mixing’s Art And Science Month in April and thoseEditor guys asked so nicely. . . . )

Armed with the specific knowledge of components ofthe re c o rding chain discussed so far in this series, l e t ’sdiscuss the actual session and our cre a t ive and tech n i c a loptions along the way. Th roughout this article I’ll be dis-pensing advice and then making the case for it; whilethis is all based on ex p e r i e n c e , bear in mind that differ-ent pro d u c e rs and artists have different ways of wo rk i n g .So don’t get mad if you disagre e .

Begin at the beginning

Perhaps the single most neglected part of making a goodre c o rding is pre p ro d u c t i o n . It is an investment that allbands and pro d u c e rs should make . And I’m a fan of invo l v-ing the re c o rding engineer during pre p roduction as we l l .

Big budget artists as well as struggling up-and-comersneed to scrounge up the time it takes to wo rk off-stage ina cheap studio, re h e a rsal space, or gara g e , and makerough re c o rdings of the songs they plan to re c o rd later.

Think about it. The way most people hear our music isby listening to the re c o rding over loudspeake rs . The waythe band listens to the music of other artists is by listen-ing to those re c o rdings over loudspeake rs . But the waythe band listens to their own music is live at the gigs, o rduring re h e a rsals and jam sessions. Th e re is an unfo r t u-nate inconsistency here .

Ta l e n t e d , passionate new artists often create a bandthat is simply thrilling live . Then the album fails to“ c a p t u re ” t h i s . Quite possibly all that has gone wrong isthat the band hasn’t had a chance to listen to them-s e l ves the same way they listen to all the bands theyl ove—the same way their future fans will listen tothem: on loudspeake rs .

G ive the artists a chance to react to themselves as theyappear in loudspeaker play b a ck and they’ll often makethe appropriate adjustments necessary to sound great ona re c o rd i n g . The same band that re a l ly wo rks the crow dl ive can often wo rk the loudspeake rs through theirre c o rdings; they just need a ch a n c e .

P re p roduction re q u i res just a few mics and a cassetted e ck . Wo rking with more mics and a DAT or 8-tra ckre c o rder is sometimes even better. The mission of pre p ro-duction is to capture the performances on tape for studyand evaluation later.

M a ny bands have never actually heard themselve suntil the first take in the studio on the first song of thef i rst session for their first album. Th e re is alre a dy a lot ofp re s s u re built in to that first studio situation. I t ’s a lot ofm o n ey. Th e re are a lot of mics all over the place. Th e re isa lot of gear in the control room with lights and metersevaluating eve ry thought the musicians have .

For the first-time re c o rding artist, an unders t a n d ab l ep a ranoia sets in. An overwhelming fear of making mis-t a kes that will be capture d , amplified and mocked byeve ry mic, meter and loudspeaker in the studio leads toa performance that is more conserva t ive , less ex c i t i n g .Th a t ’s not the sort of vibe that will lead to a Gra m my -winning perfo r m a n c e .

RECORDING MAY 2000

If the band has never heard them-s e l ves befo re , get re a dy for somech a l l e n g e s . Think back to the firs ttime you re c o rded yo u rs e l f. Wh e nyou are n ’t play i n g , and you are justl i s t e n i n g , you start to hear thingsthat have perhaps gone unnoticedfor ye a rs . I drift flat when I singl o u d , I rush during the solo, and I dothis funny thing at the end of thebridge that just sounds aw f u l — Ia lways thought it sounded awe s o m e .

The band deserves a chance towo rk these things out ahead of thealbum sessions. The fact is, mu s i c i a n swill fix many technical issues ontheir own if you just give them atape of some re h e a rs a l s . The drum-mer will stop rushing during the ch o-r u s , the singer will plan out some ofthose “ o o h s ” and “ a h s ” at the end,e t c . M a ke a rough re c o rding of thep re p roduction session for eve rymember of the band.

The songwriter also benefits fro mp re p ro d u c t i o n . Most pop music songsa re studied on paper: meter, r hy m e ,wo rd ch o i c e , and structure are eva l u-ated with the same care given ap o e m . Songs differ from poetry inthat they are set to mu s i c . The song-writer should there fo re get thechance to study his or her wo rk as itl ives on loudspeake rs . M a ke a ro u g hre c o rding for the songwriter.

The project engineer also benefitsf rom doing the re c o rding during pre-p ro d u c t i o n . The audio quality of thefinal product will improve marke d lyif eve ryone gets to hear what theyand their instruments sound likecoming back off tape.

The drummer may not notice thes q u e a ky kick pedal during perfo r-m a n c e s , but during play b a ck eve ry-one will. The guitarist may not seemto know that the strings on her gui-tar are re p l a c e ab l e , but during play-b a ck the sad, lifeless tone mightm o t ivate the effo r t .

R e c o rd the instrument and yo u ’ l lfind its eve ry we a k n e s s — g u a ra n t e e d .If the squeaky pedal and dull oldstrings are discove red befo re the bigs e s s i o n , then the problem can bea dd re s s e d . If it happens in the heatof the actual album-making session,

you’ll find yo u rself trying to smoothover and hide a problem or wa s t i n gp recious studio time and cre a t ivee n e rgy waiting for someone to run tothe music store for the $5 solution.

It is, but it isn’t

The rules for the preproductionsession:

Fi rs t , t reat it like the actual session.E ve ry one must put their hearts intothe session 100% and make it count.

The second rule—and this is iron-ic—is to make sure everyone knowsthat it’s not the actual session. Aline must be drawn between prepro-duction and session work. The bestway to extract all the benefits ofpreproduction is remove the tempta-tion to keep some of the takes orsome of the tracks.

At first it seems perfectly log i c a l :re c o rd the pre p roduction session toDAT, and if they nail a take we’ll useit on the re c o rd .R e c o rd the re h e a rs a l sto mu l t i t ra ck , and if we get a killervocal take we’ll use on the album.

B e e p / Q u a ck/Eep or whatever noiseyour computer makes when an erro ris made. Trying to rescue the vo c a lt a ke from pre p roduction and use it onthe album will draw so mu ch atten-tion and re q u i re so mu ch effort thatyou’ll fail to pro p e rly evaluate therest of the re c o rding; the whole re a-son for pre p roduction is undermined.

H ow can the songwriter change awo rd or two later if the tra ck isa l re a dy re c o rded? “ We’ll just punchin the new wo rd s ,” someone say s .

B e e p / Q u a ck / E e p. M a t ching thesound of the vocal will be a lot oft rouble when you leave the re h e a rs a lroom and go to the fa n cy, a c o u s t i cp a radise of the re c o rding studio,when you leave the live vocal mic(the Indestructo X2000) in the va nand start using the sweet vocal mic(the Delicato Tube2k) in the studio.And asking even wo rld-class drum-m e rs to ove rdub their drum perfo r-mance to an alre a dy existing vo c a lt a ke is ra re ly successful.

The point of pre p roduction is todocument with adequate sonic quali-ty all the music and perfo r m a n c eideas that the band has as of today.Then these ideas are evaluated dur-ing play b a ck over loudspeake rs ,w h e rever people prefer to listen:t u r b o - t we a ked mega-hi-fi systems, i nthe car, in headphones, a ny w h e rew h e re you do a lot of listening to there c o rdings you buy.

M i s t a kes become audible, and aremost always fixable befo re the albums e s s i o n s . And more ex c i t i n g , t h i sband that you like so mu ch live willcome up with ideas for modifying

Excerpted from the May edition of RECORDING magazine.©2000 Music Maker Publications, Inc. Reprinted with permission.



RECORDING MAY 2000

the arra n g e m e n t , song structure , g u i-tar tone, ly r i c s , and so on that willb l ow you away. G ive them a re c o rd-ing of how they sound and let themdo what they are re a l ly good at: mak-ing their own music sound gre a t .

P re p roduction also gives the pro-ducer and engineer a chance to con-tribute meaningfully to the cre a t ivemusic making pro c e s s . The jobs ofp roduction and engineering happenin the studio. P ro d u c e rs and engi-n e e rs have a familiarity with thegear of the studio like mu s i c i a n sh ave with their instruments.

The studio experience of the pro-duction and engineering teame n ables them to make musical sug-gestions that are unique to re c o rd e dmu s i c . Double the vocals in the ch o-r u s e s , a dd slap-back to the guitarduring the solo, use some gatedroom mics on the drums, run thepiano tra ck through a Leslie cab i-n e t , e t c . — t h e re is a vast sonicpalette to choose fro m .

These are creations that re ly onthe studio and its equipment to bec re a t e d . Th ey re ly on loudspeake rp l ay b a ck to be re a l i ze d . It is impera-t ive that the producer and engineerlook out for these audio concoctionsthat will contribute to the music andt ranslate it into an action that theband understands and appre c i a t e s .

The band is expected to have anopinion on how appropriate suchsounds are to their mu s i c , but it isthe job of the studio cats to be ab l eto create them. P re p roduction give sthe producer and engineer their firs tchance to start making these studiod e c i s i o n s .

S t rategize: who, what ,when, where, why?

Before the actual album sessionsbegin, the producer, engineer, andband should develop a recordingstrategy. It’s just a schedule of whorecords their instruments and when.In what order should the songs berecorded? Which tracks get recordedfirst, and which are overdubs? Thesesorts of decisions are important towork out.

L i ke so mu ch in mu s i c , t h e re isra re ly a single right way to do things.But some appro a ches are more use-

ful than others . Yo u ’ ve got to decideamong the live to 2-tra ck , l ive to mu l-t i t ra ck , b a s i c s , or ove rdub sessions.This month we discuss the live ses-s i o n s . N ext month’s ‘Nuts & Bolts’t a kes on basics and ove rd u b s .

Live to two

It isn’t always necessary to re c o rdto a mu l t i t ra ck . If you are re c o rding as i n g l e , simple instrument, you canre c o rd it straight to your 2-tra ck mas-ter mach i n e — p ro b ab ly a DAT. S o l op i a n o, vo i c e , or guitar are obv i o u sex a m p l e s .

Without the distraction of otherinstruments and perfo r m e rs , t h eengineer can re a l ly fo c u s . M i x d ow nwo n ’t be necessary, as there is noth-ing to mix the solo instrument with.Capturing the tone and adding justthe right effects is the sole priorityof a live to 2-tra ck session.

Your decision to go live to twos h o u l d n ’t be based on engineeringc o nvenience or desires alone. Th ere c o rding strategy must also fa c t o rin the musical advantages and dis-a dvantages as we l l . In a live to twothe performer is as focused as thee n g i n e e r, chasing that elusive goal—their best perfo r m a n c e .

An important musical benefit ofthe single player live to two sessionis that there are no other mu s i c i a n sa ro u n d . Other playe rs often addp re s s u re , stop take s , or re q u i rec o m p ro m i s e :

Singer: “ L e t ’s use take 17! Listento how I phrased the opening line.”

Drummer: “But I fumbled that fillin the first ch o r u s . I’m re a l ly diggingt a ke 12.”

In many live to 2-tra ck sessions itis just an engineer looking for asweet sound and a musician search-ing for his or her personal best.

Of cours e , t h e re is still opportunityto modify and enhance the live

re c o rd i n g . Post re c o rding pro c e s s i n gconsists of two options: editing andm a s t e r i n g .

You can cut and paste together (lit-e ra l ly or digitally) the best parts of allthe takes into a single best take .A n dyou can master the 2-tra ck tape justre c o rd e d .That is, you can still modifythe sound of the re c o rding with a finaldose of any effects you desire — t y p i-c a l ly equalization and compre s s i o n ,but there is no reason not to addreverb or more elab o rate effects aswe l l . Do whatever you think soundsb e s t .

To ach i eve simplicity and intimacy,we plan on a live to two re c o rd i n gs e s s i o n . But live to two isn’t just fo rsolo instruments. We can certainlyre c o rd more complex arra n g e m e n t sand bigger bands live to two tra ck s .

L e t ’s put it in context by skippingahead for a moment to that commonmu l t i t ra ck session, the ove rd u b . S ayd r u m s , bass and guitar have beenre c o rd e d . Time for a saxophoneove rd u b .

Consider the vibe at the ove rd u b .The saxophone player is all alone inthe studio, p l aying into perhaps asingle micro p h o n e , l iving in a mu s i-cal wo rld that exists within the head-p h o n e s . It isn’t easy to find the killersolo that will take over the wo rl dwhen yo u ’ re playing all alone inh e a d p h o n e s . Certain components ofmusic feed off the live interaction ofother mu s i c i a n s .

This sax solo might benefit frombeing recorded at the same timethat the rest of the band plays.Record it live.

And there are other instancesw h e re the live to two is tempting.D r u m m e rs and bass playe rs are oftenso mu s i c a l ly intera c t ive that theyp refer tra cking together (don’t missour discussion of the Basics session

n ext month). If you can re c o rd thesolo during the inspired gro ove ofthe live session, you’ll find moreex p re s s ive n e s s , m o re powe r, m o ree m o t i o n .

Certain styles of music are built on afoundation of interaction: jazz, b l u e s ,and power trios often like to be re c o rd-ed all at once. H i g h ly improv i s a t i o n a lmusic is difficult to pull off mu s i c a l lyt h rough an assembly of ove rd u b s .

The final product will improve

m a r k e d ly if everyone gets to hear

w h at they and their instruments

sound like coming back off ta p e .

Perhaps the band hasn’t had a chance

to listen to themselves the same way

they listen to all the bands they

love—on loudspeakers.

RECORDING MAY 2000

L ive to two becomes a mu ch moreintense session now. Two tra cks ofre c o rded music can easily come fro mm o re than a dozen micro p h o n e saimed at any number of instrumentsp l aying live , at once. And elab o ra t esignal processing might be re q u i re d .

Skip the coffee. You’ll have plentyof adrenaline as you adjust the leve l son all those microphones; dial inequalization that is just right for eachof them; set up compression on half, p ro b ab ly more ofthem; send the snare to a plate reve r b , the Rhodes to aquarter note delay, and the vocal to both the reverb andd e l ay; and so on.

Yo u ’ ve got to hear eve ry little thing going on micro-phone by micro p h o n e , instrument by instrument, a n deffects unit by effects unit in the live to two session.

In add i t i o n , you must somehow hear the big thing: theove rall 2-tra ck mix itself. B a ck in the day, e n t i re orch e s-t ras we re re c o rded live with a single well-placed micro-p h o n e . It can be done, but it’s always something of athrill ride. Consider these ideas to help out.

Safety net? What safety net?

Fi rs t , k n ow the tune. Try to get a ch a r t , attend are h e a rs a l , get a tape from the pre p roduction session (seeab ove ) , and/or just plain learn the tune in detail duringthe first couple of take s . Yo u ’ ve got to know what thesong is about and memorize the arrangement: know whois playing when, when the loud parts are , when the softparts are , and ride the fa d e rs accord i n g ly.

S e c o n d ,t a ke a “ l ive ”a p p ro a ch to the re c o rding tech-n i q u e . We know that in a live to two tra ck session there willbe no mixdown later. The good news is that in a live to two -t ra ck session we don’t perform ove rd u b s .

Musical issue: it’s hard on the perfo r m e rs . Th ey ’ ve gotto get the performance just right, as there can be no fix-ing of mistake s , just repeated attempts at the tune—“ O k ay. H e re we go again. Ta ke 94...ro l l i n g . . .”

But it’s good news for the engineer. I t ’s fine if the vo c a lleaks into the guitar mic and the drums leak into theo rgan mic. We’ll never re re c o rd one without the other, s os u ch co-mingling of sounds—we call it leakage—ofteni s n ’t a pro b l e m .

L ive re c o rding liberates the engineer of all thoseh e a d a ches associated with trying to separate the play-e rs and get clean tra ck s . B ye bye booths. G o o d byeg o b o s . No need to hide the guitar amp in the closet andthe bass amp in the basement (isn’t that why it’s calleda basement?).

We constantly go to such trouble to ach i eve isolation inmu l t i t ra ck sessions. And those habits die hard .

Yo u ’ ve gotta try it live and loose. S t i ck all the playe rsin one room and live it up. Th ey can arrange them-s e l ves in the way that is most comfo r t able for them—p ro b ab ly the way they re h e a rs e , the way they write thes o n g s , or the way they play live . A r ranged this way,t h ey are so comfo r t able they might fo rget they arebeing re c o rd e d . This is a good way to capture some-thing special on tape.

A r ranged this way they can see eacho t h e r. M o re ove r, t h ey can hear eachother acoustically. So you can get rid ofthe headphones. Headphones are a dis-t racting part of any session for thee n g i n e e r. Musicians don’t like ‘ e mmu ch either. Th ey don’t make for ave ry exciting or comfo r t able env i ro n-ment to jam. Headphones are a neces-s a ry evil in mu l t i t ra ck pro d u c t i o n . B u tl ive re c o rding often permits you to dis-

pense with them altog e t h e r.What should we wa t ch out for when we put the band

all in one room? Fi rst the good new s . When an instru-ment is picked up by microphones other than its ow n , amagic thing starts to happen. This leakage into othermics starts to capture a different view of the instrumentthan a closely placed mic can manage alone. When it iswo rking it starts to make the instruments come tog e t h e rinto a more compelling single sonic ensemble. The bandwill sound tighter, the song will gel.

The live re c o rding might lack the precision that cancome from well-isolated tra ck s , but it gains a more inte-gra t e d , m o re organic total sound that is often we l l -

aligned with the aesthetic of the music being re c o rd e d .The music we tend to re c o rd live to two, this highlyi m p rov i s e d , h i g h ly intera c t ive sort of music benefits fro mbeing re c o rded in this sonically integrated way.

Manning the board

R e c o rding the band all at once in a single ro o mre q u i res you to keep in mind two key issues: pro c e s s i n gand panning. If you plan to apply some heavy pro c e s s i n gto a single instrument, you need to minimize leakageinto its micro p h o n e s .

For ex a m p l e , the vocal signal is pro b ab ly going to getsome careful signal pro c e s s i n g . Perhaps you want to useeq and compression to keep it strong and audible in them i x . If there is too mu ch bass leakage into the vocal micyou might find yo u rself getting into some tro u b l e .

When you try to eq in some low end strength to thevoice itself you also bring up the unwanted low fre q u e n-cy bass leakage. If the snare stumbles loudly into thevocal mic, i t ’s going to cause the vocal compressor toreact differe n t ly. If the singer we re facing the band andsinging into a cardioid micro p h o n e , m a x i mum acousticrejection could be ach i eved and the problem avo i d e d .

The lead vocal deserves special attention. On the otherh a n d , the microphone sitting brave ly in front of theFender Twin Reverb wo n ’t notice a bit of vo c a l . Yo u ’ vegot to minimize leakage into the quiet instruments, e s p e-c i a l ly the ones getting ex t ra pro c e s s i n g , but most of theother leakage isn’t a problem at all.

The second issue to manage in a live re c o rding situa-tion is panning. As instruments are panned left and right(and rear for you surround sound ex p e r i m e n t e rs outt h e re ) , yo u ’ ve got to listen for the impact it has on thesound that has leaked into the mic.

Consider the good ol’ snare . The sound of the snare isgoing to be audible in pretty mu ch eve ry micro p h o n e

Two tracks of recorded music can easily come from

more than a dozen microphones aimed at any number

of instruments playing live at once.

Skip the coffee—you’ll have plenty of adrenaline.

musical benefits of the live session arec a p t u re d .R e c o rd with similar stra t e-g i e s .A r range the musicians to maxi-m i ze their comfort and encourage theirc re a t iv i t y. Seek advantageous blendingof the instruments in the room thro u g hs t rategic mic placement that capture sthe tone of the instruments and a gooddose of acoustic leakage. But do avo i dtoo mu ch leakage on those tra cks des-tined for a good dose of signal pro c e s s-ing or aggre s s ive panning.

The live to mu l t i t ra ck session take ssome of the pre s s u re off the engineeras the priority is all about sessionv i b e , musician comfo r t , and awe s o m eraw tra ck s . The mixing of the tra ck swill get to happen in a separa t e , l e s sc row d e d ,l ower stress session.

N ext month we ex p l o re the moretypical production process: re c o rd i n gto mu l t i t ra ck and then ove rd u bb i n ga ny number of additional tra cks sothat they can be mixed into a powe r-f u l , p o l i s h e d , and professional stere omaster re l e a s e .

Alex Case thinks they should callt h at TV show ‘ S at u r d ay Night Live toTw o .’ Request Nuts & Bolts topics viac a s e @ re c o r d i n g m a g . c o m .

that is in the same time zo n e . If thes n a re leaks loudly into the guitarand you pan the guitar to the left,then you’ll hear the snare imagedrift left. If the snare sound alsoleaks into the piano that getspanned right, the ove rall snare soundcan stay more centere d .

In fa c t , the sound of the snare inthe more distant microphones oftensounds fa n t a s t i c .You might want toplan your panning strategies so thatleakage like that of the snare can bekept under contro l .You may have tob a ck off on the ex t reme pan pot set-t i n g s , pulling things in closer to cen-ter to keep the stereophonic image ofthe band tighter.

A l t e r n a t ive ly, you might use leak-age on purp o s e .K n owing that thes n a re will leak into the acoustic gui-tar tra ck that you want to pan left,you might use an omnidirectional micon the piano panned right to pick upex t ra snare leakage on purp o s e .

With a little attention to theses t rategies on processing and pan-

n i n g , you’ll find re c o rding the bandall at once in a single room is a liber-ating way to wo rk .

Live to multi

Sometimes it just isn’t possible tomeet the audio demands of the pro-ject in a live to two. Wild and compli-cated arrangements and large bandsm a ke getting the mix right whilere c o rding nearly impossible. If yo u ’ vegot drums, b a s s , g u i t a rs , key s , a horns e c t i o n , a chorus section, lead vo c a l s ,and miscellaneous hand perc u s s i o n ,the session is pro b ab ly too complicat-ed for a live to two tra ck appro a ch .

The live feeling and sonic benefitsof a live to two session can also bec a p t u red in a mu l t i t ra ck env i ro n-m e n t . Just because the music needsto be re c o rded all at once doesn’tmean the engineering has to happenall at once. That is, we can re c o rd theband with all the live and intimatea p p ro a ches described ab ove and stillmix it down later. R e c o rd the livesession to mu l t i t ra ck . Old timers likeme call this ‘ l ive to 24,’ but as mydigital audio wo rkstation goes to (anot yet utilized) 64 tra ck s , it seemssafer to call it ‘ l ive to mu l t i .’

All or most elements of the tune arere c o rded simu l t a n e o u s ly so that the

RECORDING MAY 2000

The drums are easily the most difficult partof the live to two challenge. A small kit might con-

sist of kick, snare, hi-hat, two rack toms, a floor tom, aride cymbal and a crash cymbal.T h a t ’s a small kit. It might also have more drums, hand per-

cussion taped on, two kick drums, a couple of snares, as well as along list of additional toms, cymbals, gongs, trash can lids, cere a l

boxes, buckets and beer kegs.What a mess. It’s an engineering challenge: one player, many instruments. And these

i n s t ruments are all in close proximity to each other.

Mic placementYou might want to abandon the idea of close-miking every piece of the kit in a live to

two session. Try instead to capture the entire kit with a pair of o v e rhead m i c rophones placedabove the drum set. Placing just a couple of mics to capture so complicated an instrument isa skill acquired through experience, so try to give yourself room to make and fix mistakes.

For the stereo image to work in loudspeakers you must arrange mics symmetrically aboutthe kit. Place the mics so that the snare and kick remain as close to the center as possible.

These overheads might be coincident (the two directional capsules placed as close to eachother as possible but oriented in diff e rent directions) in either XY or MS configuration. Or theymight be a spaced pair of microphones, looking down on the kit and the room from above ando ff to the sides. Please refer to the 2/97 issue for more about stereo miking configurations.

How high? To make this determination you’ve got to wrestle with two conflicting tre n d s .First you’ve got to find the right drum sound versus room sound combination. As you movethe mics away from the drums (higher or farther) they’ll pick-up more and more of theambient sound of the room. The amount of ambience is a matter of musical judgement.Make sure it supports the mood of the song—somewhere between tight/focused andwild/out of control.

The second issue is that moving the overhead microphones will adjust the relative bal-ance between cymbals versus snare/kick/toms. When the overheads are in close to thecymbals they act more like ‘cymbal mics.’ As you back the mics away from the kit they startto view the whole kit without pre f e rence to the cymbals. With the mics down close the snaremight be four times farther away from the overhead mic than the crash cymbal; pull themic away and the relative distances from mic to snare and from mic to cymbal converg e .

So placement of overheads determines the cymbals vs. drums balance and the dru m svs. room combination. Listen to both trends as you move the overhead microphones andyou’ll find the sweet spot that solves both pro b l e m s .

Mic typeCondenser mics are the most popular choice here because they tend to be best at cap-

turing transients. The drum kit rattles out transients from down beat to fade out. Ribbon microphones, with their unique high frequency detail, are also a good choice—if

they can take the sometimes very high sound pre s s u re levels booming out of the kit. New rib-bon microphones can handle this; old (and nearly irreplaceable) ones probably can’t. Powerrock and roll playing will crush them; more laid back jazz and brushwork are no problem.

Moving coil dynamic microphones certainly are n ’t ruled out for the pair of drum over-heads. These days they often have a terrific transient response, so they are up for the jobof capturing the percussive detail of the drums accurately.

And sometimes we throw accuracy out the window. Dynamic mics can act a little likec o m p ressors when highly transient waveforms hit their capsules. And what musical com-p ressors they are! Rock drums can benefit from being carefully captured by a condenseror ribbon, and they make a make a musical statement when massaged ever so by ouri n d e s t ructible friend the moving coil dynamic.

Pick-up pattern Anything goes here. Cardioids let you ‘aim’ the microphones to tailor the sound. To o

much ride cymbal and not enough toms? Just rotate the microphone so that it faces moret o w a rd the toms and looks less directly at the cymbal.

F i g u re eight patterns let you do some ‘aiming’ as well. The bidirectional microphone ismost sensitive to sounds directly in front and directly behind the capsule; it totally re j e c t ssound incident from the sides.

While it’s typical to think of them as ‘hearing’ mostly what’s in front and what’s behind,I find it helpful to focus on what they don’t hear. Think of the bidirectional mic’s re j e c t i o no ff to the sides as your tweaking tool: for instance, for less ride cymbal, rotate the figureeight pattern so that the ride falls a little more into the side rejection area.

O n m i d i rectional patterns are also a good choice for overhead microphones, though theya re a little more difficult to place. Capturing acoustic energy from all directions, they’ll grabm o re ambiance than a cardioid or figure eight. As a result, to get the same balance of ro o msound versus kit sound, omnis will need to be closer to the kit than more directional mics.

Less obvious is the fact that the omni mic is a simpler, arguably purer device than thec a rdioid mic. Directional mics re q u i re a little signal processing to achieve rejection in cert a i nd i rections. It’s usually very careful, clever, and excellent sounding, but even acoustic signalp rocessing of the highest quality pays a price. To over-generalize gro s s l y, omnis often havea sweeter low frequency character than a lot of directional microphones, but of course thiss o rt of thing varies from one mic model to the next. There are cardioids with fantastic lowend and omnis that are low frequency deficient. What I’m really trying to say is that choos-ing between omni, bidirectional, and cardioid isn’t just about pick-up pattern, it’s also aboutf requency re s p o n s e .

C o m p a re not just the blend of cymbals versus toms and drums versus room sound, butalso the sound quality of the drums coming through the mics. Listen to the spectral andtimbral effect of choosing a diff e rent pick-up pattern .

More than two drum mics?It is possible to capture the entire kit with just a pair of overheads. In fact a single micro-

phone can work, placed either overhead or down in the kit tucked between the snare andthe rack tom opposite the hi-hat.

Using so few microphones on so broad an instrument re q u i res that you have time to re a l-ly tweak the mic placement and that you have a nice sounding room to help balance the sound.For this sort of work, first listen to the kit in the live room, then position the mics, and finallylisten in the control room. You’ve got to listen to the whole kit as well as all its individual pieces.

If you don’t like what you hear, re t u rn to the room with a specific objective in mind (e.g.too much crash cymbal, or snare pulls left) and move the mics (or change the mics, changethe pick-up pattern, move the kit, etc.) in a way that you think will help. Return to the con-t rol room, listen, and repeat...and repeat...until you love the overall balance of the kit.

This sort of judgement also re q u i res experience. Just using one or two overhead micson the drums re q u i res finesse. More typically, we support the overheads with a couple ofclose mics, even in a live to two-track session.

First, the kick drum welcomes a dedicated mic. To extract a decent amount of low endthump without too much messy room ambience, you’ve got to get a microphone in close.The kick is loud, so you’ll need a mic with the ability to handle high sound pre s s u re levels(up to and above 120 or even 130 dB SPL). Many condenser mics these days can take it,but most of the time the kick demands the robustness of a dynamic.

The snare, so important musically, also gets the special attention of a close mic. In theheat of a session you may not be able to count on the drum balance that you can pull outof the overhead microphones alone. Sticking a mic in close to the snare lets you ride afader to change the amount of snare in the live mix—a handy thing.

A dynamic cardioid mic is up to the job, and especially for rock it grabs a present tone thatwill sound exciting. Condensers and ribbons are also desirable for the high frequency detail.

If you go for a condenser, it probably needs a pad to prevent nasty distortion of them i c ro p h o n e ’s electronics. If you use a ribbon mic (especially an old one) on the snare, takeout some insurance or book the studio under a false name—one hit, one shredded ribbon.

EffectsFor multitrack sessions it is common to eq and compress every single drum track on its

way to the tape machine—in pop and rock, anyway. For a live to two, you’ve got to backo ff on this approach; it’s suicidal to dial up a stack of effects in a live to two session.

If you’ve just got overheads up, gentle compression is probably welcome. Three goals:1. S a f e t y : use compression to prevent the distortion that comes with levels to tape/diskthat are too hot.2. P u n c h : use compression to tighten up each hit of the drums and add a bit more attack.3. Care: not too much compression or you’ll hear the decay of the cymbals become unnat-ural, pumping softer then louder as the compressor rides the gain too aggressively.

These conflicting goals force us to back off the compression on the overheads significantly. If you’ve added close mics to the kick and snare, go ahead and compress them hard so

that they add punchiness, clarity, and attack to the overall sound in the overheads. The ideais to get maybe 80% of the drum sound from the overheads. Sneak in the close micro-phones to add that extra little power and detail.

This two to four microphone approach should enable you to get the kit under control inp retty short ord e r, freeing you to focus on the bass, and the guitar, and the vocal, and…

I s n ’t live to two a blast?

Learn By Drumming: Live To TwoLearn By Drumming: Live To Two

RECORDING JUNE 2000

This month we continue our discussion of the re c o rd i n gsession vo c ab u l a ry, m oving beyond pre p roduction and

l ive re c o rding and into the mu l t i t ra ck re c o rding pro c e s s .We tra ck instruments one at a time when live to two -

t ra ck and live to mu l t i t ra ck are n ’t appropriate or possi-b l e . The making of the re c o rding becomes now a 4-stepp rocess: basics, ove rd u b s , m i x d ow n , and mastering.

Mastering is the sole topic of a future Nuts & Bolts col-u m n . M i x d own was the focus in the April issue. Th i smonth we take on basics and ove rd u b s .

The basics session is simply the first tra ck - l aying ses-s i o n . As all ove rdubs will be performed around theset ra ck s , t h ey are called the basic tra ck s .

In 99.9% of pop and ro ck sessions, ‘basics’ re f e rs todrums and bass. When the band is to be re c o rded a pieceat a time through ove rd u b s , it is usually easiest to layd own the fundamental gro ove of the tune firs t .D r u m m e rs and bass playe rs usually play off each otherand there fo re like to be re c o rded simu l t a n e o u s ly.

Th a t ’s easy for the re c o rding engineer to accommodate.Drums are placed in the biggest room ava i l able in the stu-d i o / l o f t / b a s e m e n t .The bass is re c o rded through a dire c tb ox and/or through a bass cabinet isolated in any way pos-s i b l e — s t u ck in another room or booth, t u cked in a closet,or in the wo rst case surrounded by gobos and heavy blan-kets to at least minimize leakage into the drum mics.(Gobos are movable absorbent isolation barriers . )

The bass player stands in the same room as the drum-mer and the jamming commences. R e c o rded on the mu l-t i t ra ck , these drum and bass tra cks form the basic tra ck s .

Guided by voicesWith the exception of drum and bass mu s i c , p l aying a

tune that consists solely of drums and bass is mu s i c a l lynot ve ry inspiring. I t ’s easy to get lost during a take . S o

the drummer and bassist don’t get lost, we always ke e pthem on a leash.

But I digre s s . So the drummer and bassist don’t get lostwithin the tune, we also re c o rd scra t ch tra ck s . The singer,g u i t a r i s t , key b o a rd i s t , and other members of the rhy t h msection are also re c o rded during the basics session. Th e s ea dditional rhythm section tra cks are just meant as aguide to the drums and bass and will be re - re c o rded later.

The top engineering priority is the quality of the drumand bass tra ck s . S c ra t ch tra cks are compromised sonical-ly in pursuit of better basic tra cks when necessary. Fo rex a m p l e , to keep the guitar from leaking into the drumsit might be run through a small practice amp instead ofthe louder- t h a n - l o u d , Wo rld Trade Center twin towe rs ofguitar rig. The vocalist might be squeezed into a smallbooth for isolation, singing into a second-choice micro-phone because all the good mics are on the drum kit.

The point of the scra t ch tra cks is to feed the drumsand bass information and inspiration; the point of thebasics session is to get the most compelling drum andbass performance ever captured on tape. Pe r i o d .

As alway s , it is a session priority of the engineer (andeve ryone else invo l ved) to help ensure that the band isc o m fo r t ab l e . Th ey need to hear each other easily. I d e a l lyt h ey should also be able to see each other perfectly.S t u ck in headphones, t h ey’ll be gr u m py at firs t . C a re f u l lydial up a great sounding mix in the headphones, a dd i n gsome basic effects (reverb on the vocals at least), and tryto make the basics session a satisfying place to perfo r m .

You’ll know you got a keeper take when the drummerl i kes it. D r u m m e rs know eve ry single hit they ’ re laye r i n ginto the tune; they know the feel they are going for; theya lways know when they missed a fill; they’ll certainlyk n ow when they nailed it. When they ’ re happy, b r i n geve ryone into the control room for a loudspeaker listen.

BY ALEX CASE

PA RT 12

This ought to be a piece of cake after the challenges of last month’s live-to-two-track session. Right?

The Multitrack SessionThe Multitrack SessionThis ought to be a piece of cake after the challenges

of last month’s live-to-two-track session. Right?

RECORDING JUNE 2000

The other thing yo u ’ ve gotta wa t chis leve l s . When the drummer andbass player fall into “the zo n e ” t h eytend to play harder (i.e. louder) thanin all the previous take s . M a ke surewhen the band loves the take thatthe levels didn’t head too far into thered zone and distort. If you re c o rddistorted tra cks on the mu l t i t ra ck ,you can’t un-distort them later.

M e a n t i m e , the band and the pro-ducer have to make sure the play-b a ck through loudspeake rs is asinspiring as the live take wa s . It isn’teasy to keep the loudspeaker play-b a ck ex c i t i n g . While the take wa sbeing re c o rded we had the benefit ofwa t ching the playe rs .

The true test is the control ro o mp l ay b a ck . Do you still feel thatexcitement when you can’t see thep e r fo r m e rs? The basics session iscomplete when the perfo r m a n c epasses the loudspeaker play b a ck test.

Almost perfect? What if the takefeels right except for a few minorm i s t a kes? Many such mistakes aref i x able (see below ) , but first decideif they should be fixed at all. It ist e m p t i n g — ve ry tempting—to fixeve ry single flub so that the tra ck son tape re p resent some ideal, b e s tpossible ve rsion of the song.

Th a t ’s a fine appro a ch , and manybands are famous for their “ p e r f e c t ”re c o rd i n g s . Th ey are also famous fo rspending a lot of time (sometimesye a rs!) in the studio, not a luxury wecan all affo rd .

And sometimes such “ p e r f e c t ”t ra cks are accused of being too per-f e c t ,l a cking life, wa r m t h , or emo-t i o n . P i ck your spots care f u l ly so thatyour project falls at the appro p r i a t espot between the high audio craft ofS t e e ly Dan and low-fidelity on pur-pose of Tom Wa i t s .

Spike the punchTy p i c a l ly some fixes are called fo r

at the basics session. The first thingre p a i red is the bass tra ck . M i s t a ke sin the drum tra ck usually demandthat the entire take be re d o n e . Th e rea re ex c e p t i o n s , but try to avo i dp a t ching and piecing together adrum perfo r m a n c e .

Tra ck until you get a single, c o n s i s-t e n t ly strong drum take . Then eva l u-ate the bass part. Most like ly, w h e nthe bass player loves the take asmu ch as the drummer yo u ’ ve still gotsome minor re p a i rs to do. The bassistp ro b ab ly needs to punch-in a fewspots to fix funny notes—notes thatwe re early or late, notes that we res h a rp or flat, w rong notes, l o u dn o t e s , soft notes, or notes that justd o n ’t seem to wo rk .

The engineer has to make sure theaudio quality is top shelf. S o m e t i m e sthe band nails it on the first or sec-ond take . Th a t ’s good news mu s i c a l ly.But it can be bad news sonically.Perhaps you have n ’t had a chance tot weak: solo the snare to be sure it iscrisp or eq the floor tom so it soundsas full as yo u ’d like .

With experience you can set upm i c rophones and get levels onto tapethat are perfectly usable as of Ta keO n e . That is, while it might havebeen nice to tweak the eq and add adose of compression during basics,yo u ’ ve got a tra ck re c o rded we l lenough that such processing can bedialed in during the mix.

P u n ching-in is the process of goinginto re c o rd during a tra ck alre a dyre c o rd e d .You cue and play thet a p e / h a rd disk a few bars in front ofthe mistake to be re p l a c e d . The bassp l ayer plays along. You go intore c o rd while playing (typically yo uhold down the play button with onefinger and feather touch the re c o rdbutton) at just the right spot. Th ered lights go on, and the bass playe ris live , l aying down a new part whileyou erase the old part.

D o n ’t go out for pizza now, b e c a u s eyo u ’ ve got to punch out. Yo u ’ ve gotto get the mu l t i t ra ck out of re c o rd( t y p i c a l ly by pressing the play but-ton alone, without touching re c o rd )so that only the mistake is re p l a c e dand the rest of the prev i o u s ly re c o rd-ed tra ck is pre s e rve d .

P u n ching in and out on adigital audio wo rkstation isu s u a l ly simple. P u n ching inand out on a mu l t i t ra ckre c o rder is tricky business.You’ll acquire the skill onlyt h rough pra c t i c e . S o m e t i m e syou punch entire ve rs e s , b u tother times you might try top u n ch in and out on an indi-vidual eighth note of mu s i cor a single syllable of vo c a l .

At the basics session you have theunique pleasure of fixing the basst ra ck . Bass is pro b ab ly the most diffi-cult instrument of all to punch .When we loop a sample, we know tore a ch for ze ro crossing points on thewave form to avoid glitch e s . We aimfor the same target for punch points.I d e a l ly we go into and out of re c o rdat something ve ry close to a ze roc rossing point so that we don’t get anaudible click where the wave ab r u p t-ly transitions from old to new.

The low fre q u e n cy signal of bassp resents a ch a l l e n g e . It has long,s l ow moving waves that spend a lotof time, c o m p a red to high fre q u e n-c i e s , away from ze ro amplitude.S i m p ly put, high frequencies cro s sthe ze ro amplitude axis more oftenthan low fre q u e n c i e s . As a re s u l t , o u r

o dds of getting a click-less punchpoint are wo rse for lower fre q u e n cysounds like bass.

M a ny digital machines let yo uselect a cro s s fade time for punch e s .This adjusts ex a c t ly how abrupt thet ransition from the tra ck to thep u n ch will be. As in editing and loop-i n g , a cro s s fade gives you a brief win-d ow in time during which both oldand new tra ck are mixed tog e t h e r.The intent is to help dovetail onet ra ck into another.

Done corre c t ly, this helps make thep u n ch point less audible. But settinga cro s s fade beyond 20 millisecondsoften leads to other audible artifa c t s .The click you try to avoid is re p l a c e dby the flanging sound that comesf rom briefly hearing the two wave-forms (old and new) simu l t a n e o u s ly

during a slow cro s s fa d e .The key to successful

p u n ches is in selectingyour punch points care f u l-ly. Because we are lookingfor a high fre q u e n cymoment in the music sothat we can punch during aze ro cro s s i n g , t ry to per-form your punch duringthe high fre q u e n cy tra n-sients of the part.

RECORDING JUNE 2000

On bass, that means we hide ourp u n ch points in fret noise and pickn o i s e . P u n ch in and out at theinstant the bass player is articulatinga new note. The buzz and grit whenthe bass player digs in on a dow nbeat gives you an instant of high fre-q u e n cy activity where your punchcan hide.

Your punching tech n i q u eis enhanced by a second ele-ment to your stra t e g y :p u n ch when no one is look-i n g , e r, I mean listening.That is, select punch pointson bass that are going to bem a s ked by some other loudand distracting eve n t . Aslamming snare hit, fo rex a m p l e , will tempora r i lyfill the mix with so mu chnoise that a small error inp u n ching in or out over on the basst ra ck is cove red up.

C o nve n i e n t ly, the drummer isre q u i red by the Rock - N - R o l lD r u m m e rs’ Union to hit the snare onbeats 2 and 4, and the bass player isgoing to articulate a new note onmost of those snare hits so that manyp u n ch points become ava i l ab l e .

OverdubsAfter re c o rding the killer drum

t a ke of the century andp e r forming maybe a hand-ful of bass punch e s , yo u ’ vecompleted the basics ses-sion for the tune. Time tom ove on to phase two ofthe mu l t i t ra ck session:ove rd u b s .

During ove rdubs yo ure c o rd single instrumentsor small sections onto sepa-rate tra cks of the mu l t i-t ra ck . In this way you buildup the pop music arra n g e-ment around all the othert ra cks alre a dy re c o rd e d .

O ve rdubs are typically less stre s s-ful and less crazy than basics or livesession wo rk . You are given themental relief that comes from fo c u s-ing your energies on perhaps a sin-gle musician with a single instru-m e n t , using maybe a single micro-p h o n e . You get the chance to hearout the many subtleties of there c o rding discipline.

It is during the calm, late nightove rdub that you get to hear the dif-f e rence moving a mic one inchm a ke s , the change in sonic ch a ra c t e rthat comes when you change fro mone mic to another, the audibleeffect of re c o rding on a wood floorve rsus carp e t , and so on.

As engineer yo u ’ ve got a fair amountof freedom now. I encourage you to useove rdubs as a chance to ex p e r i m e n twith re c o rding ideas and refine yo u reve r- d eveloping re c o rding tech n i q u e .

For ex a m p l e , a stra i g h t fo r wa rdelectric guitar ove rdub can be assimple as sticking a single, b ravedynamic cardioid up close to theamp and hitting re c o rd . If the tone atthe amp is good, this re c o rding tech-nique will never fa i l .

But as the producer and the gui-tarist and the other band membersexperiment with alternative mu s i c a li d e a s , you can stick a few alternativem i c rophones up next to the trustydynamic alre a dy there to see howthe sound ch a n g e s . You can ex p e r i-ment with alternative mic place-m e n t s . Find out what a ribbon micro-phone in the corner sounds like . Tryan omnidirectional large diaphra g m

condenser over by the brick wall tosee if the sound wo rk s . Or whatev-e r — ch e ck out Bill Stunt’s ‘A n o t h e rArticle About Recording ElectricG u i t a rs?!’ 2/00 or Bob Ross’s‘Eclectic Electric Guitar’ 7/97 fo rsome other hints and cra z i n e s s .

During ove rd u b s ,d o n ’t be afra i dto throw experimental signals ove ron ex t ra tra ck s . If you like thes o u n d s , keep the alternative tra ckand be a hero. If it sounds we a k ,e rase it and ex p l o re other ideas ont o m o r row ’s ove rd u b s .

So many instruments rewa rd dis-tant miking, s t e reo miking, a n dexperimental miking tech n i q u e s .

Th e re is no obvious way tolearn all the options andk n ow which ones wo rk .A n dsession budgets can ra re lya f fo rd to let the engineerex p e r i m e n t . Do this sort ofwo rk on the side, while thesession is distracted bysomething else. O ve rd u b sonto spare tra cks re p re s e n ta terrific opportunity.

What does an ove rd u bsession re a l ly look like? Ifyou are picturing a single

m i c rophone in front of a singleinstrument in an otherwise emptyro o m , yo u ’ re missing out on a lot ofthe fun. The typical day of ove rd u b sfills the studio with as many micro-phones as a basics session. H e re ’sh ow it goes.

M aybe the day begins with an elec-tric guitar ove rd u b . The engineersets up the tried and true appro a chplus an experimental set of mics,should there be time or motiva t i o n

during the ove rdub to re a chfor another kind of tone.

When the guitar tra ck isdone and the session move son to the next ’dub (tam-bourine for ex a m p l e ) , l e avethe guitar setup as is. B r i n gout another mic for the tam-b o u r i n e . Of course there isroom to ex p e r i m e n t .C o m p a re a dynamic to ac o n d e n s e r, or an omni to ac a rd i o i d .E ven the humbletambourine tra ck we l c o m e ssome engineering ex p l o-

ra t i o n . Then move on to the nex tove rd u b , m aybe didgeridoo.

As the various ove rdubs are done,the room fills with micro p h o n e s . Th i sis handy for a couple of re a s o n s .Fi rs t , with the electric guitar ampp rev i o u s ly set up and re a dy to go theband and producer are free to ex p e r-iment fre e ly. Th ey can re a ch for theguitar with a second’s notice to tryout a new musical idea.

RECORDING JUNE 2000

A few hours into an ove rdub ses-sion you might have emptied the miccloset and used up all the mics t a n d s . The re c o rding room is re a dyfor action. Mandolin? Good idea.H ave a seat where we tra cked theacoustic guitar a few hours ago andwe’ll start from there . It’ll only takea second. Rain stick? Cool idea. Ju s tstep up to the tambourine mic.

The ove rdub session becomes ac o m fo r t able place to ex p l o re mu l t i-t ra ck re c o rding ideas, l i b e rating themu s i c i a n s , p ro d u c e r, and engineer towo rk fast and fre e ly.

A c c u mulating the various ove rd u ba r rangements within a single ro o mnot only makes getting the differe n tove rdubs done more quick ly and easi-

ly, it also leads to some fun ex p l o r i n gof engineering ideas. While the bandp l ays with different parts on thed i d g e r i d o o, you can open up differe n tmics in the room to see how it sounds.

That is, while they re c o rd into theintended set of micro p h o n e s , you canraise the fa d e rs over on the electricguitar mics in the corner, the tam-bourine mic in the center of thero o m , and the acoustic guitar micsby the stone wa l l , and so on.

With each different ove rdub yo u ’ l llearn a bit more about the re c o rd i n gc ra f t , because you’ll get to hear halfa dozen different kinds of micro-phones and mic placements all fro mthe comfort of your chair behind thec o n s o l e / DAW. You are occasionallyrewa rd e d . But separate from thosewelcome accidental discove r i e s , yo ua re giving yo u rself a chance to learnever more about the never endingp rocess of tra ck i n g .

L e aving the microphones set upafter each ove rdub fo rces you toex p l o re new re c o rding tech n i q u e s .Perhaps you always re c o rd tam-bourine with a condenser. Good call.Since tambourines are a perc u s s i o ninstrument full of transients and highf re q u e n cy energ y, it makes perfectsense to use the type of micro p h o n e

best suited to the task (see Nuts &Bolts Part 4, 1 0 / 9 9 , and maybe Parts 5and 6 as we l l , 11/99 and 12/99).

During an intense day of ove rd u b syou may find yo u rself faced with atambourine ove rd u b , and all your con-denser mics are spoken fo r.Try a mov-ing coil mic.You might be pleasantlys u rprised by how accurate that new -fangled dynamic you just bought is.

Or possibly the more colore dsound of El Cheapo Dynamic micmight give the tambourine the edgethe tune needs. If the tune is full ofhigh fre q u e n cy tra cks alre a dy (cy m-b a l s , acoustic guitars , s h a ke r, b r i g h tp a d s ,e t c . ) , the tambourine maysound better via a dynamic than ac o n d e n s e r.

Th e re ’s one other reason to fo l l owthis appro a ch to ove rd u b s . Some peo-ple think it’s cooler if they are hang-ing out in a room full of micro-p h o n e s . Th ey feel more like a powe rsession playe r, t h ey like the vibe thatcomes from filling a room with equip-m e n t , and they feel like they are get-ting their money ’s worth from thestudio if most of the mics get used.

Wh a t ever floats their boat. But Iam certain that this appro a ch toove rdubs gives the engineer a lotm o re pleasure .

In developing your familiarity withm i c rophone make s , m o d e l s , a n dplacement stra t e g i e s , t h e re is no sub-stitute for ex p e r i e n c e . The more timeyou spend in the studio the betteryou’ll get at it.

But it is the ove rdub session mostof all that lets you make progre s sh e re . E x p l o re multiple re c o rd i n gt e chniques at once through this‘ d o n ’t take it down until the end ofthe session’ appro a ch to ove rd u b s .H ave fun.

Alex Case wonders: in New Zealandand Au s t ra l i a , do they call themunderdubs? Request Nuts & Bolts top-ics via case@re c o r d i n g m a g . c o m .

Explore multiple recording

techniques at once through this

‘don’t take it down until the end of

the session’ approach to overdubs.

Excerpted from the June edition of RECORDING magazine.©2000 Music Maker Publications, Inc. Reprinted with permission.



RECORDING JULY 2000

BY ALEX CASE

PA RT 13

Part 1 of our look at having audio wait around a bit before hitting your ears

The DelayThe DelayPart 1 of our look at having audio

wait around a bit before hitting your ears

Fi rst some mu s i c . Hum or singalong with me the tune

‘ C o m fo r t ab ly Numb’ from The Wall byPink Floy d . You know the first line:

Hello (Hello...hello...hello). Is therea ny b o dy in there ?

This is a classic use of a long delay.The dre a my, d i s t u r b e d , out of mindstate of our friend Pink is enhancedby (the entire , brilliant ro ck and ro l la r ra n g e m e n t , including) this re p e a t-i n g , g e n t ly fading ech o.

H ow ’s it done? Perhaps the sim-plest way is to use a post fader auxsend from the vocal to the delay,w h i ch is returned on a separa t efader (see Fi g u re 1a). This sends thevo i c e , with all its compression ande q u a l i z a t i o n , to the delay. We havefull control of the return from thed e l ay as it is on its own fa d e r.

We l l , a l m o s t . Pa t ching it up thisway would add a delay to the entirevocal perfo r m a n c e , not just the wo rd‘ h e l l o.’ So we resort to an automatedsend as shown in Fi g u re 1b. N ow weh ave a fader and cut button dedicat-ed to the control of the send into thed e l ay, not just the re t u r n .

This echo send remains cut mostof the song. You briefly open it upfor the wo rd ‘ h e l l o,’ and pre s t o —that single wo rd starts to ech o.Put this into automation andyou’ll get the perfectly tailore dd e l ay eve ry time.

Fi g u re 1 explains what’s going onat the console. H ow do you set upthe delay unit itself? Most delay sh ave ava i l able the controls shown inFi g u re 2: input and/or output leve l ,d e l ay time, and re g e n e ration contro l .

Input/output levels are self-ex p l a n a t o ry. D e l ay time can be fixedor va r i able—using the three modula-tion controls (ra t e , depth andshape)—as we’ll see in later ex a m-p l e s . The Regeneration contro l ,sometimes called the feedback con-t ro l , sends the output of the delayb a ck into itself. That is, a delaye dsignal can be further delayed by run-ning it back through the delay again.This is how the delay is made torepeat more than once, as happensto the wo rd ‘ h e l l o.’

As we’ll see throughout this andn ext month’s article, the simple con-t rols of Fi g u re 2 empower the delayto become a fa n t a s t i c a l ly dive rse sig-nal pro c e s s o r. In the case of the longd e l ay used by Pink Floy d , we need toset the delay time to the appro p r i a t elength of time and add enough re g e n-e ration to make the echo repeat af ew times. The other controls on thed e l ay are n ’t necessary here .

How long is long?99.9% of the time these ech o e s

should be set to a time that make smusical sense. That is, d o n ’t just pick

a random delay time, dial in a mu s i-cal delay time instead. Should itrepeat with a quarter note rhy t h m ,an 8th note, a triplet,...?

One way to do this is simply by lis-t e n i n g . Ty p i c a l ly, we use the snare to‘tune’ a delay—to set a musical delayt i m e . E ven if you plan to add delayto the vo c a l , the piano, or the guitar,it is usually easiest to use the snarefor setting the delay time bothbecause it is a rhythm instrumentand because it hits so often. So mu chof pop music has a back beat—thes n a re falling re g u l a rly on beat twoand beat fo u r.

Send the snare to the delay andlisten to the ech o. Starting with along delay time of about 250 mil-l i s e c o n d s , adjust the delay time untilit falls onto a mu s i c a l ly re l eva n tb e a t . This can be mighty confusing.It may help at first to pan the snareoff to one side and the delay re t u r nto the other.

I t ’s pretty jarring to hear a delayfall at a non-musical time interva l .But when you adjust it into the timeof the mu s i c , you’ll instantly feel it.It is easiest to find a quarter noted e l ay, but with practice and concen-t ra t i o n , you can dial in triplet anddotted rhythms too.

Sometimes we c a l c u l ate a delaytime instead. H ow is this calculated?Bear with me here , as some equations

RECORDING JULY 2000

a re about to appear. If you know thetempo of the song (we’ll call it T) inbeats per minute (BPM) and you wa n tto calculate the length of a quarternote delay in milliseconds (Q), do thefo l l ow i n g :- First convert beats per minute into minutes per beat bytaking the re c i p ro c a l :

T beats per minute becomes 1/T minutes per beat.- Then convert from minutes to milliseconds:

1/T minutes per beat x 60 seconds per minute x 1,000 milliseconds per second.

- The length of time of a quarter note in milliseconds perbeat is:

Q = (60 x 1,000)/T = 60,000/TFor example, we know a song with 60 beats per minuteticks like a watch, with a quarter note occurring exactlyonce per second. Let’s try using the equation.

T = 60:Q = 60,000/T = 60,000/60

= 1,000 msec (one second) per quarter noteDouble the tempo to 120 bpm.

T = 120:Q = 60,000/T = 60,000/120

= 500 msec (half a second) per quarter noteI use milliseconds because that is

the measurement most delay unitsex p e c t . K n owing the quarter noted e l ay makes it easy to then calculatethe time value of an 8th note, a 16thn o t e , dotted or triplet va l u e s , e t c .

Some newer delay dev i c e s , l i ke theTC D-Two rev i ewed elsew h e re in thisi s s u e , let you display delay times ineither milliseconds or bpm dire c t ly,but remember that you need to knowb e fo re you look at a bpm value if thed e l ay is calculating a quarter note orsome other length.

In the ‘ C o m fo r t ab ly Numb’ ex a m-ple ab ove , t h ey cleve rly use a dotted8th note delay. It is worth tra n s c r i b-ing it for some production insight.The tune is dre a my and lazy int e m p o, m oving at about 64 bpm. Th et wo syllables of ‘hello’ are sung as16th notes.

To appreciate the perfection inPink Floy d ’s dotted 8th note delayt i m e , l e t ’s consider two other, m o reo bvious choices: a quarter note delayor an 8th note delay (consult Table 1).

The quarter note delay stro n g lye m p h a s i zes the time of the song; it’so rd e rly and pers i s t e n t . Sing it toyo u rself as a quarter note delay :Hello x x hello x x hello. . . . Th i swould make it seem like Pink isbeing nagged or pushed aro u n d .S i n gthe 8th note delay and you find therepeats fall one after the other, w i t hno rest in between the wo rds: hellohello hello hello hello. This is justplain annoy i n g .

The delay time they chose has theeffect of inserting a 16th note rest inb e t ween each repeat of the wo rd .Hello is sung on the dow n b e a t .

The echo never again falls on ad own beat. Fi rst it anticipates beatt wo by a 16th note, then it falls onthe “ a n d ” of beat two. It then falls a16th after beat thre e . Fi n a l ly, it dis-a p p e a rs as the next line is sung.

This timing scheme determinesthat ‘hello’ won’t fall on a beat againuntil beat four, by which time thenext line has begun and ‘hello’ is nolonger audibly repeating. It’s reallya pattern of three in a song built onfour. This guarantees it a dreamy,disorienting feeling. It remains trueto the overall ‘numb’ feeling of ourhero Pink, keeping an uncertain, dis-connected feel to the story told.

The result is a pre-calculated cre-ation of the desired emotional effect.And it’s a catchy hook—a real PinkF l oyd signature . ( Th ey ’ ve used thist r i ck befo re , to devastating effect:‘Us And Them’ from Dark Side Of Th eM o o n , ‘ D ogs’ fro m A n i m a l s. . . . . )

Figure 1a) Constant Send - Use an Aux Send (literally an Echo Send) or Spare Track Bus to send the Vocal to the Delay

Table 1: evaluating the musical timing of delays

1 e & a 2 e & a 3 e & a the beat

Hel lo sung word

x x x x hel - lo quarter note delay

x x hel - lo 8th note delay

x x x hel - lo dotted 8th note delay

x x x hel - lo x hel - lo x hel - lo with regeneration

Hel lo x hel - lo x hel - lo x hel - lo net effect

RECORDING JULY 2000

Long delayI t ’s a funny idea, a dding an echo to

a singer, p i a n o, g u i t a r, or whateve r. I td o e s n ’t seem to have any motiva t i o nbased on re a l i t y. The only way tohear an echo on the vocal of a songis to go to a terrible ve nue (like anice hockey rink or the Gra n dC a nyon) and listen to mu s i c .

The sound of an echo across thee n t i re mix is in fact not a pleasantex p e r i e n c e . It is messy and distinctlyn o n - mu s i c a l .

The echoes we find in pop mu s i ctend to be used with more re s t ra i n t .In some cases the echo is added to asingle tra ck , not the whole mix. A n di t ’s mixed in fa i n t ly so as to bealmost inaudible.

In other cases the delay is add e do n ly to key wo rd s , p h ra s e s , or lick s .

SupportIf a constant echo is to be added to

an entire tra ck , the echo needs to bemixed in almost subliminally, n e a rlyh i dden by the other sounds in them i x . A soft echo underneath the leadvocal can give it added richness ands u p p o r t . This appro a ch can stre n g t h-en your singer, e s p e c i a l ly when them e l o dy heads into falsetto territory.P u l s i n g , subliminal echoes feeding along reverb can create a soft and del-icate sonic foundation under thevocal of a ballad.

Then there ’s the vulnerable ro ckand roll singer in front of his mate’sM a rshall stack . After the last ch o r u sthe singer natura l ly wishes to scre a m“ Ye a a a a a a a a a h ! ” and hold it for acouple of bars . It isn’t easy to ove r-come the guitarist’s wall of sound.Help the singer out by pumping somein-tempo delays into the scre a m .

The best “ Ye a a a a a a a a a h ! ” eve rre c o rded in the history of ro ck androll (and I have this from a re l i ab l es o u rce) is Roger Daltrey ’s in “ Wo n ’tGet Fooled A g a i n ” by The Wh o. Th es c ream occurs right after the re -i n t roduction when those cool key-b o a rds come back in, and rightb e fo re the line “Meet the new boss.The same as the old boss.”

This scream is a real ro ck and ro l lc l a s s i c . Listen care f u l ly (especially atthe end of the scream) and yo u ’ l lhear a set of delayed screams under-n e a t h . I t ’s Roger Daltrey, o n ly more

so: it’s half a dozen Roger Daltrey s .This makes quite a statement.

You can do this too. All you need isR oger and a long delay with somere g e n e ra t i o n .

SlapA staple of ’50s ro ck is sometimes

part of a contempora ry mix: slap-b a ck ech o. You never heard Elviswithout it. Solo John Lennon there-fo re often had it. And guitarists play-ing the blues tend to like it.

Start with a single audible ech os o m ew h e re between 90 ms and 200m s . On a vocal you’ll instantly add adistinct re t ro feeling to the sound. O nguitar it starts to feel more live , l i keyou are in the smoky bar yo u rs e l f.

B e fo re the days of digital audio acommon appro a ch to creating this

sort of effect was to use a spare ana-l og tape machine as a generator ofd e l ay. During mixdown the mach i n eremains in re c o rd . Signal is sentf rom the console to the input of thetape machine in ex a c t ly the sameway yo u ’d send signal to any othereffects unit: using an echo send ors p a re tra ck bus.

That signal is re c o rded at the tapem a ch i n e , and milliseconds later it isp l ayed back . That is, though the tapem a chine is re c o rd i n g , it remains inrepro mode so that the output of thetape machine is what it sees at thep l ay b a ck head.

As Fi g u re 3 show s , the signalgoes in, gets printed onto tape, t h etape makes its way from the re c o rdhead to the play b a ck head (takingtime to do so), and finally the sig-

Figure 1b) Automated Send - Uses two different Aux Sends or Track Buses. The SendChannel is a Console Strip dedicated to controlling exactly what is sent to the Delay.

RECORDING JULY 2000

nal is played back off tape andreturned to the console. The re s u l tis a tape delay.

The signal is delayed by theamount of time it takes the tape tot ravel from the re c o rd head to there p ro head. The actual delay timethen is a function of the speed of thetape and the particular model oftape machine you are using (whichdetermines the physical distanceb e t ween the two heads).

Want to lengthen the delay time?S l ow the tape machine dow n . Yo umight have two, m aybe three ch o i c e sof tape speed: 7 1/2, 1 5 , or 30 inch e sper second (lov i n g ly called ‘ i p s ’ ) .None of these delay times seemex a c t ly right? Maybe your tapem a chine has vari-speed that lets yo ufind tape speeds slightly faster ors l ower than the standard speeds list-ed ab ove .

C a n ’t make these delay times fitinto the rhythm of the song? Nop ro b . Go buy another analog 2-tra ckm a ch i n e , one with a different heada r rangement so that the delay timewill be differe n t .

Wait a second. A single tapem a ch i n e , w h i ch might cost seve ra lthousand dollars , is capable of just af ew different delay settings? Yu p. A3-speed tape machine used this wayis like a re a l ly ex p e n s ive effectsd evice with three pre s e t s . D ra g .

Tape delay was originally usedbecause it was one of the only ch o i c-

es at the time. To help out, m a nu fa c-t u re rs made tape delay s , w h i ch we retape machines with a loop of tapei n s i d e . The spacing between there c o rd and play b a ck heads wa sa d j u s t able to give you more flex i b i l i-ty in timing the delay.

Here in the year 2000 we havemore options. Life is good. We canbuy a digital delay that is easilyadjustable, wonderfully flexible,probably cheaper than a tapemachine, and it either fits in one ortwo rack spaces or exists convenient-ly in a pull-down menu on our DAW.

But if you have a spare tapem a chine that has perhaps been sit-ting unused ever since you made thei nvestment in a DAT mach i n e , yo u ’ vegot the opportunity to create tapes l a p. This can even be a cassetted e ck if it has a tape/monitor sw i t chto let you monitor the play b a ck headwhile you re c o rd .

Why bother? Some people are sim-p ly turned on by anything re t ro. Ta p ed e l ay is more tro u b l e , m o re ex p e n-s ive , and we know some great oldre c o rds used it. That is re a s o nenough for some engineers . I pers o n-a l ly am not into re t ro for re t ro ’s sake ;I take the trouble to use a tape delaywhen I want that ‘ s o u n d .’

An analog tape machine intro-duces its own subtle color to thes o u n d . M a i n ly, it tends to add a lowf re q u e n cy hump into the signal,depending on the tape mach i n e , t h e

tape speed, and how it is calibra t e d .If you push the level to the taped e l ay into the re d , you intro d u c ethat signature analog tape compre s-s i o n , and at hotter levels still, a n a l ogtape saturation distortion.

Tape delay becomes a more com-p l ex , ve ry rich effect now. It isn’t justa delay—it is a delay plus equalize rplus compressor plus distortiond ev i c e .

This can be darn difficult to simu-late digitally. I t ’s sometimes the per-fect bit of nuance to make a tra ckspecial within the mix.

EmphasisA dding a long delay to a key wo rd ,

as in the Pink Floyd ex a m p l e , is away to emphasize a particular wo rd .It can be obv i o u s , l i ke the ‘hello’ thatbegins the song. S i mulating a calland response type of ly r i c , the delayis often a hook that people singalong with.

A l t e r n a t ive ly, it can be more sub-t l e . A set of emphasizing delays hitskey wo rds throughout “ S y n ch ro n i c i t yI I ” on The Po l i c e ’s final albumS y n ch r o n i c i t y. The first line of eve rychorus ends with the wo rd ‘ away,’w h i ch get a little delay based boost.

Listen also to key end words in theverses: ‘face,’ ‘race,’ and, um, ‘crotch.’These are a quick dose of severalechoes, courtesy of the regenerationcontrol.

The Wa l l f l owe rs’ “One Headlight”on Bringing Down the Horseo f f e rs agreat example of re a l ly hiding thed e l ay s . Listen care f u l ly to the thirdve rse—the wo rds ‘turn’ and ‘ b u r n ’e a ch get a single subliminal dottedquarter note delay.

I t ’s not unusual to low pass filterthese sorts of delay s . R e m oving thehigh fre q u e n cy content from eachrepeat makes it sink deeper into them i x . Nice delay units provide yo uwith this filter as an option.

M o re ove r, t h e re is often the ab i l i-ty to double the delay time on out-b o a rd digital delays by pressing abutton labeled “ X 2 ,” m e a n i n g‘times two.’ This cuts the samplingrate in half. With half as many sam-ples to keep tra ck of, the amount oftime stored in a fixed amount ofm e m o ry effective ly doubles, h e n c ethe ‘times two’ lab e l .

Figure 2) Signal Flow Through a Typical Delay Unit

RECORDING JULY 2000

Halving the sample rate also low-e rs the upper fre q u e n cy capability ofthe digital dev i c e . You know this ifyou are fo l l owing the sampling ra t ewa rs: 44.1 kHz, 48 kHz, 96 kHz andm o re . The key benefit of incre a s e dsampling rate is improved high fre-q u e n cy resolution (A ‘Nuts & Bolts’column dedicated to digital audio isfo r t h c o m i n g ) .

While sampling rates are cre e p i n gup on all our digital toys (especiallyDAWs and mu l t i t ra ck re c o rd e rs ) , wesometimes lower the sampling ra t eon our digital delay s . L ow pass fil-tering the delay is often a desirab l emix move .

GrooveB eyond support, s l a p, and empha-

s i s , we sometimes re a ch for delay tofill in part of the rhythm tra ck of as o n g . Reggae is famous for its clich ée ch o. Drum progra m m e rs have beenk n own to put in an 8th or quarternote delay across the entire gro ove .

Guitarists use delay too. U 2 ’s Th eEdge has made delay a permanentpart of his guitar rig. A classicexample is apparent from the ve ry

i n t roduction of U2’s “ Wide Awa ke ”on The Unforgettable Fi re. The quar-ter note triplet delay isn’t just ane f f e c t , i t ’s part of the riff.

The Edge has composed the delayelement into the song. Ditto for “ I nthe Name of Love ” f rom the samea l b u m . An echo isn’t just an echo anym o re , i t ’s a part of the tune.

Make it shortThe delays discussed ab ove are

all audible as ech o e s , repeats of ane a rlier musical phra s e . D e l ays aresometimes so short that they are n ’tp e rc e ived as ech o e s . That is, as thed e l ay time falls below about 50m i l l i s e c o n d s , the sound of thed e l ay is no longer an ech o. We stillhear the delay, but it takes on an ew persona as the delay time getsthis short.

N ext month we ex p l o re theseshorter time effects.

Alex Case wonders: why are flightd e l ays always long delays? RequestNuts & Bolts topics via case@re c o r d-i n g m a g . c o m .

Figure 3) Tape Delay - The Tape Machine is always rolling, In Record. The distance between the Record Head and the Playback Head as well

as the selected Tape Speed determine the Delay Time.

Excerpted from the July edition of RECORDING magazine.©2000 Music Maker Publications, Inc. Reprinted with permission.




N ow start shortening the delay. 100 ms, 80 ms, 6 0 m s , 2 0m s , 10 ms, 5 ms, d own to 3 ms and below. Listen care f u l lyas you do this. What the heck is going on?

The long delay is just an ech o. The ve ry short delay s(15 ms and lower) sound stra n g e , sometimes hollow,sometimes boomy. At one short delay setting there ’sex t ra low end, then at a slightly different delay time, al a ck of low end. This mix of a bass sound with a ve ryshort delay sounds like it’s been equalize d .

G ra d u a l ly lengthen the delay time and listen for thepoint at which it starts to sound like a distinct echo again.Depending on the bass sound, you may hear the delays e p a rate from the bass into an echo somew h e re betwe e nabout 60 and 80 milliseconds. In between the ve ry longand the ve ry short delay times, we l l , i t ’s hard to describe.

N ext try a snare sound. Again start with a long delayand gra d u a l ly pull it down to a short delay. Again we findit is a distinct echo at long settings. The delay intro d u c e sa strange timbral change at short delays and somethingtough to describe as it transitions between the two. Wh i l ewe ’ re here , do the same experiment with an acoustic orelectric guitar tra ck , or a string patch on the sampler.

Welcome to the real wo rld of delay s . Th ey are n ’t justfor echoes any m o re . When delays become shorter thanabout 50 or 60 milliseconds (depending on the type ofsound you are listening to, as demonstrated ab ove) theya re no longer repeats or echoes of the sound. The samed evice that delays a signal starts to change the color, t h es p e c t ral content of the signal.

L e t ’s ch e ck out how it wo rk s .

Sine of the timesConsider first a pure tone (no fun to listen to, but help-

ful to study ) . Mixing together—at the same volume andpan position—the original signal with a delayed ve rs i o nof itself might have results like the two special cases

BY ALEX CASE

The Short Delay, Pa rt I:The Short Delay, Pa rt I:Untangling the Comb Filter

The humble delay is a powerful production tool. Yo us e e , not all delays sound alike . Long delays sound

ve ry different from short delay s .“No duh, C a s e ,” you think to yo u rs e l f. Let me ex p l a i n .

The sonic difference between a long delay and a shortd e l ay isn’t just the apparent length of the delay.

Long delays are pretty intuitive; they sound like ane ch o, perhaps repeating a few times. Short delay s , on theother hand, a re n ’t heard as ech o e s .Ve ry short delay sh ave an important spectral effect on the sound. Th e nt h e re are the delay times in between long and short.Th ey have a more complex , t ex t u red effect.

So we classify delays into three broad categories, c l ev-e rly called long (greater than about 50 ms), m e d i u m( b e t ween about 50 ms and 20 ms), and short (less thanabout 20 ms). We cove red long delays in last month’s col-umn; medium and short delays are so darn cool that we ’ l ldedicate this and next months’ columns to them.

Make it shortAs delay times fall below about 50 milliseconds, t h ey

t a ke on a new pers o n a . If you are actually reading this inyour studio, t ry the fo l l owing ex p e r i m e n t . ( Those of yo ureading this on an airplane or tour bus are out of luck .That’ll teach you: never leave your studio, eve r. )

Pa t ch up a sampler loaded with a variety of sounds orfind a mu l t i t ra ck tape with a good variety of tra ck s . O nyour mixer/DAW, set up a delay fed by an aux send thatreturns to your monitor mix at about the same volume asthe synth or original tra ck s . Pan both the source audioand the return from the delay dead center.

Listen care f u l ly to the mix of each source soundwhen combined with the output of the delay. Start witha bass line. C h e ck out the combination of the bass witha long delay, m aybe 200 milliseconds. Yu k . I t ’s a blurry,chaotic mess.

Excerpted from the August edition of RECORDING magazine. ©2000 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301Tel: (303) 516-9118 Fax: (303) 516-9119 For Subscription Information, call: 1-800-582-8326


s h own in Fi g u re 1. ( Just look at thesolid lines for now; we’ll come backto the dashed lines and what theymean in a minu t e . )

If the delay time happens to beex a c t ly the same as the period ofthe sine wave , we have the con-s t r u c t ive interference shown inFi g u re 1a. That is, if the delay timewe set up on our delay processor isex a c t ly equal to the time it take sthe sinusoid to go through onecy c l e , then they combine coopera-t ive ly, and the result is a signal ofthe same fre q u e n cy but with twicethe amplitude.

The situation in Fi g u re 1b re p re-sents another special case. If thed e l ay time happens to be equal tohalf a period (half the time it take sthe sine wave to complete ex a c t ly onecy c l e ) , then the original sound andthe delayed sound move in oppositionto each other—they are 180 degre e sout of phase. The combination re s u l t sin ze ro amplitude—pure silence.

If you have access to a sine waveoscillator (either as test equipmentor within your synthesize rs or com-p u t e r ) , g ive it a try. I recommend 500Hz as a starting point—it isn’t quiteas piercing as the standard test tone

of 1000 Hz, and the math is easy. Th etime it takes a pure 500 Hz tone tocomplete one cycle is 2 milliseconds( Period = 1 / Fre q u e n cy = 1 / 500 =0.002 seconds = 2 milliseconds).

So mixing together equal amountsof the original sine wave and a 2millisecond delayed version will cre-ate the case shown in Figure 1a. Setthe delay to 1 millisecond, creatingthe situation of Figure 1b, and you’llfind that the sine wave is essentiallycancelled.

N ow look at the dashed-line waveforms on Fi g u re 1. Th ey show thatthese doublings and cancellationshappen at certain other higher fre-quencies as we l l . For any given delayt i m e , certain frequencies line up justright for perfect constructive ord e s t r u c t ive interfere n c e .

The math wo rks out as fo l l ow s . Fo ra given delay time (t ex p ressed ins e c o n d s , not milliseconds) the fre-quencies that double are describedby an infinite series: 1/t, 2 / t , 3 / t , . . . .The frequencies that cancel are: 1/2t,3 / 2 t , 5 / 2 t , . . . .

Using these equations we confirmthat a 1 millisecond delay (t = 0.001seconds) has peaks at 1000 Hz, 2 0 0 0H z , 3000 Hz, . . . and nulls at 500 Hz,1500 Hz, 2500 Hz, . . . . This is consis-tent with our observations in Fi g u re1b of how a 1 millisecond delay can-cels a 500 Hz sine wave .

In Fi g u re 1a, the dashed line is the2/t (constructive) case, and in 1b, t h edashed line is the 3/2t (destructive )c a s e .A g a i n , you can see how thepeaks and dips in the waves eithera dd up or cancel out.

A 2 millisecond delay has ampli-tude peaks at 500 Hz, 1000 Hz, 1 5 0 0H z , . . . and nulls at 250 Hz, 750 Hz,1250 Hz, . . . . We looked at the re s u l t sof this 2 ms delay for the single fre-q u e n cy of 500 Hz in Fi g u re 1a. Th emath reveals that the peaks and dipshappen at seve ral fre q u e n c i e s , n o tjust one. Of cours e , the only re l eva n tpeaks and va l l eys are those that fa l lwithin the audible spectrum fro mabout 20 Hz to 20,000 Hz.

To ex p l o re this further, return toyour mixer setup combining a sinewave with a delayed ve rsion of itselfset to the same amplitude. S we e pthe sine wave fre q u e n cy higher andl owe r, wa t ch your meters , and listenc a re f u l ly. With the delay fixed to 1m i l l i s e c o n d , for ex a m p l e , sweep thef re q u e n cy of the sine wave up slow lybeginning with about 250 Hz.

You’ll hear the combination of thed e l ayed and undelayed waves disap-pear at 500 Hz, re a ch a peak at 1000H z , disappear again at 1500 Hz,re a ch a peak again at 2000 Hz, a n dFigure 1: Combining Sine Waves


so on. We ’ ve got a d e l ay (not ane q u a l i zer) changing the fre q u e n cycontent of our signals. We ’ ve got ad e l ay (not a fader or a compre s s o r )changing the loudness of our mix.

L e t ’s ride the fa d e rs in the fo l l ow-ing ex p e r i m e n t . On your mixer, o n efader has the original sine wave at500 Hz panned to center. And the sinewave is also sent to a delay unit set toa delay time of 1 millisecond. A n o t h e rfader controls the return from thisd e l ay, also panned to center.

Start with both fa d e rs dow n .R a i s ethe fader of the source signal to a re a-s o n able leve l .N ow raise the secondfa d e r. As you make the delayed signall o u d e r, your mix of the two waves getsq u i e t e r. As you add more of thed e l ayed sine wave , you get morea t t e nuation of the original sine wave .

This is the phenomenon shown inFi g u re 1b. And the mix re a ches itsm i n i mum level when the two signalsa re at equal amplitude.

Time for musicStupid parlor trick or va l u ab l e

music production tool? To answer thisquestion we have to get rid of thep u re tone (which pretty mu ch neve rhappens in pop music) and hook upan electric guitar (which pretty mu cha lways happens in pop mu s i c ) .

Run a guitar signal—live , f ro myour sampler, or off tape—thro u g hthe same setup ab ove . With thed e l ayed and undelayed signals set tothe same amplitude, listen to whath a p p e n s .

Can you find a delay time settingthat will enable you to completelycancel the guitar sound? Nope. Th eguitar isn’t a pure tone (thank God).It is a complex signal, r i ch withsound energy at a range of fre q u e n-c i e s . No single delay time can cancelout all the frequencies at once.

But mixing together the guitarsound with a 1 millisecond delaye dve rsion of the guitar sound definitelydoes do something, and what hap-pens is definitely cool. It would benice to understand what is going on.

We alre a dy saw a 1 millisecondd e l ay re m ove the 500 Hz sine wavee n t i re ly. In fa c t , it will do the samething with guitar (or piano, o rd i d g e r i d o o, or any t h i n g ) . M u s i c a linstruments containing a 500 Hz com-ponent within their ove rall sound willbe affected by the short 1 millisecondd e l ay; the 500 Hz portion of theirsound can in fact be cancelled. Wh a tremains is the tone of the instrumentwithout any sound at 500 Hz.

But wa i t ,t h e re ’s more . Try the 2 mil-lisecond delay. In the case of the 500Hz sine wave , we saw that the signal

got louder when we added this delay.In the case of the guitar, the 500 Hzportion of the signal gets louder.

Taking a complex sound like gui-t a r, w h i ch has sound energy at a va s trange of different fre q u e n c i e s , a n dmixing in a delayed ve rsion of itselfat the same amplitude will cut cer-tain frequencies and boost others .This is called comb filtering (seeFi g u re 2) because the alteration inthe fre q u e n cy content of the signallooks like teeth on a comb.

Combining a musical wave fo r mwith a short delayed ve rsion of itselfra d i c a l ly modifies the fre q u e n cy con-tent of the signal. Some fre q u e n c i e sa re cancelled, o t h e rs are doubled.The intermediate frequencies ex p e r i-ence something in between outrightcancellation and full-on doubling.

So short delays are less like ech o e sand more like equalize rs; they aretoo short to be perc e ived as ech o e s .In fact they are so short that theystart to interact with discreet compo-

Figure 2a

Figure 2b


nents of the ove rall sound, a dd i n gsome degree of constructive (i.e.a dd i t ive) or destructive (i.e. s u b t ra c-t ive) interference to different fre-quencies within the ove rall sound.

Fi g u re 1 demonstrates this for asine wave . Fi g u re 2 summarizes whathappens in the case of a complexwave like guitar, p i a n o, s a x o p h o n e ,vo c a l , e t c .

The spectral result is that the com-bining of a signal with a delayed ve r-sion of itself acts like a ra d i c a lequalization move: a boost here , acut there , another boost here , a n o t h-er cut there , and so on. In theory yo ucould simulate comb filtering withan equalize r, dialing in care f u l ly thea p p ropriate boosts and cuts.

Th a t ’s the theory. In fa c t , one ra re lyc o u l d . To fully imitate the comb filtereffect that a 1 millisecond delay cre-a t e s , yo u ’d need an equalizer withabout 40 bands of eq (20 cuts and 20boosts within the audible spectrum).I ’ ve never seen so crazy an equalize r(other than in softwa re ) .

In fa c t , part of the point of usingshort delays in your mix is to cre a t esounds that you can’t create with ane q u a l i ze r. I t ’s pretty impre s s ive . Asingle short delay creates a wildlyc o m p l ex eq contour.

Short delays offer a ve ry intere s t-ing ex t ra detail: they create mathe-matical—not necessarily mu s i c a l —changes to the sound.

S t u dy Fi g u re 2, comparing part 2ato part 2b. Th ey show the same info r-m a t i o n . But Fi g u re 2a presents thei n formation with a logarithmic fre-q u e n cy axis. This is the typical wayof viewing mu s i c , because it’s howour ears hear: double the fre q u e n cy,go up an octave . Double it again, g oup another octave , and so on. Th i srelationship is why, for ex a m p l e , yo ugo up a half step with each fret on aguitar but the frets get closer tog e t h-er as you go up the neck .

But if you look at comb filteringwith a linear (and non-musical) fre-q u e n cy axis, you see that the peaksand dips in the filter are spaced per-f e c t ly eve n ly. It isn’t until you viewthe implications of the short delay inthis linear way (Fi g u re 2b) that yo usee why it is in fact called a comb fil-t e r. You’ll get a better hairdo usingthe comb in Fi g u re 2b instead of 2a.

This highlights another unique fea-t u re of using short delays to shapethe harmonic content of a sound. Th edistribution of the cuts and boosts isa mathematical peculiarity, all equal-ly spaced in terms of the linear nu m-ber of Hz. It is distinctly non-mu s i c a l .

Pa t ch up the comb filter with aspecial radical effect in mind. If yo uwant more careful tailoring of sound,use an equalizer with its log a r i t h m i c ,m o re musical contro l s .

Time for reflectionI t ’s still fair to ask: why all this

talk about short delays and theireffect on a signal? After all, h owoften do we use delays in this way ?

It is essential to understand thesonic implications of these shortd e l ays because all too often theys i m p ly can’t be avo i d e d .

Consider re c o rding an electric gui-t a r. With the amp in the middle ofthe room on a beautiful wo o d e nf l o o r, we place a sweet tube micro-phone a few feet away and try toc a p t u re the natural sound of theamp in the ro o m . This is a gooda p p ro a ch , s h own in Fi g u re 3a.

P roblem is, the sound reflected offthe floor and into the microphone willa r r ive a split second later than thesound that went straight from amp tom i ke .The path is longer via thereflected path, i n t roducing some delay.

The result is some amount of combf i l t e r i n g . R e c o rding a sound and asingle reflection of that sound is alot like mixing a tra ck or samplewith a delayed ve rsion of itself, as inour discussions ab ove . Comb filter-ing is a part of eve ry d ay re c o rd i n g .

Fo r t u n a t e ly the sound reflected offthe floor will also be a little quieter,reducing the comb filter effect. I fthe floor is carp e t e d , the comb filter-ing is a little less pro n o u n c e d . P l a c eab s o rption at the point of the re f l e c-t i o n , and the comb filtering is eve nless audible.

An important part of the re c o rd i n gc raft is learning to minimize theaudible magnitude of these re f l e c-tions by taking advantage of ro o macoustics in placing musical instru-ments in the studio and stra t e g i c a l lyplacing ab s o rp t ive materials aro u n dthe musical sourc e . This is onea p p ro a ch to capturing a nice soundat the micro p h o n e .

Better ye t , learn to use thesereflections and the comb filteringt h ey introduce on purp o s e . For ex a m-p l e , raising the microphone willm a ke the difference in lengthb e t ween the reflected path and thed i rect path even longer. Raising them i c rophone there fo re lengthens theacoustic delay time differe n c eb e t ween the direct sound and thereflected sound, t h e reby ch a n g i n gthe spectral locations of the peaksand va l l eys of the comb filter effect.

Of cours e , raising the micro p h o n ealso pushes the microphone furtheroff-axis to the amp, changing the tim-b re of the electric guitar tone asp i cked up by the micro p h o n e .You canraise the amp up off the floor, p e r-haps setting it on a piano bench . Yo ucan tilt the amp back so that it fa c e sup towa rd the raised micro p h o n e .

But then again, you can flop theamp on its belly, facing stra i g h td own into the floor if that soundsg o o d .A lways do what sounds good.D e l ay-induced comb filtering is onlypart of the equation.

Another common appro a ch tore c o rding a guitar amp (and pre t t ymu ch any other instrument) is to usea combination of two or more micro-phones to create the sound as yo ure c o rd it onto a single tra ck .

Figure 3a: Reflections Cause Comb Filtering


Consider the session shown inFi g u re 3b: two micro p h o n e s , o n et ra ck . H e re we have a close micro-phone (pro b ab ly a dynamic) gettingthe in yer face gritty tone of the ampand a distant microphone capturingsome of the liveness and ambience ofthe ro o m .You might label the fa d e rc o n t rolling the close micro p h o n esomething like “ c l o s e ” and the fa d e rg overning the more distant mic some-thing like “ ro o m .” You adjust the twofa d e rs to get the right mix of closeand room sounds and print that to asingle tra ck of the mu l t i t ra ck .

Th a t ’s only half the story. As yo uadjust the fa d e rs controlling theset wo micro p h o n e s , you not onlychange the close/ambient mix, yo ualso control the amount of comb fil-tering introduced into the guitart o n e . These two mics pick up ve rysimilar signals, but at differe n tt i m e s . In other wo rd s , t h ey act ve rymu ch like the signal plus delay sce-nario we ’ ve been discussing.

M oving the distant microphone to as l i g h t ly different location is just likechanging thetime settingon the delay

u n i t . It effective ly selects differe n tkey frequencies for cutting and boost-ing using the exact same principleswe ex p l o red in Fi g u res 1 and 2.

Sound travels a little farther thana foot per millisecond. To lengthenthe delay time difference by about am i l l i s e c o n d , m ove the distant micb a ck about a fo o t . To get a ten mil-lisecond delay incre a s e , m ove thedistant mic back about ten feet. I t ’sthat simple.

N a t u ra l ly, t h e re ’s too mu ch tokeep tra ck of. E a ch of these micro-phones re c e ives reflected soundsf rom the floor, the ceiling, and allthe other room boundaries—all ina ddition to the obvious direct soundf rom the amp.

So we get a complex interaction ofthe many components of guitar

sound radiating out of the amp. Th ed i rect sounds into multiple micro-phones arrive at different times,leading to some amount of comb fil-t e r i n g . The reflections from the va r i-ous room boundaries into eachm i c rophone arrive at a later timethan the direct sound, a dding stillm o re comb filtering.

Th e re is an infinite number of va r i-ables in re c o rd i n g . In theory, were c o rding engineers like this com-p l ex i t y. ( For certainty, become a taxa c c o u n t a n t . For endless opportuni-ties of ex p l o ra t i o n , become a re c o rd-ing musician.) Understanding combfiltering is part of how we master thevast re c o rding pro c e s s .

The myth of the sweet spotPerhaps you want a tough, h e av y,

l a rger than large guitar tone. M ay b ea comb filter derived hump at 80 Hzis the ticke t . Or should it be 60 Hz?

You decide. E x p l o re this issue bym oving the microphones aro u n d .P l a c et wo microphones on the amp as show nin Fi g u re 3b. Keep the close mic fixedand move the distant one slow ly.Yo u rgoal is to introduce a fre q u e n cy peakat some powerful low fre q u e n cy.

If you have the luxury of an assis-tant engineer, h ave him or her slow lym ove the microphone around whileyou listen to the combined close/dis-tant microphone mix. If you lack ana s s i s t a n t , re c o rd a take onto tapewhile you slow ly move the micro-p h o n e , as quietly as you can.

When those comb filter peaks andnotches fall into frequency rangesthat complement the tone screamingout of the guitar amp, you’ll havefound a sweet spot. No dumb luck.No magic. Finding the mic place-ment that captures the tone thatpleases the guitarist simply requiresa bit of patience—and an under-standing of the spectral implicationsof short delays.

The art of microphone placementre q u i res mastery of room acoustics,

musical acoustics, and psych o -a c o u s t i c s . To ach i eve pre d i c t ab lygood sounding results you needre c o rding ex p e r i e n c e , an unders t a n d-ing of microphone tech n o l og i e s ,k n owledge of microphone soundq u a l i t i e s , ex p o s u re to the va r i o u ss t e reo miking tech n i q u e s , and manyother topics.

In other wo rd s , you need a sub-scription to R e c o r d i n g .And an essen-tial tool in mic placement is the useof comb filtering for fun and pro f i t .Avoid it as necessary. Or use it onp u rpose when you can.

Electric guitar, w h i ch my mom andsome scientists would classify asb roadband noise, responds well tocomb filtering. With energy across arange of fre q u e n c i e s , the peaks anddips of comb filtering offer a dis-t i n c t , audible sound property to bem a n i p u l a t e d .

Other instruments rewa rd thiskind of ex p e r i m e n t i n g . Try placing asecond (or third , or fourth…) micro-phone on acoustic guitar, p i a n o, a ny-t h i n g . Experiment with the comb fil-t e r- d e r ived signal processing to get asound that is natural—or wa cky.

One day you may find yourself ina predicament: the amp sounds phatout in the live room, but thin in thecontrol room. Perhaps the problemis that, courtesy of the short delaybetween two microphones, you’vegot a big dip in frequency right at akey low frequency region. Undo theproblem by changing the spectrallocation of the frequency notch:move a microphone, which changesthe delay, which changes the fre-quencies being cancelled.

E ve ry time you re c o rd with morethan a single micro p h o n e , m a ke itpart of your routine to listen for thecomb filter effect. C h e ck out eachmic alone. Then combine them, l o o k-ing for critical changes in the timbre .

What fre q u e n cy ranges disappear?What fre q u e n cy ranges get louder?The hope is to find a way to get ridof unwanted or less interesting partsof the sound while emphasizing them o re unique and more appealingcomponents of the sound.

And make short delays part ofyour mixing bag of trick s . For subtletone shaping or a radical speciale f f e c t , the short delay is a powe r f u lsignal pro c e s s o r. Mastering it willlead dire c t ly to better soundingre c o rd i n g s .

Alex Case knows the differe n c ebetween a comb filter and an oil filter.

Request Nuts & Bolts topics viac a s e @ re c o r d i n g m a g . c o m .

Figure 3b: Multiple Microphones Lead To Comb Filtering


Fi g u re 1 re i t e rates the controls on a standard digitald e l ay device (we looked at this some in our two earl i e rcolumns on delay ) . This month we focus on the modula-tion section of this delay unit. These controls let uschange the delay time in contro l l e d , c l ever way s .

Fi g u re 2a describes a fixed delay time of 100 millisec-o n d s . I t ’s a slap echo as discussed last month. The delayunit takes whatever signal you send it, holds it for thed e l ay time you set (100 ms), then sends it out. Th a t ’s it.Th roughout the song, all session long, the delay timeremains ex a c t ly 100 milliseconds; all signals sent to it—

g u i t a r, vo c a l , or didgere e d o o — experience theexact same amount of delay.

Th a t ’s a delay without modulation.Some great effects begin when you start

using some of the delay modulation controls.Usually three basic controls are found: Rate,Depth, and Shape.

R ate c o n t rols how quick ly the delay time isch a n g e d . Fi g u re 2b gives a graphic re p re s e n t a t i o nof what happens when this control is ch a n g e d .You’ll find cases when you’ll want to sweep thed e l ay time imperc e p t i b ly slow ly (the dashedl i n e ) , and other times where you dial in a fa s t ,ve ry audible rate (the solid line).

D epth c o n t rols how mu ch the delay is modulat-e d . Fi g u re 2c gra p h i c a l ly contrasts two differe n ts e t t i n g s .That fixed delay time might be incre a s e dand decreased by 5 milliseconds (the dashedl i n e ) , 50 milliseconds (the solid line), or more .

The third contro l , S h a p e ,describes how thed evice moves from one delay setting to the nex t .

BY ALEX CASE

The Short Delay, Pa rt II–The Short Delay, Pa rt II–Flange and Chorus

This month’s Nuts and Bolts column wraps up ourt h ree-month look at delay. Your writer hopes that

by the end of this column you’ll have re n ewed interest inusing the humble delay, a simple effect with powe r f u lsonic capab i l i t y.

In the preceding two-month discussion on delay s — l o n gand short—we ’ ve spoken almost entire ly about f i x e dd e l ays (i.e. setting the signal processor to a singleu n changing delay time). Things re a l ly get interesting asyou start to change the delay time as part of the effect.M a ny effects are built on va rying delay s .


As Fi g u re 2d show s , it can sweep in aperfect sinusoidal shape (the dashedline) back and forth between theupper limit and the lower limit yo uset (those upper and lower limits yo uset with the Depth contro l ) . Or yo umight want a square wave sort of tra-j e c t o ry between delay times, in whichthe delay time snaps—instead ofswe e p s — f rom one setting to the other.

Fi g u re 2d highlights a commonf e a t u re of the Shape control: it letsyou use a shape that is some mixtureof the two—part sinu s o i d , p a r ts q u a re . Your delay may also have arandom setting in which the delaytime moves less ord e rly between thet wo delay ex t re m e s .

Fi n a l ly, some delay units let you usea combination of all the ab ove , fo rexample va rying the delay time in as l i g h t ly ra n d o m ,m o s t ly sinusoidal gen-e ral pattern.The Shape control letsyou mix these options and set a con-tour for how the delay moves betwe e nits highest and lowest settings.

These three controls let you takec o n t rol of the delay and play it like amusical instrument.You set how fa s tthe delay moves (ra t e ) .You set thelimits on the range of delay timesa l l owed (depth). And you determineh ow the delay moves from its shortestto its longest time (shape). F l a n g i n g ,ch o r u s i n g ,d o u b l i n g , and re l a t e deffects are now yo u rs for the taking.

FlangeDialing in a ve ry short delay time

and modulating it via these thre ec o n t rols lets you create flanging. Th eo n ly rule is that the delay timeneeds to be less than about 20 mil-liseconds—in fact I recommend start-ing with a delay time of 10 millisec-onds or less. This ensures that audi-ble comb filtering will occur. Set thed e l ay modulation controls to taste.

The flanger’s ringing, w h o o s h i n g ,ear tickling sound comes from thecomb filter effect we discussed lastmonth in combination with the mod-ulation controls we just went ove r.

Recall from last month what hap-pens when you combine a signalwith a delayed ve rsion of itself.When the delay time is belowabout 20 milliseconds, certain fre-quencies are constructive ly re i n-fo rc e d . Other frequencies opposee a ch other, a t t e nuating or eve ncanceling each other out. Th a t ’sgood old comb filtering.

Those peaks and va l l eys in the fre-q u e n cy spectrum introduced by ashort delay offer a distinct sound.The specific frequencies where spec-t ral boosts and cuts occur depend onthe specific short delay time we use.One delay setting causes the peaksand va l l eys to occur at one set of fre-q u e n c i e s . A different delay settingresults in a different set of peaksand va l l eys (see Fi g u re 3).

The way cool effect comes fro mmodulating the delay time. As yo u rmodulated delay sweeps from oned e l ay time to another, the comb fil-ter bumps and notches sweep also.Fi g u re 3 shows the result: flanging.

Sometimes you don’t hold back :the entire mix gets flanged. O t h e rtimes you might apply the effect to asingle instrument, l i ke the drum kit.You might flange just a single tra ck .Or you might limit the effect to justone section of that tra ck (e.g. o n ly onthe bridge). Flange to taste.

Pop music is full of examples off l a n g i n g .“ Then She Ap p e a re d ”f ro mX T C ’s last effort as a band on thealbum N o n e s u ch o f f e rs a good cases t u dy of a gently sweeping flange.E a ch time the wo rds “ Then shea p p e a re d ”o c c u r, a bit of your tra d i-tional flange begins courtesy of a shortvocal delay being slow ly modulated.

In this example the flange comesand goes throughout the song, o f f e r-ing us a good chance to hear thevocals with and without the delayt re a t m e n t . Feel free to take a moresubtle appro a ch , as on Mich a e lPe n n ’s “ C over Up” f rom the albumR e s i g n e d. A wa cky flange appears onthe vocal for the single wo rd‘guests’ near the end of the secondve rs e . Th a t ’s it. No more flange onthe vocal for the rest of the tune.I t ’s just a pop mix detail to makethe arrangement that mu ch morei n t e re s t i n g .

The flange effect actually softensthe rather hard sounding, s i b i l a n t ,and difficult to sing ‘sts’ at the end ofthe wo rd ‘ g u e s t s .’ And it makes mesmile eve ry time I hear it. The simpleeffect that comes from mixing in as h o r t , modulated delay offers a bro a drange of audio effects. F l a n g i n gi nvites your cre a t ive ex p l o ra t i o n .


DoubleThe first Nuts & Bolts of Delay col-

umn (Ju ly 2000) focused on longd e l ays—those greater than about 50m i l l i s e c o n d s . We saw how long delay sa re used for a broad range of ech o -based effects.

Short delays of about 20 millisec-onds or less create the ra d i c a lcomb filtered effect that, e s p e c i a l-ly when modulated, we call flang-i n g . What goes on in between 20and 50 milliseconds? Natura l ly, t h ebest way to answer that question isto listen to the effect of combininga signal with a medium delay some-w h e re between about 20 and 50m i l l i s e c o n d s .

The result is neither an echo nor af l a n g e .The delay is too short to bep e rc e ived as an ech o. It happens toofa s t . But the delay is too long to leadto audible comb filtering.

What do these medium delays do?Try a 30 millisecond delay on a vo c a l

t ra ck for a good clue. This kind ofmedium delay sounds a little bit likea double tra ck — l i ke two tra cks of thesame singer singing the same part.

It is a common mu l t i t ra ck pro d u c t i o nt e chnique to have the singer double at ra ck .You re c o rd a killer take ,t h e nh ave the singer re c o rd the part againon a separate tra ck along with him orh e rs e l f.The resulting sound is stro n g e rand rich e r. It even shimmers a little.


If you are unfamiliar with thesound of doubled vocal tra ck s , aclean example can be found at thebeginning of “ You Never Give MeYour Money ” on the Beatles’ A bb e yR o a d.Ve rse One begins with solovo c a l . On the wo rds “ f u n ny paper”the doubling begins. The vo c a lremains doubled for the next lineand then the harmonies commence.

Speaking of harmonies, a m o n gother things Roy Thomas Baker isfamous for taking doubling to theh i l t .C h e ck out the harmonies, d o u-bled and tripled (pro b ab ly mu chm o re ) , t h roughout The Cars’ firs ta l b u m . For ex a m p l e , listen to thef i rst harmonies on the first song“Good Times Roll,” when they singthe hook “good times ro l l .” It soundsdeep and immense; the vocals takeon a slick , hyped sound.

This layering of tra cks borrow sf rom the tradition of forming instru-mental sections in orch e s t ras andch o i rs . The value of having mu l t i p l einstruments play the same mu s i c a lpart is indescribab ly magic. A dd i n gm o re playe rs doesn’t just cre a t em o re volume—the combined soundis rich and ethere a l . It tra n s p o r t sthe listener.

A contempora ry application ofdoubling can be found on MacyG ray ’s “I Try ” f rom the album O nHow Life Is. Ty p i c a l ly, d o u b l e t ra ck ssupport the vo c a l , a dding their inex-p l i c able ex t ra bit of polish. Th ey areg e n e ra l ly mixed in a little lower inl evel than the lead vo c a l , re i n fo rc i n gthe principal tra ck from the centeror panned off to each side.

The Macy Gray tune turns this onits head. At the ch o r u s , w h e re yo uneed a good strong vo c a l , the vo c a lt ra ck panned dead center does some-thing quite brave: it all but disap-p e a rs . The chorus is sung by double-t ra cks panned hard left and right.

I t ’s brilliantly done. Rather thansupporting the vo c a l ,t h ey become thevo c a l .The chorus doesn’t lose stre n g t hand the tune doesn’t sag or lose ener-gy one bit.The doubled tra ck s —panned hard but mixed aggre s s ive lyfo r wa rd—offer a contagious hook thati nvites the listeners to sing along.

Pop vo c a l s — e s p e c i a l ly back gro u n dvo c a l s — a re the instrument mostoften doubled, fo l l owed fa i rly closelyby the rhythm electric guitar. Th esame part is re c o rded on two differ-ent tra ck s . On mixdow n , t h ey appearpanned to opposite sides of thes t e reo field.

The two parts are nearly identical.Sometimes you sw i t ch to a differe n tg u i t a r, a different amp, a differe n tm i c rophone or slightly change the tone


of the doubling tra ck in some otherway. M aybe the only differe n c eb e t ween the tra cks is the perfo r m a n c e .

E ven ro ck and roll guitar legendsa re human (mostly ) , leading to a pairof guitar tra cks that va ry slightly int i m i n g . The ‘ chugga chugga’ of theleft guitar tra ck is slightly early inone bar, s l i g h t ly late in the nex t .

Th rough the interaction of the twoguitar tra ck s , our ears seem to pickup on these subtle delay ch a n g e s . A ttimes the two tra cks are so similart h ey fuse into one meta instrument.Then one tra ck pulls ahead and wenotice it. Then the other tra ck pullsahead in time and tempora r i ly draw sour attention.

Doubled guitars are part audio illu-s i o n , part audio roller coaster. Th eya re an audio tre a t , plain and simple.

L ayering and doubling tra cks canbe simulated through the use of amedium delay. If it isn’t conve n i e n t ,a f fo rd ab l e , or phy s i c a l ly possible toh ave the singer double the tra ck , j u s trun it through a medium delay.Modulate the delay so that the dou-bled tra ck moves a little. This helps itsound more org a n i c , not like a clonec o py of the original tra ck .

The result is the beginning of as l i ck , mu l t i t ra ck effect. A dd a bit ofre g e n e ration (the lower control inFi g u re 1), and you’ll create a few lay-e rs of the tra ck underneath the pri-m a ry one.

Some delay units have the ability tooffer seve ral delay times simu l t a n e-o u s ly (called a multitap delay ) . D i a lin seve ral slightly different delaytimes in the 20 to 50 millisecond are aand you are synthesizing the rich n e s sof many laye red vo c a l s . S p read themout to different pan positions for awide wall of vocal sound.

Fun stuff. Just make sure thesound is appropriate to the song. Th esolo folk singer doesn’t usually bene-fit from this tre a t m e n t . Neither doesthe jazz trumpet solo. But many poptunes welcome this as a specialeffect on lead vo c a l s , b a cking vo c a l s ,key s , s t r i n g s ,p a d s , b a s s , and so on.

ChorusAn alternative name for the dou-

bling effect is ch o r u s . The idea isthat you could add this delay effectto a single vocal tra ck to simu l a t e

the sound of an entire ch o i r —ch o r u s— t h rough the use of mediumd e l ay s . N a t u ra l ly, s t a cking up 40medium delays of a single vo c a l i s twill not sound conv i n c i n g ly like achoir of 40 people. Think of itinstead as a special electronic effect,not an acoustic simu l a t i o n .

And it isn’t just for vo c a l s . Jo h nS c o f i e l d ’s tra d e m a rk tone includes as t rong dose of chorus (and distortion,and a sweet guitar, and brilliantp l ay i n g , among other things). Yo u ’ l loften hear a bit of chorus on theelectric bass. This medium delay con-coction is a powerful tool in the cre-ation of musical tex t u re s .

To see how “out there” the effectcan be made, spin Throwing Copperby Live and listen to the beginningof the tune “Lightning Crashes.” I’veno idea what kind of craziness isgoing on. The guitar sound includesshort and medium delays, among thepanning, distortion, and phase shift-ing effects going on. To my ear thedelay is being modulated between ashort flanging sort of sound (around10 milliseconds) and up to a longer,chorus sort of delay time (aroundmaybe 40 milliseconds).

Note especially the sound of theguitar in the second ve rs e , when theeffect and the re l a t ive ly clean sound

of the guitar are mixed together atsimilar vo l u m e s . You get a good tasteof ch o r u s . And you get great inspira-tion to do more with it.

A simple delay unit offers a broadrange of audio opportunities, repre-senting a nearly infinite number ofpatches. Short delays create thatfamily of effects called flanging.Medium delays lead to doubling andchorusing.

We can take a quick tour of all ofthe ab ove with a single album: K i ckby INXS. To hear a terrific use off l a n g e , listen to “ M e d i a t e .” Th ey re a l-ly went for it. True doubling? Listento “ S we a t ” and those hard pannedq u e s t i o n s ,“ H ow do you feel? Wh a tdo you think? Wh a t cha gonna do?”

Fi n a l ly, the same album gives us aclassic application of chorus to anelectric guitar. C h e ck out the rhy t h mguitar in “ N ew Sensation” and thes t e e ly cool tone the chorusing add s .

Of cours e , when the delay is longenough it separates from the original

signal and becomes its own perc e p t i-ble event: an ech o. F l a n g e , ch o r u s ,and ech o — t h ree ve ry different kindsof effects that come from a singlekind of effects device: the delay.

In fa c t , t h e re ’s more . This Nuts andBolts series will soon discuss reve r band pitch shifting, t wo more classesof effects that, at their heart, a rebased on the delay.

Alex Case used to sing in a ch o r u s .He hopes to one day sing in a flange.


Excerpted from the September edition of RECORDING magazine. ©2000 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119 For Subscription Information, call: 1-800-582-8326


D is the point of maximum negative amplitude ando c c u rs 3 milliseconds after the beginning of the cy c l e .And E has our sine wave returning 4 milliseconds afterpoint A to what looks ex a c t ly like the starting point: theamplitude is ze ro, and incre a s i n g . O k ay so far? We aregoing to fo l l ow these points through some signal pro c e s s-ing and move them around a bit.

Fixed delayRun this sine wave through a fixed delay of, s ay, 1 mil-

l i s e c o n d , and you get the situation described in Fi g u re1 i i . Vi s u a l ly, you might say the sine wave slips along theh o r i zontal time axis by 1 millisecond.

Looking point by point, Table 1 shows us what hap-p e n s . Point A originally occurred at a time of ze ro.I n t roduce a fixed delay of one millisecond and Point An ow occurs at time equals one millisecond.

The other points fo l l ow. Point D, for ex a m p l e , o c c u r re du n d e l ayed at a time of 3 milliseconds. After it has beenrun through a fixed, u n changing delay of one millisecond,Point D is fo rced to occur at a time of 4 milliseconds.

Accelerating delayH e re ’s the mind bender. What happens if the delay

i s n ’t fixed? What if the delay sweeps from a starting timeof 1 millisecond and then incre a s e s , and incre a s e s , a n di n c re a s e s . . . ?

Table 2 summarize s . H e re the delay changes at a ra t eof one millisecond per millisecond. S ay what? For eve rymillisecond that ticks by during the course of this ex p e r i-m e n t , the delay gets longer by one millisecond. Fo r

P i tch Shift i n gP i tch Shift i n gBY ALEX CASE

We all know what happens when you play an audiotape at a faster speed than it was re c o rded: the

p i t ch of the re c o rding goes up. P l ay it slower than there c o rded speed and the pitch goes dow n . S o m ew h e re inthis simple principle lies an opportunity for audio ex p l o-ration and entertainment.

Can it be done digitally? Of cours e . H ow is it done?Using a delay. Ye s , I said delay. L oyal re a d e rs of Nuts &Bolts h ave just spent the last three issues reading ab o u td e l ays: ech o, ch o r u s , f l a n g e , and their many cousins. Ye twe are n ’t finished discussing delay, because even pitchs h i f t e rs are built on this effect.

To see how, yo u ’ ve got to put up with a bit of math( w h i ch I seem to sneak into eve ry article) and fo l l owalong with Fi g u re 1. Fi g u re 1i shows a simple sine wavewith a chosen fre q u e n cy of 250 Hertz. This sine wavecompletes a cycle eve ry 4 milliseconds, confirmed cour-tesy of the fo l l owing familiar equation:

Period = the time to complete one cy c l ePeriod = 1 / Fre q u e n cyPeriod = 1 / 250 HzPeriod = 0.004 seconds or 4 millisecondsFi g u re 1i labels some key landmarks during the cours e

of a single cy c l e . We start the clock at point A .H e re , at time equals ze ro, the sine wave is at an ampli-

tude of ex a c t ly ze ro and is incre a s i n g . It re a ches its posi-t ive peak amplitude at B, taking ex a c t ly 1 millisecond todo so. It has an amplitude of ze ro again at the halfwaypoint (time equals 2 milliseconds) labeled C. This make sit look a lot like point A , but while the amplitude is thes a m e , n o t ab ly ze ro, it is decreasing this time.


ex a m p l e , if at one point the delay is 10 milliseconds,then five milliseconds later the delay unit is operating at15 milliseconds.

If you have n ’t had the chance to study phy s i c s , yo umight be puzzled by the idea of changing the delay timeat a rate of 1 millisecond per millisecond. I find it help-ful here to get in my car.

S ay you are driving at a speed of 8 5 (edited for your safe-ty) 55 miles per hour and accelerating at a rate of 1 mphper hour (our Canadian neighbors should use kilometersper hour for similar re s u l t s ) . That means that for each hourthat passes by your speed increases by one mile per hour.D r iving 55 mph now becomes 56 mph an hour from now.Wh o a !

If you subscribe to the idea that cops wo n ’t pull yo uover for speeding until you are at least 10 miles per hourover the speed limit, then—starting at the speed limit—you can drive most of the way across Texas (it’s about 600miles from Dallas to El Paso) without getting a ticke t .

B a ck to the mu s i c . H e re we are increasing the delay by1 millisecond each millisecond. And the surprising re s u l tis a change in the pitch of the tra ck .

Table 2 shows the location of our sine wave landmark sboth befo re and after thei n t roduction of thiss t e a d i ly increasing delay.Point A initially occurs ata time of ze ro. At thistime the delay is alsoze ro. Point A thenremains unchanged ando c c u rs at time ze ro.

Skip to point C. It orig-i n a l ly occurs at a time oft wo milliseconds. By thistime the delay hasi n c reased from ze ro tot wo milliseconds. Th i sd e l ay of two millisecondsleads point C to finallyoccur at a time of fo u rmilliseconds after thebeginning of this ex p e r i-m e n t .

Do the math point bypoint and you get a sinewave that looks likeFi g u re 1iii.The key land-m a rks are identified.The result is clearly still a sine wave .But since it takes longer to complete the cy c l e , we know thep i t ch has ch a n g e d .

B a ck to our trusty equations. Looking at the new sinewave at the bottom of Fi g u re 1, l e t ’s calculate its fre-q u e n cy. The sine wave in Fi g u re 1iii takes a full 8 mil-liseconds to complete its cy c l e .

Period = the time to complete one cy c l ePeriod = 8 milliseconds or 0.008 secondsFre q u e n cy = 1 / Pe r i o dFre q u e n cy = 1 / 0.008Fre q u e n cy = 125 HertzThat’s right.The constantly increasing delay caused

the pitch of the signal to change. We sum it up as fol-lows.A 250 Hertz sine wave run through a delay thatincreases constantly at the rate of 1 millisecond per mil-lisecond will be lowered in pitch to 125 Hertz. Strange,but true.

Skipping the details—though you are encouraged top rove these on your ow n — we find a number of finerpoints on pitch shifting. Our example demonstrated that

an increasing delay lowe rs the pitch . It is also true that ad e c reasing delay raises the pitch .

In add i t i o n , our example found that changing the delayat a rate of 1 millisecond per millisecond moved the pitchby an octave . It is also possible to change the pitch by twoo c t ave s , or a minor third , or a perfect fifth—whatever yo ud e s i re; one need only change the delay at the correct ra t e .

So the underlying methodology of pitch shifters isreve a l e d . A pitch shifter is a device that changes a delayin specific controlled ways so as to allow the user toaffect the pitch of the audio.

N a t u ra l ly, it ain’t easy. We ’ ve got a key pro b l e m . R e t u r nto our example in which we lowe red the 250 Hz sine waveby an octave through a steadily increasing delay. If weimagine applying this effect to an entire three and a halfm i nute tune, not just a single cycle of a sine wave , wefind we are increasing the delay from a starting point ofone millisecond to a final delay time of 210,000 millisec-onds (31/2 m i nutes equals 210,000 milliseconds).

That is, at the start of the tune we add an incre a s i n gd e l ay: 1 ms, then 2 ms, and so on. By the end of the tune,we are adding a delay of 210,000 milliseconds. This high-lights two pro b l e m s .

Fi rs t , we need a ve ry long delay. Most delays are capa-ble of a one second delay (1,000 milliseconds) at them o s t . Super cool mega turbo delays might go up tom aybe 10 seconds of delay. But a delay of hundreds ofthousands of milliseconds (hundreds of seconds) is a lotof signal processing hors e p ower that is ra re ly ava i l ab l e —RAM isn’t that ch e a p.

S e c o n d , our song, w h i ch used to be 31/2 m i nutes long, d o u-bles in length to seven minutes as we lower the pitch by oneo c t ave . Consider the last sound at the ve ry end of the song.

B e fo re pitch shifting it occurred 31/2 m i nutes (210,000milliseconds) after the beginning of the song. By this timeour pitch shifting delay has increased from 1 millisecondto 210,000 milliseconds. Th e re fo re the final sound of thep i t ch shifted song occurs at 210,000 milliseconds (originaltime) plus 210,000 milliseconds (the length of the delay ) .That is, the song now ends 420,000 milliseconds (that’ss even minutes!) after it began. The 31/2 m i nute song isl owe red an octave but doubled in length.

S i m p ly increasing the delay fo rever as ab ove is ex a c t-ly like playing a tape back at half the speed it wa s


re c o rd e d . The pitch goes down andthe song gets longer.

Wh e re pitch shifting signal pro c e s-s o rs differentiate themselves fro mtape speed tricks is in their cleve r n e s sh e re . Digital delays can be manipulat-ed to always incre a s e , but also to re s e tt h e m s e l ve s .

In our sine wave ex a m p l e , what hap-pens if our digital delay increases at arate of ex a c t ly one millisecond per mil-lisecond but never goes over 50 mil-liseconds in total delay? That is, eve rytime the delay re a ches 50 millisecondsit resets itself to a delay of ze ro andc o n t i nues increasing from this newd e l ay time at the same rate of one mil-lisecond per millisecond.

The result is pitch shifting that neve ruses too mu ch delay and never make sthe song more than 50 milliseconds longer that the un-p i t ch shifted ve rs i o n . After all, our analysis showed itwas the rate of change of the delay that led to pitch shift-i n g , not the absolute delay time itself. A ny delay time

that increases at rate of 1 millisecondper millisecond will lower the audioby one octave .A ny delay time. So whynot keep it a small delay time?

The devil is in the details. G e t t i n gthe pitch shifter to reset itself in thisway without being noticeable to the lis-tener isn’t easy. It is a problem solve dby clever softwa re engineers who findways to make this inaudible.

Older pitch shifters ‘ g l i t ched’ as they tried to return tothe original delay time.These day s — we are lucky to bea l ive in audio in the year 2000—those glitches are mostlyg o n e .To d ay we simply re a ch for a device labeled “ p i t chs h i f t e r,” dial in the desired settings (gimme a shift up of amajor third mixed in with a shift down of a perfect fo u r t h ) ,and get to wo rk . Life is good.

Side effectsB e fo re we get into the special effects we create with

p i t ch shifting dev i c e s , it is worth noticing that pitchshifting is a natural part of some effects we ’ ve alre a dyi nvestigated here in the Nuts & Bolts s e r i e s . Recall thechorus effect that comes from adding a slow ly modulatedd e l ay of about 30 milliseconds.

As you listen to the richness that the chorus effecta dds to a vocal or guitar, listen for a subtle amount ofp i t ch shifting. Th a t ’s right, p i t ch shifting is a componentof that effect we call ch o r u s .

Since a chorus pedal relies on a modulating delay, i ti n t roduces a small amount of pitch shifting. As the delaytime sweeps up, the pitch is slightly lowe re d . As thed e l ay time is then swept dow n , the pitch is then ra i s e d ,ever so slightly. Repeat until thicke n e d .

Special effectsThe Nuts & Bolts rev i ew of the a basic pop mix, ‘ M i x i n g

by Numbers’ in the 4/00 issue, i n t roduced a commoneffect built in part on pitch shifting: the spre a d e r.

A quick rev i ew of this effect: the spreader is a “ p a t ch ”that enables you to take a mono signal and make it a littlem o re stere o - l i ke .You ‘ s p read’ a single tra ck out by sendingit through two delays and two pitch shifters .

The delays are kept short, e a ch set to different va l u e ss o m ew h e re between about 15 to 50 milliseconds. To oshort and the effect becomes a flange/comb filter (as wediscussed last month). Too long and the delays stick out asdistinct ech o e s . So our window for acceptable delay timesin this effect is between about 15 and 50 milliseconds.


In using a spreader, the return ofone delay output is panned leftwhile the other is panned right. Theidea is that these quick delays add adose of support to the originalmonophonic track. In effect, thesetwo short delays simulate some earlysound reflections that we wouldhear if we played the sound in a realroom. The spreader takes a singlemono sound and sends it to twoslightly different short delays to sim-ulate reflections coming from theleft and right.

Th a t ’s only half the story. Th eeffect is taken to the next leve lcourtesy of some pitch shifting.Shift each of the delayed signalsever so slightly, and the fo r m e rlyboring mono signal becomes a mu chm o re immers ive , i n t e resting loud-s p e a ker cre a t i o n .

Detune each delay by a nearlyi m p e rceptible amount, m aybe 5 to 15c e n t s . The goal of the spreader is toc reate a stereo sort of effect. As are s u l t , we try to keep the signal pro-cessing on the left and right sidesever so slightly different from eacho t h e r. Just as we dialed in uniqued e l ay times for each side of thise f f e c t , we dial in different pitch shiftamounts as we l l — m aybe the left sidegoes up 8 cents while the right sidegoes down 8 cents.

L i ke so mu ch of what we do inre c o rding and mixing pop mu s i c , t h eeffect has no basis in re a l i t y. B ya dding delay and pitch shifting, wea re n ’t just simulating early re f l e c-tions from room surfaces any m o re .The spreader makes use of our signalp rocessing equipment (delay andp i t ch shifting) to create a big stere osound that only exists in loudspeake rmu s i c . This sort of thing doesn’t hap-pen in symphony halls, o p e ra houses,stadiums or pubs. I t ’s a studio cre-a t i o n , plain and simple.

Ta ke this effect further and yo uend up with what I think of as a‘ t h i cke n e r.’ Why limit the patch tot wo delays and two pitch ch a n g e s .What if you have the signal pro-cessing hors e p ower in your DAW orin your ra cks of gear to ch a i nt ogether eight or more delays andp i t ch shifts?

Try it. While it’ll sound unnaturalwhen used on vocals, many keyboardparts respond well to the thickeningtreatment. Modulate those delayslike a chorus and, guess what? Morepitch shifting is introduced.Addedin small, careful doses, this denselypacked signal of supportive, slightlyout-of-tune delays will strengthenand widen the loudspeaker illusionof the track.


Big timeEnough with these subtle pitch

ch a n g e s . L e t ’s add a serious amountof pitch shifting.

Hammond B3 organs and manyblues guitars are often sent thro u g ha rather wa cky device: the LeslieC ab i n e t . The Leslie is a hybrid effectthat is built on pitch shifting, vo l u m ef l u c t u a t i o n , and often a good dose oftube ove rd r ive distortion.

It is essentially a tricked out gui-tar or keyboard amp in which thespeakers rotate. Honest. The highfrequency driver of a Leslie is hornloaded, and the horn spins aroundwithin the amp.

C razy as it sounds, the engineerswho came up with this we re re a l lyt h i n k i n g . It would be ve ry difficult tospin the large woofer to continue theeffect at low fre q u e n c i e s . I n s t e a dt h ey enclosed the woofer inside ad r u m . The drum has holes in it andro t a t e s . The result is a low fre q u e n cys i mulation of what the Leslie isdoing with the horns at higher fre-q u e n c i e s , e r, we l l , sort of.

The Leslie is too funky a device toc over in detail now, but we mentionit because it is part of our pitchshifting toolkit.

So what’s it sound like? With thedrum and horn ro t a t i n g , the loudnessof the music increases and decre a s e sat any given listener position—amplitude modulation. And with thehigh fre q u e n cy horn spinning by, aDoppler effect is created: the pitchi n c reases as the horn comes towa rdthe listener/microphone and thend e c reases as the horn travels away.

The typical example used in thes t u dy of the Doppler effect is a tra i ngoing by, horns ab l a ze . That classicsound of the pitch dropping as thet rain passes is based on this princi-p l e . Sound sources appro a ching witha ny appre c i able velocity willi n c rease the perc e ived pitch of thes o u n d . As the sound source departs,the pitch similarly decre a s e s . The netresult of the Leslie system then is aunique fluttery and wo bb ly sound.

The Leslie effect is used whereve rB3s and their ilk are used. Ty p i c a l lyoffering two speeds of ro t a t i o n , yo ucan hear a fast Leslie and a slowLeslie effect, as well as the accelera-tion or deceleration in betwe e n .

Listen to the single note org a nline at the introduction to ‘ Ti m eand Time Again’ on the CountingC rows’ first re c o rd , August AndEverything After. The high notee n t e rs with a fast rotating Leslie.As the line descends, the speed isre d u c e d . Listen care f u l ly thro u g h-out this song, this album, and other


B3-centric tunes and you’ll hear theLeslie pitch shifting vo c ab u l a rythat key b o a rdists love .

Of cours e , you can apply it to anyt ra ck you like — g u i t a r, vo c a l , o b o e —if you have the device or one of itm a ny imitators or simu l a t o rs .

The hazard with an obvious pitchshift is that it can be hard to getaway with mu s i c a l ly. Yo u ’ ve heardspecial effects in the movies and onsome re c o rd s , w h e re a vocal is shift-ed up or down by an octave or more .Too low and it conjures up imagesof death robots invading the mix toeat Shreve p o r t . Too high and yo u rsinger becomes a gre m l i n - o n - h e l i u msort of disaster.

In the hands of talented mu s i c i a n s ,a g gre s s ive pitch shifting re a l lywo rk s . TA N K A PA (The Artist NowK n own As Prince Again) lowe rs thep i t ch of the lead vocal tra ck andt a kes on an entire ly new persona inthe song, Bob George from The BlackA l bu m. The effect is obv i o u s . Th eresult is fa n t a s t i c .

No effort was made to hide theeffect in the bass line ofSledgehammer on Peter Gab r i e l ’sclassic ‘ S o.’ The entire bass tra ckseems to include the bass plus thebass dropped an entire octave .A n dthe octave down bass line is mixedright up there with the original bass.Nothing subtle about it.

You can even use a pitch shifter toa dd two - ,t h re e - , or fo u r-part harmonyif you are so inclined. But get outyour arranging book, because the

p i t ch shifter makes it easy to add ac o n s t a n t ,n ever ending, n ever ch a n g-i n g , just plain annoying harmony line.

Dial in a pitch change that is amajor third up and add it to the leadvo c a l . If the song is entire ly diatonicwithin a major key and is a ve ryh a p py song, I mean ve ry happy, t h i smight wo rk . Otherwise it is going tosound cloy i n g ly swe e t , l i ke add i n gmaple syrup to the ice cream you puton top of your shoo fly pie.

The trick to creating harmoniesusing pitch shifting is to composemusical harmonies. And a staticp i t ch shift will ra re ly cut it.

Fo r t u n a t e ly, d evices and softwa replug-ins to facilitate this abound (pro b-

ab ly the most famous is the DigiTe chVocalist series).The pitch shifting cane s s e n t i a l ly be tied to MIDI note com-mands enabling you to dictate the har-monies from your MIDI contro l l e r.Th ep i t ch shifter is processing the vo c a lline on tape or disk according to thenotes you play on the key b o a rd .

This results is a harmony or coun-t e r m e l o dy line with all the harmonyand dissonance you desire . I t ’s builton a single vocal tra ck , and re l i e salmost entire ly on good soundingp i t ch shifting.

Go beyond harmonies. Use pitchshifting to turn a single note into ane n t i re ch o rd . String patches cansometimes be made to sound moreo rch e s t ral with the judicious add i t i o nof some perfect octave and perfectfifth pitch shifting (ab ove and/orb e l ow) to the patch .


And don’t stop with simple inter-va l s .C h o rds loaded with tensions areo k ay too, used we l l . Yes put it fro n tand center in ‘Owner of a LonelyHeart’ on the album 90125. S i n g l e -note guitar lines are tra n s formed intosomething more magic and less gui-t a r- l i ke using pitch shifters to cre a t ethe other notes.

A final obvious pitch shiftingeffect worth mentioning is the stoptape effect. As analog tape risksex t i n c t i o n , this effect may soon belost on the next generation of re c o rd-ing mu s i c i a n s .

When an analog tape is stopped, i td o e s n ’t stop instantly; it takes aninstant to decelera t e . L a rge reels oft a p e , l i ke two inch 24 tra ck , a re pre t-ty darn heav y. It takes time to stopthese large reels from spinning. I fyou monitor the tape while it tries tostop (and many fa n cy mach i n e sresist this, a u t o m a t i c a l ly muting to

avoid the distraction this causes dur-ing a session), you hear the tape slowto a stop. S ch l u m p. The pitch dive sd own as the tape stops.

This is sometimes a musical effect.And it’s not just for analog tape, a sGarbage demonstrates via a ProTools effect between the bridge andthe third Chorus of ‘I Think I’mPa ranoid’ on their second album.

Surgical effectsP i t ch shifting is also used to zo o m

in and fix a problematic note. We ’ veall been there .

In the old days of mu l t i t ra ck pro-duction (about a year ago), we usedto sample the bad note. Then wetuned it up using a pitch shifter. I twas raised or lowe red to taste.Fi n a l ly, the sampled and pitch shift-ed note was re - re c o rded back ontothe mu l t i t ra ck . With the pro b l e m a t i cnote shifted to pitch perfection, n oone was the wiser.

That was then. N ow, c l ever softwa retaking advantage of powerful hard-

wa re permits pitch shifting to be donea u t o m a t i c a l ly (Antares Au t o Tu n e[ h a rdwa re and softwa re ve rs i o n s ] ,Wave Mechanics Pitch Doctor, T CE l e c t ronic Intonator, e t c . ) . That is, t h eeffects device can monitor the pitch ofa vo c a l ,v i o l i n , or didgeridoo. When itdetects a sharp of flat note, it shiftsthe pitch automatically by the amountn e c e s s a ry to re s t o re tuning. Wow. A n dit re a l ly wo rk s .

But please be careful with thesed ev i c e s . Fi rs t ,d o n ’t over polish yo u rp ro d u c t .P i t ch shifting eve rything intoperfect tune is ra re ly desirab l e .Vi b ra t ois an obvious example of the de-tuningof an instrument on purp o s e .

And if Bob Dylan had been pitchshifted into perfect pitch , w h e rewould folk music be now? Th e re is alot to be said for a musical amountof ‘ o u t - o f - t u n e n e s s .’ Remove all thewa r t s , and you risk re m oving a lot ofemotion from the perfo r m a n c e .

S e c o n d , d o n ’t expect to create ano p e ra singer out of a lounge cro o n e r.Th e re is no replacement for actualmusical ab i l i t y. If the bass playe rc a n ’t play a fre t l e s s , g ive her one withthose pitch-certain things called fre t s .If the violin player can’t control hisi n t o n a t i o n ,h i re one who can.

D o n ’t expect to rescue poor mu s i-cianship with automatic pitch corre c-t i o n . People want to hear your mu s i c ,not your effects ra ck .

Out of timeThis month re p resents our fo u r t h

month of discussion on delay. A re wedone yet? Natura l ly, n o. We continu eour tour of the delay in a future Nuts& Bolts installment when we take adetailed look at reve r b .

Alex Case strapped his pitch shifter tohis gear shifter and drives by ch a n t i n g .


If Bob Dylan had been

p i t c h - s h i fted into perfect pitch,

where would folk music be now?

Excerpted from the October edition of RECORDING magazine.©2000 Music Maker Publications, Inc. Reprinted with permission.




Reverb, Pa rt 1Reverb, Pa rt 1BY ALEX CASE

What does space sound like ?

Deep space, w h e re NASA probes and Holly wood starsh ave often ve n t u re d , d o e s n ’t sound like mu ch . S o u n dt ravels through air, wa t e r, s t o n e . . . a ny t h i n g . The onlything that doesn’t propagate sound is... n o t h i n g . That is, aroom full of nothing, a va c u u m , will not have sound,w h i ch needs an elastic medium through which to trave l .Outer space is there fo re completely silent.

Fo rget outer space then. What does the sound of ap hy s i c a l ,a rch i t e c t u ral space sound like? What does ac a t h e d ral sound like? An opera house? A hall? A club?As you no doubt have ex p e r i e n c e d , all these spaces addtheir own signature to whatever sound happens withint h e m . When we listen to music in anything other than ana n e choic ch a m b e r, we listen to the sound of the mu s i cplus the sound of the ro o m .

The room acts like a signal processor: music in, mu s i cplus effects out. In fa c t , thanks to many different mu s i c -gear make rs , a rch i t e c t u ral spaces have essentially beens q u e e zed into ra ck spaces. R everb units are signalp ro c e s s o rs acting like acoustic spaces.

BlueprintAn understanding of the sound of a physical space

begins with a look at the floor plan of a ro o m . Fi g u re 1s h ows a source (S), w h i ch might be a singer, c e l l o, o rd i d g e r i d o o. It makes a sound in the ro o m . Time passesand the re c e iver (R) hears it.

As the rays show, we hear first the direct sound fro ms o u rce to re c e ive r. I t ’s the shortest path. But the re f l e c-tions are audible too. So that direct sound is fo l l owed bya quick vo l l ey of re f l e c t i o n s . S h o r t ly after the sound com-mences the listener is immersed in a field of thesereflections—too many to be identified discre e t ly. Th i sreflected sound energy in a room is reve r b e ra t i o n .

These reflections are different from the direct sound intime of arriva l , angle of arriva l , and spectral content.

Since the reflected sounds travel along a longer path thanthe direct sound, t h ey re a ch the re c e iver after the dire c ts o u n d . Bouncing off wa l l s ,f l o o r, and ceiling (and furni-t u re , music stands, and other mu s i c i a n s ) , t h ey also gener-a l ly arrive at a different angle than the direct sound.

A Look under the hood

Figure 1: Room reflections. The heavy line is thedirect sound, solid lines are single bounces,dashed lines double bounces, and the dottedline shows one of millions of multiple-bouncepaths that make up the reverberant sound.


Fi n a l ly, due to the energy thesound loses as it travels through theair and bounces off various ro o mb o u n d a ry surfa c e s , the amplitude ofthe signal at different fre q u e n c i e sch a n g e s . Air and fuzzy surfaces (likec a rp e t , f i b e rglass and foam) tend toabsorb high fre q u e n c i e s . F l ex i b l es u r faces (like ve ry large windows orpanels of wood) tend to absorb agood amount of low fre q u e n c i e s .

All said, the room intro d u c e sd e l ay, changes the angle of arriva l ,and manipulates the loudness ands p e c t ral content of a signal.

TimeH ow delayed are the reflections? It

depends on the room size and geome-t ry. The reflections in larger ro o m st a ke longer to re a ch the listener thanthe reflections in smaller ro o m s . I fthe source or re c e iver is particularlyclose to a room surfa c e , that ch a n g e sthe pattern of re f l e c t i o n s .

Listening to a sound fo l l owe di m m e d i a t e ly by its reflections seemsl i ke ly to be a rev i ew of the Nuts &Bolts Delay Tr i l ogy just completed( Ju ly–October 2000). We discussedh ow a delay of about 5 millisecondsi n t roduces comb filtering when com-b i n e d , in approx i m a t e ly equal parts,with the undelayed signal.

Because sound travels at ro u g h lyone foot per millisecond, that meansthat a signal whose reflected path isabout five feet longer than the dire c tpath will create comb filtering.Right? Not necessarily.

Try taking a harmonically richsound like a piano patch or tra ck .Send it to a short delay of about 5m i l l i s e c o n d s . Monitor both at ab o u tthe same vo l u m e . With both signalspanned to the same location in thes t e reo landscape, h a rd left for ex a m-p l e , the comb filter alteration to thef re q u e n cy content of the signal isu n m i s t a k ab l e .

N ow pan the delay to hard right.P resto—the comb filtering seems tod i s a p p e a r. Instead we get a localiza-tion cue: the delay seems to shift theimage of the piano towa rd the unde-l ayed signal.

Fo l l ow that thought and slow lyd e c rease the delay time. As thed e l ay time appro a ches ze ro, t h eplacement of the stereo image headst owa rd the center. All the while, t h ecomb filtering effect is gone.

This points out an enigmatic pro p e r-ty of short delays: the angle of arriva lm a t t e rs! Short delays dire c t ly com-bined with their undelayed bre t h re nwill create comb filtering. S h o r t


d e l ays re a ching the listener from ave ry different direction do no sucht h i n g . In our look at reve r b e ra t i o n ,this leads us to ask:

Where do the room reflectionscome from? This too depends on thephysical geometry of the room. Youneed only patience and a ruler tofigure out which reflections reachthe receiver.

Fi g u re 1 shows the first handful orso of re f l e c t i o n s . Some reflections wehear after a single bounce off a sur-fa c e . Other reflections strike two,t h re e , or more surfaces befo re finallyre a ching our ears . The direction fro mw h i ch they come seems to be a lotabout luck , s t a t i s t i c s , and/or thep hysical geometry of the ro o m .

Yet our personal audio analy s i ssystems (ears and brains) can makesense of this. Though it isn’t intu-i t ive , it is important to know that wea re n ’t distracted or confused bythese re f l e c t i o n s .

L e t ’s make thes o u rce a singer, s i n g i n gyour re c e n t ly pennedt u n e , “Insulate theA t t i c .” We zoom in onthe first wo rd of thec a t chy chorus for thish i t - wa i t i n g - t o - b e - d i s-c ove re d . She sings “ Fi b e rg l a s s . . .” a n dfor the sake of analysis we slow timed own like a Holly wood mov i e . Th ere c e iver hears the wo rd first dire c tf rom the sourc e ,“ Fi b e rg l a s s .” Then areflected ve rsion of the wo rd arrive sf rom one side, then the other, t h e nf rom behind, “ f i b e rg l a s s . . . f i b e r-g l a s s . . . f i b e rg l a s s .”

This ought to be confusing, but iti s n ’t . As you know from listening tomusic and conve rsations in re a ls p a c e s , the reflections coming fro mall around do not stop us from know-ing—at all times—where the singeris and what she’s singing.

R e s e a rch e rs have teased this outof various ex p e r i m e n t s . We localizethe source based on the angle ofa r r ival of the first wave form and t h epattern of reflections that immedi-a t e ly fo l l ow; we synthesize an opin-ion about the room in which thesound event happens based on thea m p l i t u d e , q u a l i t y, and angle ofa r r ival patterns of these supportingre f l e c t i o n s .

I t ’s tempting perhaps to think thatthe reflections from all around arei g n o red so as not to confuse our per-sonal audio analysis system. Q u i t ethe opposite. Sounds without thesupport of reflections are difficult tolisten to, difficult to localize , a n dsound just plain stra n g e .

We don’t hear sounds in anech o i cch a m b e rs ve ry often, after all, s oour hearing mechanism isn’t tai-l o red to that ex p e r i e n c e . If yo u ’ veh e a rd sound in an anechoic env i-ro n m e n t , you know it’s unnerv i n gand a little confusing. In fa c t ,re s e a rch has shown our localizationabilities suffer without some add i-tional re f l e c t i o n s , even though theycome from directions different thanthe direct sound.

Using amplitude, time of arriva l ,and spectral content, we make use ofthe clues these reflected sound wave so f f e r. Our personal audio analy s i s

system has developed the ability toabsorb a complex sound field, ex t ra c tthe direct sound, i n c o rp o rate thereflected sound field, and add it allup into a complete perception of asound in a space. P retty darn cool.

Synthesized spaceTo create the sound of a ro o m

without the use of an actual ro o mone need only assemble the set ofreflections a room would add to ad i rect sound.

A grotesque ove rs i m p l i f i c a t i o n .But even simplified, the illusionwo rk s . E a ch reflected sound suffersa bit of delay and attenuation hav-ing traveled farther than the dire c ts o u n d , and a bit of equalizationdue to air and boundary energ yab s o rp t i o n .

The only processes at wo rk arechanges of amplitude, e q , d e l ay, a n dangle of arriva l . Good new s , b e c a u s eeffects ra cks and pull-down menu sa re full of that sort of capab i l i t y —t raditional studio signal pro c e s s i n g

c a n , c l eve rly employe d , s i mu l a t ereve r b e ra t i o n . R a ckmount units dis-p l ay the wo rd ‘hall’ and do a fun jobof sounding like one.

Digital reve r b e ra t o rs are , to sum-m a r i ze , ve ry shrewd vo l u m ea d j u s t e rs , spectrum manipulators ,ch a n g e able panners , and va r i ab l ed e l ay s . An audio wave form goes inand triggers a nearly infinite set offa d e d , e q u a l i ze d , p a n n e d , a n dd e l ayed ve rsions of itself.

N a t u ra l ly, some equalize rs soundbetter than others , some delay unitssound better than others . And thewhole algorithm used to simulate thec o m p l ex pattern of sound energy isgoing to have an audible effect onthe sound of the reve r b .

Not surp r i s i n g ly then, some reve r bd evices sound better than others . A tthe ve ry least, most reverb dev i c e ssound different from most others onthe marke t . E a ch manu fa c t u re ro f f e rs its own appro a ch , c reating itsown sound; our studios benefit fro mh aving many different reve r b s . Th e reis no single best, just a broad paletteof reverbs awaiting our cre a t ive use.

Reverbus ex machinaL iving as we do at the edge of a

n ew millennium in a thriving digitale c o n o my full of dotcom mira g e s , we mayfo rget about lifeb e fo re audio was dig-i t i ze d . But some-w h e re between thetime all those crittersb o a rded ship withNoah and the pre s e n t

d ay, we had a period of non-digitala u d i o.

While it is fa i rly trivial today for acomputer to do a decent job simu l a t-ing the sonic ch a racter of a space, i tis ve ry difficult to do so with analoge l e c t ro n i c s . R e s o u rceful equipmentd e s i g n e rs looked for physical sys-tems that could sustain a sound likea decaying acoustic space wo u l d .Th ey found some success using twod evices: the spring and the plate.

The spring reverb offers an intuitivea p p ro a ch . Initiate subtle vibration in aspring using your audio wave fo r m ,and b o i n g ,let it go.The spring contin-ues to vibrate for a time, a bit like ahall sustains a single violin note.

We l l , s o r t a . The fact is, s p r i n g sd o n ’t ex a c t ly behave like ro o m s .Th ey are elastic and can respond tomu s i c , but the simulation ends there .

H oweve r, the musical value doesn’t !Just because a spring doesn’t soundl i ke the Musikve reinssaal in Vi e n n ad o e s n ’t mean it isn’t good enough fo rJimi or Stevie or Yo u . Leo Fender put

The room acts like a signal process o r :

music in, music plus effects out.


spring reverbs in electric guitar amps,and there ’s been no turning back .

Spring reverb rings with its ow ndistinct ch a ra c t e r. S u b t ly used, it fillsin underneath a tra ck , a dding sup-port and shimmer. O ve rd r ive n , i tc rashes and wo bbles (ever move aguitar amp while it was cra n ke da n d — c r w u waw u wawoing—the springgets jostled?).

Taking the spring idea and makingit two-dimensional leads us to theplate reve r b . This device is essential-ly a sheet of metal with a drive ra t t a ched to it to initiate vibra t i o nand a sensor or two or more to pickup the decay that ensues. ( Will sur-round sound lead to mu l t i ch a n n e lplates? I fear the answer is ye s . )

The plate is another mech a n i c a ls i mulation of an acoustic space. B a n gon a sheet of metal and it rings for aw h i l e , again somewhat like the soloviolin in a symphony hall. And likethe spring, as a simulation of anactual space the plate falls short.

But as a pop music effect it is asweet success.

Sweet sound, funky smellWhen an actual large hall isn’t fea-

sible and a spring or plate reve r bi s n ’t ava i l ab l e , t h e re ’s always theb a t h ro o m . L a rge spaces are reve r b e r-ant in part because they are larg espaces (I get paid to say this sort oft h i n g ? ) . That is, the reve r b e rance ofa space is dire c t ly proportional tothe size of the ro o m . M a ke the ro o mw i d e r, l o n g e r, and/or higher, and thereverb time increases (because thereflections have farther to trave l ) .

The other key driver of reve r b e ra-tion in a physical space is theab s o rp t ivity of the room surfa c e s .A b s o rp t ive materials on the floor,wa l l s , or ceiling will lower the reve r bt i m e . H a rd re f l e c t ive surfa c e si n c rease the reverb time.

The trouble with using reverb fro ma hall during a studio production isthat there isn’t usually a hall aro u n d .So lacking a large space with itsassociated long reve r b e ration time,we go to the only room around withre a l ly hard shiny surfaces: the tiledb a t h ro o m .

Because the tiles reflect sound ener-gy more than your typical room finisht reatments like gypsum wall board orc a rp e t i n g , the bathroom has a littlereve r b e rant kick .K i t chens sometimesa re a close second place. R a re ly car-p e t e d ,t h ey have a decent amount ofh a rd surfaces: countertops, a p p l i-a n c e s , wood cab i n e t s , and such .E l evator shafts and high rise fires t a i rs have contributed a big reverb tothe studio that could get away with it.


N a t u ra l ly, some studios built reve r-b e rant bathrooms on purp o s e .L o s ethe plumbing fixtures and make theroom a little bigger and yo u ’ ve got areverb ch a m b e r. Put in loudspeake rs(inputs) and microphones (outputs)and yo u ’ ve got a physical spacereve r b e ra t o r.

What it lacks in physical vo l u m e —i t ’s now h e re near the size of an operahouse—it makes up for in highlyre f l e c t ive surfaces of stone, t i l e ,c e m e n t , beer bottles, and such . Th ere s u l t , of cours e , i s n ’t an opera houses i mulation on the ch e a p, but a whollyd i f f e rent kind of reve r b e ra t i o n .

C h a m b e rs offer their own uniques i g n a t u re to the audio sent to them.The art of building and maintainingthem has distinguished a select fewstudios that get bookings partly fo rthe sound of their ch a m b e rs .

Breaking it on downR everb in all its flavo rs — p hy s i c a l

p e r formance spaces, digital effectsd ev i c e s , m e chanical resonating sys-t e m s , and acoustic ch a m b e rs—can beb ro ken down into a few para m e t e rs .But I must pre face it with this: allreverbs offer unique and subtlesonic contributions to your audiothat defy measure m e n t . Ta ke two dif-f e rent reverbs and set them to thesame patch , dialing in the same va l-ues for all their adjustable para m e-t e rs , and they’ll still sound differe n t .

No symphony hall sounds ex a c t lythe same as any other. No platesounds ex a c t ly like any other. A lway slisten for what you like; it’s just thesound of the add e d , s y n t h e s i ze dambience that matters , not thereverb time, not the algorithm, a n dc e r t a i n ly not the reverb make andmodel nu m b e r.

Reverb timeE a s i ly the most cited descriptor of

reve r b e ration is Reverb Ti m e .Sometimes called RT 6 0 , reverb timem e a s u res the number of seconds nec-e s s a ry for the sound in a room tod e c ay by 60 dB. P ra c t i c a l ly and histor-i c a l ly speaking, RT60 measures howlong a sound lingers in a room after

an impulse (e.g. a sharp clap, g u ns h o t , balloon pop, or electro n i c a l lys y n t h e s i zed click) until you can’t hearit any m o re (ro u g h ly 60 dB quieter).

Some of the most famous symphon-ic halls have reverb times ave ra g i n gjust under two seconds; opera housesex t ract better speech intelligibility byshortening reve r b e ration to just ove rone second. Digital reve r b s ,s p r i n g s ,and plates empower you to dial in anyreverb time you like . H ave fun.

SpectrumL i s t e n , in your mind, to the sound

of a room decay i n g . Cut that soundup into different fre q u e n cy ra n g e sand create a reverb time measure-ment for each spectral region ofi n t e re s t .

RT60 typically re f e rs to the decayof the octave band centered on 1 kHz.

But there is nothing stopping us fro mmeasuring the RT60 at the octavebands below and ab ove 1 kHz.

In fa c t , a rch i t e c t u ral acousticiansm e a s u re and calculate the reve r btime at all audible fre q u e n cy bands.L i ke using a tone contro l , a c o u s t i-cians design spaces with differe n treverb times at different fre q u e n-cies to satisfy musical taste, not sci-entific purity.

A c t u a l ly, halls are distinctly not flatin the spectral content of their reve r b .Halls for classical and romantic mu s i cre p e r t o i re typically have low fre q u e n-cy reverb times that are a bit longerthan the mid fre q u e n cy reverb times.This gives the halls a degree ofwarmth that seems to support the typeof music that will be played there .

You’ll see this ex p ressed inacoustics litera t u re and reverb signalp rocessor manuals as Bass Ratio.Bass Ratio mathematically compare st wo octaves of low fre q u e n cy reve r b(125 Hz and 250 Hz) to two octaves ofmid fre q u e n cy reverb (500 Hz and1000 Hz). The resulting ratio quanti-fies a hall’s wa r m t h , what we mightcall its Phatness.

Hall designers are finding whatwo rks for a Gore cki symphony and aPuccini opera . But only you know the

Real spaces alw ays have some predelay .

If they have it, why shouldn’t reverb patc h e s ?


color of reve r b e ration that wo rks fo rt o n i g h t ’s tra ck , “Insulating theA t t i c .” Experiment with the tonecolor of your reverb by adjusting itsBass Ratio if it offers one. A BassRatio of 1.2 will warm up the reve r-b e rant wash of ambience by tellingthe reverb to create a low fre q u e n cyreverb time that is 1.2 times as longas the mid fre q u e n cy reverb time.

Some reverbs don’t offer bass ra t i oc o n t ro l . Shape the color of yo u rreverb by using eq on the reve r breturns on your mixer or on the sendto the reve r b . C o n t rol the low end toa dd wa r m t h , not mu dd i n e s s . Or ifyou are going for a brighter reve r b( w hy not?), find some magic shim-mer and airiness but avoid painfulsizzle and sharp n e s s .

Predelay and Early ReflectionsB eyond the length and color of the

reve r b , t wo other fundamental pro p-erties of reve r b e ration are the time itbegins (pre d e l ay) and the timing ofthe first few bounces—singlebounces from the source to a wall tothe listener (early re f l e c t i o n s ) . If yo uuse a spring or plate reve r b , the wa s hof decay commences the instant yo u rsound starts. In a large hall (or gym-n a s i u m , or canyo n , or domed stadi-um) it takes an instant befo re thereverb begins, and there are one orm o re distinct bounces befo re thewash of reverb sets in.

By adjusting the parameter identi-fied on most devices as Pre d e l ay, wecan adjust the time gap betwe e nsound start and reverb start. P re d e l ays i m p ly inserts a delay between thed i rect sound and the reve r b e ra t i o na l g o r i t h m . In the wo rld of digitala u d i o, a dding a delay is fa i rly triv i a l ,so pre d e l ay controls are found onalmost any digital reverb dev i c e .

In the realm of analog audio, d e l ayi s n ’t so easy. Plates and springs there-fo re ra re ly give you this feature .When using a plate or spring reve r b —or a bathro o m — h ave the best of bothwo rlds by inserting a digital delay onyour reverb send so that you can adda contro l l able amount of delay befo rethe reverb begins. Tape delay is acommon feature in this role as we l l .

E a rly reflection control is commoneven on the most inex p e n s ive digitalreve r b s , and has been for a long time.

In genera l , the simplest units let yo uc o n t rol the proportion of the earlyreflections by setting their re l a t ivevo l u m e . Their pattern depends on theshape of the room yo u ’ ve selected

What good do these para m e t e rsdo? The answer is built in two wo rl d s :p hysical acoustics and psych o a-c o u s t i c s . Fi rs t , real spaces alway sh ave some amount of physical pre d e-l ay because it takes time for thesound to travel out to all the ro o mboundaries and bounce back at thel i s t e n e r, f i rst in distinct early re f l e c-tions and then in an enveloping wa s hof reve r b . If real spaces have it, w hys h o u l d n ’t reverb patch e s ?

S e c o n d ,p re d e l ay is ve ry va l u able toour personal auditory analysis system.Listen care f u l ly to the sound of a

fixed amount of reverb with and with-out pre d e l ay. As pre d e l ay separa t e sthe reve r b e rant decay from the initialsound in time, it also separates themin our mind. It is easier to hear thereverb after a bit of pre d e l ay. Wi t h o u tp re d e l ay, the direct sound masks thereve r b , making it less appare n t .

This suggests two important cours e sof action when you want a touch morereverb on a tra ck: raise the reve r bsend so that the reverb gets louder, o rlengthen the pre d e l ay so that thereverb that yo u ’ ve alre a dy put in themix becomes more audible. It isn’ta lways the case, but often you can addthe feeling of reverb by adding somep re d e l ay.This appro a ch is clever as ita dds more of the desired effect with-out adding clutter to a mix.

E a rly reflections do their part bysuggesting the rough shape of thereve r b e rant space: is it a toilet stall ora cathedral? Selecting a room shapebecomes a critical choice in cre a t i n grealism due to these re f l e c t i o n s .

N ext month we stir up these reve r bi n gre d i e n t s — ch a m b e r, p l a t e , s p r i n g ,RT 6 0 , bass ra t i o, p re d e l ay, e t c . — i n t oa few different concoctions and seeh ow they combine to create a terrificvariety of pop music flavo rs .

Even in kindergarten, Alex Case wasm o re interested in reverb time thanp l ay time. Sustain N&B articlesthrough case@re c o r d i n g m a g . c o m .

Bass Ratio = RT Lows divided by RT Mids = (RT60@125Hz + RT60@250Hz) / (RT60@500Hz + RT 6 0 @ 1 0 0 0 H z )

Excerpted from the November edition of RECORDING magazine.©2000 Music Maker Publications, Inc. Reprinted with permission.



RECORDING DECEMBER 2000

Reverb, Pa rt 2Reverb, Pa rt 2BY ALEX CASEDistilling the Options

hat does reverb sound like? Th e re are so manykinds; Fi g u re 1 breaks it down into some log i c a l

c a t e g o r i e s .So far so good. Once we learn what a hall sounds like ,

and a plate sounds like , we’ll start to master the topic ofreve r b e ra t i o n . We ’ ve got our wo rk cut out for us, t h o u g h ,p re c i s e ly because there are so many kinds. And we ’ re alldying to know what sort of reverb they used on the newTattooed Waif album, Pierce Me Here. L e t ’s break it dow n .

R everb devices in general might be bro ken down intofour broad categories: spring, p l a t e , digital reve r b e ra t o r,and special effects. We discuss the first three here , s av-ing special effects for next month.

As we discussed in last month’s Nuts & Bolts thrillr i d e , reverbs that re ly on a mechanical device like aspring or a plate to generate ambience define their ow nclass of reve r b . Th ey each have such a unique sound thatt h ey deserve a category to themselve s . Learn what theysound like and re a ch for them whenever the cre a t iveu rge hits yo u .

S t evie Ray Vaughn offers a case study on bothspring and plate throughout his debut album, Te x a sF l o o d. In genera l , his guitar hasclassic spring reverb and his vo c a lhas plate reverb—with pre d e l aythat sounds like ly to be tape-b a s e d . From the opening guitarnotes and vocal line on the firs tt u n e , ‘ L ove Struck Baby,’ theset wo classic reverb sounds maket h e m s e l ves know n . And there ’s noreason not to send the guitar tothe plate, the snare to the plate,and so on.

But that right-most category onFi g u re One, digital reve r b , is a lit-tle va g u e . When the reverb comesin a digital box , as small as half ara ck space, it becomes trickier toc l a s s i f y.

Just modifying a single reverb patch opens up a near-ly infinite set of possibilities. R everb times can ra n g ef rom maybe a couple hundred milliseconds up to 20 or30 seconds. P re d e l ay is adjustable from 0 to maybe asecond or two.

Part 1 of this series on reverb introduced a number ofreverb para m e t e rs: bass ra t i o, p re d e l ay, e q u a l i z a t i o n , f i l-t e r i n g . Wh e re do we begin?

Time & space Digital reverbs can be defined based on the size of the

a rch i t e c t u ral space they simulate: large hall or smallro o m . In betwe e n , we l l , t h e re ’s medium ro o m , big bro t h-e r ’s room (which is larger than my ro o m ) , the laundryro o m , the basement, and the gym. A a ck .

It goes on: stadium, c a nyo n , l o cker ro o m , l ive ro o m ,e t c . So we draw a line in the sand separating larg ef rom small. R everb times (RT) greater than about 1.5seconds (and they can go as high as a positive ly insane30 seconds or more) make up the ‘ l a rge’ reve r b s .N a t u ra l ly, reverb times of about 1.5 seconds and lessa re ‘ s m a l l .’

Figure 1

W


L a rge takes many names: hall,warm hall, bright hall, c a t h e d ra l , Ta jM a h a l , and such . Small includesthings like ch a m b e r, medium ro o m ,tight booth, and such . As each has itsp u rp o s e , it isn’t a bad idea to start asession with one large and one smallreverb set up and re a dy to go.

The names of the reverb pre s e t smight seem nearly meaningless; yo uk n ow they can all be adjusted toalmost any reverb time. M e d i u mro o m . RT = 1.3 seconds. I t ’s no bigdeal to change it to 2.2 seconds andc o nvert it into a hall, r i g h t ?

Not ex a c t ly. Th e re ’s a bit more to itthan reverb time. A hall sounds dif-f e rent than a ro o m . R everb designersh ave gone to the trouble to capturethose differences—the time delayb e t ween the direct sound and theonset of reve r b e ration is greater in ahall than in a room because the wa l l sa re farther apart.

And as the distance between ro o mboundaries is greater on ave rage fo ra hall than it is for a ro o m , the gen-e ral pattern and density of earlyreflections is different for a hall thana ro o m . Th e re are countless, h oweve rs u b t l e , d i f f e rences between a larg ehall and a small ro o m . Our ears (andb rain) are excellent at catching thoses u b t l e t i e s .

As a re s u l t , reverb designers go togreat trouble to capture and/or simu-late those magic little differences thatdefine a space as a hall, an operah o u s e , a medium sized ro o m , and soo n . So when you dial in a preset reve r bthat says hall, not ro o m , be assure dthat someone has taken the time to tryto capture those differe n c e s .

G o rgeous (i.e. ex p e n s ive) hall pro-grams will sometimes sound flat outbad if you shrink their ‘ s i ze’ down toro o m - l i ke dimensions. L i kew i s e ,lengthening a great sounding ro o mp a t ch to hall-like reve r b e ration willoften lead to an unnatura l , u n c o nv i n c-ing sound full of strange artifa c t s .

H aving said that, I can be pre t t ys u re you are all going try it on yo u rn ext mix. Th a t ’s okay, because mu s i cand music tech n o l ogies rewa rd thatsort of innovation and chance taking.But it’s important to know when yo ua re stre t ching boundaries and whatto look out fo r.

So what do we do with a long’ ve r b , a short ’ve r b , and so on? Th a t ’sa little bit like asking “ Wh a t ’s a Dminor 7 ch o rd fo r ? ”You use it whenit sounds right to yo u . And you canuse it when the theory supports it.

What fo l l ows is some discussion ofgood uses of different types ofreve r b s . Listen care f u l ly to re c o rd-ings you like and learn by ex a m p l e .


Try similar appro a ch e s , and armed with that ex p e r i e n c e ,c reate your own bag of reverb trick s .

Magic dustSprinkle long reverb onto a vocal or a piano or a string

pad for some hy p e , p o l i s h , and glitter. It will almost cer-t a i n ly put the ‘studio’ stamp on your re c o rd i n g , but thes l i ckness of a huger-than-huge reverb can add a bit ofp rofessionalism to the re c o rding you are trying to make .

Typical modifications to the standard large hall comecourtesy of the bass ratio control (discussed in lastm o n t h ’s column) and good ol’ equalization. Brighten it,warm it up, or both.

Bright reverbs are often a standard patch in your digitalreve r b e ration dev i c e .The slightly peculiar thing is that theyd o n ’t re a l ly exist in natural spaces. As sound travels thro u g hthe air, the highest frequencies attenuate firs t . As the soundp ro p og a t e s , it is the lowest frequencies you hear last.

You’ve heard the dominance of lowfrequencies over high frequencies ifyou’ve ever stood beside a busystreet and listened to the sound ofthe car radios leaking out of thevehicles. You can hear the thumpand rumble of the kick and bass—but not much of the rest of themusic—from one car. As for the talk

radio addict sitting in the other car nearby, it sounds alot like the teacher in Charles Schultz’ Peanuts cartoons,“Wawa waaaawuh waaa wo wo wawa waaaaa.” That’s thesound of speech that is mostly vowels (lower frequen-cies) and that lacks consonants (higher frequencies).Assound breaks out of these cars and into your neighbor-hood, the low frequencies start to dominate; the highfrequencies start to evaporate.

B e l i eve it or not, the bright reve r b , full of sizzle ands h i m m e r, is a ro ck and roll pro t e s t . It is the sound of anacoustic space that doesn’t natura l ly ex i s t . I t ’s what itwould sound like if high frequencies won out over low s .And for some applications it sounds pretty good.

Paul Simon has such good diction that, rumor has it, h eis de-essed at tra ck i n g , m i x i n g , and mastering. Using hissuper human S’s to zing a bright reverb was too intere s t-ing an effect to pass up. Listen to the down tempo songson R hythm of the Saints. A shot of high fre q u e n cy energ y

ripples through the reverb with each hard con-sonant Paul sings.

The other option, if you are n ’t brighteningthe reve r b , is to fatten it. A dding a low fre-q u e n cy bias to the sound of your long hallreverb patch adds a wa r m , r i ch foundation toyour mix. This comes closer to phy s i c a l , a rch i-t e c t u ral reality as it is often a design goal ofp e r formance halls to have the low fre q u e n cyreverb time linger a bit longer than the midf re q u e n cy reverb time. And if it’s good enoughfor Mozart, i t ’s good enough for pop.

N a t u ra l ly, we are allowed to select all of theab ove for a warm and spark ly reverb sound. B ec a re f u l , t h o u g h . If the decay of the reverb fillsthe entire spectral range of your mix, high andl ow, it will leave no room for the bass, the cy m-b a l s , the vo c a l s , the strings, and so on.D ivvying up the spectral real estate is a con-stant challenge in pop music mixing. A n d

while it might always be tempting to use a full-band-width reverb that sings across the entire audible spec-t r u m , it can be wiser to limit the harmonic ‘ s i ze’ of thesound of the reverb and assemble a full mu l t i t ra cka r rangement that, in sum, fills the spectral landscape.

The third principal va r i able after reverb time andreverb tone is pre d e l ay. That gap in time between when asound begins and when a physical space is energ i zed andstarts reve r b e rating is an excellent parameter to manipu-l a t e . To change it in physical space re q u i res moving wa l l sand raising ceilings.

The results are ethere a l .Think ballad. Start with a longreverb preset on a vo i c e ,m aybe the “ O o o h ” or “A a a a h ” of ab a ck ground vo c a l . Listen care f u l ly as you stre t ch the pre d e-l ay from maybe 20 milliseconds to 40 milliseconds, 60 mil-l i e s e c o n d s , on out to 100 milliseconds or more .The feelingof reve r b e ration certainly increases as you lengthen thep re d e l ay. So does the feeling of distance and loneliness.

H e re we ’ ve stumbled onto one the most intere s t i n gparts of the re c o rding cra f t . By manipulating pre d e l ay,w h i ch is a va r i able in the studio (but not in the operahouse) we ’ ve created the feeling of a longer reve r bFigure 2


without lengthening the reve r b . If it sounds like we getto violate the laws of physics and arch i t e c t u re in thes t u d i o, i t ’s because we do.

If yo u ’ ve ever suffered from a mix that became ove rlyc row d e d , c o n f u s i n g , and messy as all the tra cks andeffects we re add e d , you may wish to remember this: pre-d e l ay can be used to separate the reverb tail from thed i rect sound by a little ex t ra bit of time. This slight sepa-ration makes the reve r b e ration easier to hear. The re s u l tis the addition of ex t ra reverb in feeling, without theactual addition of mix-mu ddying ex t ra reverb in re a l i t y.

Far outA dding reverb to some tra cks is like adding garlic to

some sauces: yum. S o m e t i m e s , t h o u g h , we are a littlem o re strategic in our motivations to use reve r b .

With the help of Fi g u re 2, p i c t u re in your mind’s earsthe sound of a vo i c e over you just re c o rded in your studio.For this example we close-miked the talent in a re l a t ive lydead ro o m .P l ay back the tra ck and you hear, we l l , t h esound of that person speaking, and he or she sounds near-by. R e c o rded by a microphone may b esix inches away from the vo i c e ove ra r t i s t , it isn’t surprising that the vo i c esounds close and intimate.

N ow add a good dose of reve r b(hall-type patch with a reverb timeof about 2.0 seconds). Pe rc e p t u a l ly,the voice now sounds more distant.The loudspeake rs didn’t move , b u tour image of the sound coming out ofthem sure did. As we use pan pots tolocate discrete tra cks of audio left tor i g h t , we use reverb to locate ele-ments of the music front to back .Your mixes take on an unreal depthas you master this tech n i q u e .

GelThe sound of the immediate space

a round a band can be ve ry evo c a t iveo f, u m , the band in a ro o m . C o m m o non drums and almost any section( s t r i n g s , h o r n s , ch o i r, k a zo o s ) , ro o mambience can help unite 32 tra cks ofove rdubs into a single, c o m p e l l i n gw h o l e . Dial in a room patch with areverb time of about 1.3 seconds orless and start gluing tra cks tog e t h e r.The trombone lines that we re re c o rd-ed two months and and two hundre dmiles away from the original saxo-phone parts will fall into the mix.

As reverb gets this short, it is timeto ask ours e l ves “ Why synthesize it?”R e c o rding studios, l a rge liv i n gro o m s , c o nverted gara g e s , and re n o-vated barns can make a contributionto the sound you are re c o rd i n g . I tm a kes sense, t h e re fo re , to re c o rd it.

When you find yo u rself noticingand liking the ambient sound of aro o m , c a p t u re it in your re c o rd i n g .Two appro a ches: place micro p h o n e sso as to capture a satisfying blendof the instrument and the ro o m , o rplace microphones to just plainc a p t u re the ro o m .

The first appro a ch is one of thej oys of re c o rd i n g . To re c o rd the music and the ro o m , yo uabandon the pop music tradition of close miking andstart re c o rding instruments from a distance. A m b i e n tmiking appro a ches abound and are a topic of an upcom-ing Nuts & Bolts column.

It is worth mentioning that this ain’t easy. To pull them i c rophones away from the instrument is to ab a n d o nsome control and consistency in our re c o rding cra f t .

Perhaps you’ve recorded your husband’s ukulele a mil-lion times and know exactly where to put the mic to cap-ture the sweet ukulele tone that always satisfies yourclients. You’ve worked hard to find that perfect micplacement location that works anytime, anywhere, anygig. It is no doubt a mic position placed very close to theinstrument—so close to the ukulele itself that it‘ignores’ the sound of the room. There is comfort in theclose miking approach.

But exploring ambient miking techniques will pay div i-d e n d s , sometimes setting the vibe for the entire tune.Capturing those tra cks re q u i res ex p e r i e n c e , q u a l i t ye q u i p m e n t , and good acoustics—and a bit of good luck

d o e s n ’t hurt. E x p l o re this path onlywhen a project has the time andc re a t ive motivation to do so.

The second, s l i g h t ly safer optionfor capturing actual acousticreverb instead of simulating it isto re c o rd the ambience of theroom onto separate tra ck s . Place amic or two any w h e re in theroom—the other side of the ro o m ,on the floor, at the ceiling, in ac l o s e t , d own the hall... R e c o rd theroom in a location you think offersa musical contribution to thesound of the instruments.

Of course you need spare tra ck sfor this, but it enables you toc l o s e - m i ke the instruments as yo ua lways have and to capture someof the sound of the room too. Yo um ay end up with the opportunityto create unique sounds on mix-d ow n .

S p r i n g , P l a t e , L a rge Hall, a n dSmall Room. Those are the obv i o u sreve r b s . And they offer a limitlessset of sonic possibilities. N ex tmonth we’ll look at the moreb i z a r re reverb tactics: to reve rse it,distort it, c o m p ress it, and whok n ows what else. H o p e f u l ly theaudio police wo n ’t pull us ove r.

Alex Case wonders: before Reverb,does it just Verb? Offer help viac a s e @ re c o r d i n g m a g . c o m .

TechniqueMajor warning: a classic mistake that inexperienced re c o rd i n gengineers make is to add too much reverb. For me, learn i n ghow to use reverb was a little bit like when I learned to makechocolate milk a couple (maybe more) years ago. On the sec-ond try (without mom watching) I doubled the recipe. On thet h i rd try (sorry mom), the chocolate to milk ratio went decid-edly in favor of chocolate (who needs the milk part anyway?).Such is the life of a kid.This ‘more is better’ approach to life might work for chocolatemilk, but it doesn’t work for reverb. Too much of a good thingsounds cheap and poorly produced. It’s literally the calling cardof a young engineer. D o n ’t sweat it, though. Reverb will fool you the first few times,but here ’s how to outsmart it. Do a mix and add as muchreverb as you want. Don’t hold back. Tu rn up the reverb untilyou hear it and like it. Print the mix. Three days later, listen tothe mix. There ’s nothing like the passage of time to clear ourears and let us hear things as we’ve never heard them before .You’ll say “What was I thinking?” as your mix swims in re v e r-berant ooze. We’ve all been there. It’s pretty fascinating that we could be in the studio, leaninginto the speakers, ears wide open, adding what sounds like anappropriate amount of reverb only to discover, well, oops. It’ssomething of an audio illusion. The more you listen for it, theharder it is to hear it. You get control of the reverb (and othereffects) in your tracks only when you learn to listen confi-dently. Relaxed, you’ll hear everything you need to hear, and,with experience you’ll know how to adjust the equipmentaccordingly.The fact is, reverb is something we have to learn to hear. Formost humans reverb is not a variable, it is a fact. Our hearingsystem hasn’t evolved with the concept that reverberation isadjustable. Recording engineers must discover and develop thisa b i l i t y. So much of audio (especially compression and equal-ization) is this way. Give yourself the chance to learn by mak-ing some fat, juicy mistakes!

Excerpted from the December edition of RECORDING magazine. ©2000 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119 For Subscription Information, call: 1-800-582-8326


Reverb, Pa rt 3Reverb, Pa rt 3BY ALEX CASEReverb on the edge

e know that on most sessions, a dding reverb to at ra ck is usually a stra i g h t fo r wa rd task. We have

an almost infinite range of conventional solutions tochoose from—but there are also some long-standing andu nusual studio reverb concoctions wo r t hy of study. We ’ l lstart with back wa rds reve r b , l ov i n g ly called “ b reve R .”

A n a l og tape machines rewa rd ex p l o ra t i o n . Tape can bec u t , s p l i c e d , sped up, s l owed dow n , a n d — ye s — p l aye db a ck wa rd s . Try it. Put your mu l t i t ra ck tape on upsided own (swapping the supply and take-up reels) and roll it.It wo n ’t hurt the tape or the tape mach i n e . And—as Jimi,the Beatles, M i chael Pe n n , and others have show n — i tcan sound pretty cool indeed.

H e re ’s the reverb part. With the mu l t i t ra ck tape play-ing back wa rd s , a dd and re c o rd some reve r b . Fi rst andmost important find an empty tra ck . Be ve ry, ve ry surei t ’s empty. This isn’t easy when yo u ’ ve flipped the tapeover for reve rse play.

If you have an 8-tra ck analog mu l t i t ra ck re c o rd e r, t ra ckone is on top and tra ck eight is on the bottom. Tra ck oneof the tape mach i n e , w h i ch you are pro b ab ly monitoringon channel one of your mixer, is actually playing backt ra ck eight off tape. Tra ck two moves to seve n , t h ree tos i x , and so on.

If you have the privilege of using an analog 24-tra ckmu t i t ra ck tape machine it gets even more confusing. A n dit isn’t easy to identify tra cks just by pushing up thefa d e rs and listening. K i ck , s n a re , b a s s , and piano soundclose to what you might ex p e c t . But it is darn difficult toidentify vo c a l s . Is this take one, t a ke two, or what?

For reve rse effects, I tempora r i ly label the tra ck sheetwith the new tra ck nu m b e rs by manu a l ly starting at thehighest tra ck , l abeling it Reve rse Tra ck One, and countingup from there .Then my tra ck sheet makes clear that vo c a lt a ke two on tra ck 17 will appear back wa rds on tra ck 8.

Once you know ex a c t ly what tra ck you are going tou s e , push up the source signal fa d e r. Use an aux send toget it into your reverb of ch o i c e . And re c o rd the outputof your reverb to the empty tra ck ( s ) . A good starting

point is to use an instrument prevalent throughout thes o n g , s ay a snare tra ck or vo c a l .

M aybe the singer sang “La la, B aby.” P l ayed back wa rd syou hear the nonsensical, “ybaB al aL.” A dd ’verb andt h e re is a decaying sound after each back wa rds wo rd .Print that reve r b .

N ow the fun part: flip the tape back over and play themu l t i t ra ck as originally re c o rd e d . The tra n s c e n d e n t a lline is re s t o red: “La la, B aby.” But push up the fa d e rsc o n t rolling the back wa rds reverb you just re c o rd e d , a n da we i rd ‘this doesn’t happen in nature’ sort of thing hap-p e n s . The decay now comes befo re the wo rd that causedi t . R eve rse reverb is an effect that stra n g e ly anticipatesthe sound about to happen.

Fi g u re One shows what’s going on. For simplicity weconsider a basic snare back beat falling on beats two andfour (Fi g u re 1a). In fo r wa rd play, the reverb you addd e c ays after each hit of the snare (Fi g u re 1b). This cre-ates the expected combination a dry, close miked snareplus reverb (Fi g u re 1c). Th a t ’s the typical appro a ch . Th i stype of reverb adds a natural ambience or perhaps ahyped ex p l o s iveness to the mix.

L e t ’s fo l l ow these same steps for brever (i.e. reve rs ereve r b ) . When playing the tape back wa rd s , we observeour snare hitting on beats four and two (Fi g u re 1d).Th a t ’s the same back beat, o n ly back wa rd s . R e c o rdsome reverb from this back wa rds playing snare (Fi g u re1 e ) . Return to normal, fo r wa rd play and ch e ck out howthe back wa rds reverb now occurs befo re each snare hit( Fi g u re 1f).

This elab o rate process is tedious and more than a littledisorienting at firs t .D o n ’t experiment with this for thef i rst time in a high pre s s u re session in front of your mostdifficult client. And definitely don’t attempt this at 3 a.m.after an 18-hour session. The risk of accidentally erasing at ra ck while re c o rding on an upside-down reel is too gre a t .But after some practice on other sessions or on your ow nmu s i c , you’ll be able to re a ch for this appro a ch comfo r t-ab ly and add a bit of uniqueness to part of the pro j e c t .

W


It takes a fair amount of trialand error to get the effect yo uwa n t . I t ’s hard to predict how itwill sound when you dial in areverb while the vocal sings gib-b e r i s h , “htnom yreve stlob dnastun daer.” I t ’s not until you re c o rdthe reverb and play it back fo r-wa rds that you can re a l ly tell ifyou like the reverb type, reve r bt i m e , p re d e l ay, bass ra t i o, e t c . U s e dc a re f u l ly and sparingly you canoffer your listeners a wild ride.( This wo rks with ech o, t o o, by theway — ch e ck out the incoming/out-going vocal echo effects on “It CanH a p p e n ” on 9 0 1 2 5 by Ye s . )

Variations on the themeB a ck wa rd s - l i ke reverb effects

appear as presets on some reve r bd ev i c e s . Often called ‘ n o n - l i n e a r ’reve r b s , these reverbs don’t decayf rom loud to soft after a sound. I nfa c t , t h ey do the opposite. I n s t e a dof getting gently softer as theyd e c ay, non-linear reverbs get louderas they decay.

S ay what? I know it’s we i rd . S i n c edigital reverbs are controlled bys o f t wa re , not room acoustics, t h eycan do some pretty bizarre , n o n -i n t u i t ive things. A regular decay i n greverb can be compressed andamplitude-modulated (with a singlecycle of a sawtooth wave) as show nc o n c e p t u a l ly in Fi g u re 2, making areverb swell soft to loud. Pa t ch thisup or look for a preset in your digi-tal reverb to create this effect.

Of cours e , you can use non-linearreverb wherever you like , but lookf i rst at percussion instruments inpop music settings. The sound of ac o n g a ,t r i a n g l e , c l ave , or other sharpp e rcussion instrument lasts merem i l l i s e c o n d s . It is a mixing ch a l-lenge to make such a short wave-form noticeable in a crowded popmix full of synths, s t r i n g s , g u i t a rs ,and laye rs of back ground vo c a l s .

Use the non-linear ’verb tolengthen the perc e ived duration ofthe percussion event slightly, m a k-ing it easier to hear and there fo reeasier to slide into the mix. A heav ydose of the non-linear reve r bsounds like a wa cky effect—some-times appro p r i a t e , sometimes not. Asubtle dose can retain the natura l-ness of the instrument and stillaccomplish the mix goal of gettingthe sound noticed. C reate the soundyou like best for the tune at hand.

P l aying tape back wa rds to cre a t ereverb that in turn is played fo r-wa rds is a lot of tro u b l e . Ta p em a chine manu fa c t u re rs have some-times built in the ability to play and

Figure 1


re c o rd back wa rds to make this ex e rcise a little easier.But armed with a sampler or digital audio editor, you canre c o rd , reve rs e , c u t , and paste with ease. All your effectsunits just doubled the number of patches they have . Yo ucan sample them and play them back wa rd s .

Squished reverbThe inex p l i c able magic of the delicate decay of sound

within an ornate European music performance hall alsoresponds well to—I’m serious here — c o m p re s s i o n . Whythe heck not? We discussed the Nuts & Bolts ofC o m p ression back in Part 9, 3 / 2 0 0 0 . Using compre s s i o nto alter the way a wave form attacks and decays is oldh a t . R everb is the decay of a sound. C o m p ressing reve r be n ables you to change the decay of this decay. As thec o m p ressor changes the amplitude of the reve r b e ra n twa s h , the musical impact of the reverb changes too.

For ex a m p l e , it is perfectly normal to re c o rd a tam-bourine in a dry (i.e. no natural reve r b e ration) booth,b e d ro o m , or basement. No pro b l e m .A dd some bright hallto it at mixdow n , right? So far so good.

But maybe yo u ’ ve experienced the problem of a dis-t a n t , weak reve r b . That is, a dding reverb to a ro ck ’n ro l ltambourine diminishes the impact of the perc u s s i o ni n s t r u m e n t , a dding distance between the tambourineand your listener.

This isn’t surprising.As we discussed last month, wesometimes use reverb with the intent of pushing a par-ticular sound farther back toward the sonic horizon.Adding reverb to our tambourine can rob it of its power,sliding its contribution to the groove away from therhythm section and away from the listener.

Slamming drums, huge bass, a wall of guitars, scream-ing vocals... and that dude way back over there tappinghis tambourine. Not so compelling, as rock and rollstatements go.

C o m p ression to the re s c u e . Slam the reverb through ac o m p re s s o r, and it turns into an entire ly new kind ofs o u n d . L ow thre s h o l d , high ra t i o, fast releasing compre s-sion changes reverb into a burst of noise and energ yassociated with eve ry hit of the tambourine (or slam ofthe snare , or strum of the guitar, . . . ) .

With apologies to the engineers who so care f u l ly fig-u red out how to digitally simulate the sound of that gor-geous symphony hall, squish it hard with compre s s i o n .Change the sound of your dry tambourine into a driv i n g ,gro ov i n g , a g i t a t i n g , i n - yo u r- face tambourine surro u n d e dby the surg i n g , d i s t o r t i n g , fizzling sonic aura of com-p ressed reve r b .

Gated reverbSend the snare drum to an aggre s s ive ly compre s s e d ,

ve ry long reverb patch (maybe a plate program modifiedto a ridiculous reverb time of five seconds or so) and yo u

can create a bed of noise that seems never to decay. E a chs n a re hit re - e n e rg i zes the reve r b . The long reverb timea l t e red by heavy compression makes sure the sound lastsand lasts.

Do this in a mix, and you’ll find that after snare hitnumber one it is no longer possible to hear the guitars oru n d e rstand the vo c a l s . Bad new s . This reverb takes ove r,o b l i t e rating all delicate elements of your arra n g e m e n tthat dare to come near it. The reverb essentially becomesa new, loud noise floor.

Figure 2

Figure 3: Changing the Shape of the Delay


Seems a little irrational to addnoise to a mix, d o e s n ’t it? Yu p. S owhen you add this mu ch noise to am i x , also use a noise gate. Gates getrid of the noise. Fi rst we add ani n s a n e ly long reverb to the mix. Th e nwe compress it to bring the level ofthat reverb/noise up. And finally wea dd a noise gate to get rid of most ofthe wa cky reverb we cre a t e d .

The result is a gated reve r b , s h ow nin Fi g u re 3. The snare drum hits. Th enoise gate opens up (triggered bythe snare ) . The burst of reverb com-m e n c e s . An instant later (at a timeset by you on the gate) the noisegate closes. The noise goes away,revealing those other elements ofthe mix (ya know, l i ke the vo c a l s ) .The snare hits again. R e p e a t .

A dding a gated burst of reve r b e ra-tion through this fa i rly elab o rate sig-nal path can convert a wimpy snareblip into the powerful snare of God.I t ’s common to set the gated reve r bto a musical note va l u e — m aybe giv-ing the decay on the snare a dottedeighth note time feel, for ex a m p l e .Then the gated reverb isn’t just loudand energ e t i c , i t ’s also gro oving hard .

E x p l o re compressed and gatedreverb and you’ll see how the non-lin-ear reverb patches we discussedab ove are cre a t e d . Th ey don’t playreve r b e ration back wa rd s , t h ey justa g gre s s ive ly manipulate the loudnessof the decaying reverb over time.

As Fi g u re 4 show s , t h e re is a lot top a t ch up to make it wo rk . It alsot a kes time to tweak it into contro l .Yo u ’ ve got to find a good soundingreve r b . Gated reverb ra re ly soundsn a t u ra l , so you are free to chose awild sounding reverb patch to startwith; skip the swe e t , high fidelityones and go for the row dy, out of thiswo rld sounds.

N ex t , yo u ’ ve got to dial in the rightamount of compre s s i o n . Set thet h reshold well below the level of theinitial burst of reverb so that thec o m p ressor is still attenuating thesignal well after the initial snare

sound hits. Dial in a ve ry fast re l e a s etime so that the compressor pulls-upthe sonic detail of the decaying tailof the reve r b .

Fi n a l ly, h a rdest of all, yo u ’ ve gotto get the noise gate to cooperate sothat it opens only on the snare . I fi t ’s MIDI tra cks yo u ’ re using, i t ’sp retty stra i g h t fo r wa rd to find thet h re s h o l d , a t t a ck , h o l d , and re l e a s etimes for the noise gate that makemusical sense.

If yo u ’ re using live drum tra ck s ,the trick is to make sure the gatei s n ’t fooled into opening when othern e a r by instruments play — l i ke thek i ck or the hi-hat that might be leak-ing into the snare mic. Often a sim-ple filter set lets you re m ove thosesounds that are mostly lower (e.g.

k i ck drum) or mostly higher (e.g. h i -hat) than the instrument your areusing to open the gate (e.g. s n a red r u m ) . Filter out the lows of the kickand the highs of the hat that leake dinto the snare signal you are using totrigger the gate, and you’ll be able toget the gate to coopera t e .

And what’s good enough for artifi-cial reve r b e ration is good enough fo rn a t u ral reve r b e ra t i o n . If you havere c o rded some natural room soundon to other tra cks during the session,remember it will respond well toc o m p ression and gating too.

In the end, reverb isn’t an effect.I t ’s a fa m i ly of effects—some obv i-o u s , some not so obv i o u s . It rewa rd sthose who take the presets in differ-ent directions and those who dareto combine it with some eq, c o m-p re s s i o n , g a t i n g , d e l ay, f l a n g i n g , d i s-t o r t i o n , and so on. Th e re are nob o u n d a r i e s .

Alex Case (case@re c o r d i n g m a g . c o m )is an arch i t e c t u ral acoustician atC avanaugh Tocci Associates in NewE n g l a n d .D o n ’t tell his boss wh at ’s inthis article—especially the back w a r d sreverb part. Th a n k s .

Figure 4: Signal Flow

Excerpted from the January edition of RECORDING magazine.©2001 Music Maker Publications, Inc. Reprinted with permission.




Studio Documentat i o n ,Studio Documentat i o n ,BY ALEX CASEPart 1—The Track Sheet

the next two columns we'll look at ways to docu-ment eve ry detail of each studio pro j e c t . Ta ke

s h e e t s , setup sheets, and recall sheets are all useful partsof the well-documented studio, and we'll get to thosen ext month. This month we begin with the best-known ofall studio documents, the tra ck sheet.

Identifying tracksThe tra ck sheet's most obvious and vital function:

identifying what's been re c o rded on which tra ck s .What's on tra ck 1? "Hi-hat." What's on tra ck 19?" B a ck ground vocal #3—LowPart." This labeling must bedone so meticulously that to seean empty space on a tra ck sheetis to know with 100% certaintythat it is a blank tra ck ava i l ab l efor re c o rd i n g .

Then there is other info r m a-tion that belongs on a tra cks h e e t , m o s t ly fa i rly obv i o u sitems that nevertheless some-times are omitted. What good isit to know that tra ck 1 containsthe hi-hat when you can't tell what the song is? So startfilling in your tra ck sheet by writing down the songt i t l e . Don't leave it at that—list the artist, p ro d u c e r,e n g i n e e r, and assistant engineer. On the off-chance thatthe tra ck sheet gets separated from the mu l t i t ra cktape—something that should never happen—all thisi n formation will come in handy.

If you are the studio, the engineer, the pro d u c e r, a n d / o rthe artist, put a phone nu m b e r, email add re s s , or both oneve ry single document having anything at all to do withthe pro j e c t . M a ke it easy for anyone who finds the docu-ment to find yo u .

You can buy blank tape for $X. But once you startputting music and studio time on tape, that tape quick ly

becomes nearly priceless, l i t e ra l ly and figura t ive ly. B yincluding all of this information you minimize the ch a n c eof losing your inve s t m e n t .

All this is important, but the point of this article isthe not-so-apparent information that should beincluded on each and eve ry tra ck sheet. Of cours e ,not eve ry project is re c o rded on tape, let alone ana-l og tape, and digital audio wo rkstations take care ofa lot of the housekeeping for yo u . But the centra lconcepts should be obvious enough that you cana p p ly them to other media.

How fast was I going,Officer?It is essential that the play-

back speed of the tape be clear-ly indicated. Can you actuallyplay back a tape at the wrongspeed? Yep. Does it really everhappen? You betcha.

On analog mach i n e s , t h a tmeans noting the speed ini n ches per second (ips). Ty p i c a lspeeds are 7-1/2 ips, 15 ips, a n d30 ips. G e n e ra l ly speaking, t h e

faster tape speeds lead to increased dynamic ra n g e .But rolling tape at faster speeds also leads to highertape costs—each tick up in speed will double yo u rtape costs. If the project is on a tight budget or if theband is long-winded and aiming for a double album,this can be a big deal. You make this decision befo rethe first session, and then you document it on eve ryt ra ck sheet.

Th e re is a similar parameter on digital tape and harddisk re c o rd e rs: sample ra t e , w h i ch must be noted (ink H z ) . Most common are 44.1 kHz, 48 kHz, and incre a s-i n g ly 96 kHz. As with tape speed, higher sample ra t e sa rg u ab ly lead to better sounding master re c o rd i n g s . B u tthe higher sample rates re q u i re more tape or hard disk

I n

the point of this article

is the not-so-apparent

information that should be

included on each

and every track sheet.


space to store the increased data.The machine will usually know atonce what the sample rate is, b u tyou don’t—so if you need to matchrates from tape to tape (or disk tod i s k ) , write it dow n .

Are you my master?Note on Fi g u re 1 that the tape

m a chine used is identified (justab ove the sample ra t e ) . I can’tove re m p h a s i ze this point: alway s ,a lways note the make and modelnumber of the machine that cre a t e sa ny master tape—be it 24-tra ck , 8 -t ra ck , or even 2-tra ck . This not onlyidentifies the format (um, it wo n ’t fitin an A DAT-type mach i n e ) , it alsoidentifies the specific model nu m b e r.

In a perfect wo rld this wo u l d n ’t ben e c e s s a ry. All tapes played on allcompatible tape machines wo u l dp e r form without a hitch . Bad new s :i t ’s not ex a c t ly a perfect wo rl d .Sometimes a tape re c o rded on onem a chine wo n ’t play back on anotherm a chine without glitch e s . If yo ukeep tra ck of the type of mach i n eu s e d , you can lower the odds thatthis problem will haunt yo u .

When the tape wo n ’t play on Bob’sm a ch i n e , it may be because it is ad i f f e rent model. Fi n d , re n t , or bor-row a machine of the same make andmodel originally used during tra ck-ing and the tape might play backagain without erro rs and drop outs.Sometimes the only solution is to gob a ck to the original re c o rd i n gm a chine itself. In this case, m a ke asafety copy onto a different mach i n eas soon as you can. And be thankfulyour tra ck sheet identified thes o u rce mach i n e .

In the analog tape machine wo rl d ,identifying the source machine is

Can you actually

play back a tape at

the wrong speed?

Yep. Does it

really ever happen?

You betcha.


a rg u ab ly even more important.A n a l og tapes will play back fine onmost any type of mach i n e . The dra-matic muting on and off and the sig-n a t u re ‘zipper’ noise that only digitalre c o rdings gone wrong can makewo n ’t dog your analog pro j e c t .

But analog re c o rdings genera l lysound best when played back on thesame type of machine that did there c o rd i n g . Mastering studios oftenh ave seve ral different makes of analogtape machines for this re a s o n . Th eycan match the same make and model

you used to get the best sound offtape possible. Or the mastering engi-neer can resort to a different tapem a chine on purpose (not by accident)to find a different sound. As you cans e e , noting the tape machine used is agood idea.

Figure 1


How ’bout a date?Note the date of the first tra ck i n g

on the tra ck sheet. As you get intoove rdubs later, c a p t u re the date ofthose individual tra cks too. H av i n gthe date can help you hunt down andidentify pro b l e m s . Months aftermaking these re c o rd i n g s , you willstart mixing them. You may notice atm i x d own that the acoustic guitarsounds brighter in one song than ina n o t h e r. A little investigation reve a l sthat the strings we re brand new onone song, and two days of heav yp l aying older on another.

This is an important observa t i o n .When you start mixing a third song,you can glance at the date of theacoustic guitar ove rdub and knowb e fo re listening whether you have abright or dull tone to start with.

The date of each tra ck can answe ra range of other, similar questions:

For a given piano tra ck ,h ow longhad it been since the piano was tuned?

Was this backing vocal cut befo reor after she had her cold?

Was that tra ck re c o rded befo re orafter we cleaned the heads on themu l t i t ra ck ?

The dates essentially provide anaudit tra i l , should you want toa n swer some of these kinds of ques-tions as sonic peculiarities unfo l d . I tis quite possible you’ll never needthe dates. Keep tra ck of them justin case. Some problems are darn

subtle and might go unnoticed fo rd ay s , we e k s , or even months. B u tonce you discover that the pedal onthe kick drum has developed a fa i n tbut powe r f u l ly annoying squeak,you’ll want to figure out when inthe course of the project this start-e d , what songs might need fixing,and which ones are safe.

Signal pathAs you can see from the hiero-

g lyphs on Fi g u re One, we squeezestill more information onto thet ra ck sheet. I d e a l ly, t ry to describethe settings of each piece of gear inthe signal path.

The Lead Vocal on Tra ck 24 offersa good ex a m p l e . This particular ove r-dub was re c o rded through an A K G414 in cardioid pattern, without ap a d , and without a ro l l - o f f. Th em i c rophone preamp settings andc o m p ressor settings are shown too.G ra n t e d , it is shown in a ve ry abb re-viated fo r m , but it tells me what Ineed to know. D evelop your ow ndetailed code.

If eq had been used, I ’d turn thet ra ck sheet over and make notest h e re too. Should we have to re -t ra ck part of the vo c a l — w h i chcould easily happen: the songwriterchanges a line, the singer wants tochange the phra s i n g , a prev i o u s lyunnoticed mistake now seemsu n b e a rable and must be fixed—we’ll be able to match the soundp retty closely and re c o rd anychanges we wish. The entire signalpath has been documented. M a t chthose settings on the equipment,let the singer do a few takes tom a t ch his or her earlier perfo r-m a n c e , and you are re a dy to re -re c o rd any or all of the vocal tra ck .

As you can see, for this session Ia lways documented the vocal tra ck sf u l ly.That is standard operating pro c e-d u re; the vocal tra cks are importantenough to demand it.The tambourinet ra ck , on the other hand, o n ly indi-cates the mic and date. I’m not re a l lyworried that I’ll have to modify apiece of this tra ck . Noting the micreminds me of what sort of sound wewe re going fo r, and I can get closeenough to that sound again if need be.

The electric guitar (noted EGT ont ra ck 10) needs a fuller description.The guitarist brought in maybe halfa dozen guitars , and two amps.M o re ove r, the studio has five guitarsand three other amps. The tra cksheet there fo re notes the guitar, t h ea m p, the micro p h o n e s , and any sig-nal processing going on.

Of cours e , guitarists do a lot toshape their sound through the va r i-ous tone and pick-up settings on thei n s t r u m e n t , as well as the many set-tings on the amp and any stompb oxes being used. This gets tricky.Most guitarists I’ve had the pleasureof wo rking with have given a lot ofthought to their tone. Th ey ’ vemapped out all these settings fo re a ch and eve ry song they tra ck . Th eycan dial them up consistently with-out writing them dow n . In this case, Ilet them keep tra ck of their settingson the guitar rig mentally, and theassistant and I make notes of our set-tings in the studio manu a l ly.

Less experienced guitarists mightneed you to capture their settingst o o. This can slow down a session sig-n i f i c a n t ly, e s p e c i a l ly if you don’t

Table 1:Some suggested abbreviations (make up and use your own):

Kick Drum --------------------K Snare Drum----------------------------Sn

Hi-Hat--------------------------HH Drum Overhead Microphones------O/H

Rack Tom 1 ------------------R1 Rack Tom 2 ----------------------------R2

Floor Tom----------------------Fl Acoustic Guitar ------------------------AGT

Electric Guitar ----------------EGT Piano------------------------------------PNO

Tambourine ------------------Tambo Lead Vocal------------------------------LV

Background Vocals ----------BGV Double ----------------------------------DBL

Do Not Use--------------------DNU Do Not Erase --------------------------DNE

To Be Erased------------------TBE Serve Pickles Often ------------------SPO

In no time you’ll have six tracks

dedicated to the guitar solo,

and a dozen tracks for alternative,

possible, ‘I think so’ lead vocal tracks.


h ave an assistant engineer. In theses i t u a t i o n s , I encourage the guitaristto wo rk with the guitar tone contro l sset to wide open (turned all the wayup so that the tone controls are n ’tshaping the signal). This typicallyleads to a better tone any way, and it

is easier to repeat at a later date.The various settings on the amp

a re manu a l ly transcribed onto asheet of paper. This sort of note tak-ing in a session will be discussed indetail next month. But be fo re-warned: it is often necessary to write

d own the settings of guitar amps,c o m p re s s o rs , e q u a l i ze rs , e t c . Th e s enotes are taken on a specialized stu-dio document called a recall sheet.This enables you to, you guessed it,recall any studio setting that yo um ay have re c o rd e d .

Figure 2


On the tra ck sheet we ’ ve noted agood deal of information about thep roject and each individual tra ckre c o rd e d . When there isn’t room todocument the entire signal path for ag iven tra ck , we turn the tra ck sheetover or re a ch for recall sheets. In thisway, we have the paper-based supporti n formation needed for eve ry bit ofaudio we are putting on tape or disk.

For the session shown in Fi g u re One,e a ch drum tra ck has recall sheets asso-ciated with it documenting the settingsof the equipment we used that day

during basics. So in the end, the tra cksheet only shows settings for selectove rd u b s .Th e re are pages and pages ofnotes (not shown here of course) asso-ciated with the other tra ck s .

A good tra ck sheet presents thisi n formation clearly. Try not only towrite neatly, but also to wo rk visual-ly. I use symbols and abb rev i a t i o n sw h e rever possible to make the docu-ment easier to re a d . By the end ofthe pro j e c t , i t ’s going to be a crow d-ed piece of paper—Fi g u re 1 make sthis ab u n d a n t ly clear.

But for the engineer, the pro d u c e r,and anyone else who wo rks with us,the document communicates a lot ofi n formation in a little space. Table 1highlights some of the commonabb reviations I use. M a ny engineersdoodle cartoon-like pictures for cer-tain instruments. C reate your ow ns y s t e m . Just make sure you can re a dit in low light and write it quick lyduring more frenzied sessions.

Don’t fear erasureDuring an ove rdub session, it is

p retty common to re c o rd mu l t i p l et a kes of the same part. EGT Solo,t a ke one. Sounds gre a t . S ave it ont ra ck 11 and do another take ont ra ck 12. Ta ke two is just okay. E ra s eit by re c o rding over it. R e c o rd ana dditional take over again on tra ck1 2 . The next take onto tra ck 13, t h e nt ra ck 14. On and on it goes untileve ryone is satisfied that they havethe killer take of the unive rs e .

S ave the good takes and wipe (i.e.e rase and re c o rd over) the so-so

you’re now talking

about dozens—

or hundreds!—

of takes...


t a ke s . This process is often an effec-t ive way to capture a guitarist’s bests o l o. The reason this is important isthat unlike live perfo r m a n c e s , w h i chdisappear as soon as they occur, are c o rded performance must stand upto repeated play i n g . It is essential tothe success of the re c o rding that lis-t e n e rs continue to enjoy the soloeven after they ’ ve heard it on theradio 17 times. And the psych o - l oya lfans are going to copy, t ra n s c r i b e ,and critique the performance noteby note, string bend by string bend.

Collecting takes onto differe n tt ra cks is a decent appro a ch . You caneven edit together the best parts ofvarious takes into a single meta-solo.H owever—and this is ve ry, ve ryimportant—the process is a total fa i l-u re if you don’t take the necessarysecond step.

Step Two: after you select the ke e p-er take , get rid of the others . Filling upthe mu l t i t ra ck with ‘safety solos’ thatyou are afraid to erase will come backto haunt yo u . In no time you’ll have sixt ra cks dedicated to the guitar solo, a n da dozen tra cks for alternative ,p o s s i b l e ,‘I think so’ lead vocal tra ck s .

This doesn’t leave room for theother elements of the arra n g e m e n t .The featured perfo r m e r, the engineer,the band, and/or the producer shouldcommit to the take as soon as possi-b l e . I recommend designating thefavorite solo right there at the ove r-dub session. At my most genero u s , Imight let the band think about it andlisten to it ove r n i g h t . But the next ses-sion begins with a designation of thekeeper tra ck , and all the others getl abeled TBE (“to be era s e d ” ) .

It is pretty common during a pro-

ject to inv i t e / h i re a special guest tosing or play across a number of tuneson the album. Yo u ’ ve got may b ee l even different songs. In the cours eof this ove rdub session the guest tal-ent flies from one song to the nex t .“Nice take , l e t ’s try that sort of thingon the other ballad.” Yo u ’ ve got tozip to the next song, pull-up a gre a tsounding rough mix in the contro lro o m , dial up a terrific sounding mixin the headphones, and pre p a re tore c o rd the dub onto a free tra ck .

Th a t ’s a lot to do all at once. Th et ra ck sheet needs to commu n i c a t ec l e a rly ex a c t ly where all the tra ck s

MediaThe track sheet—and all studio documents, for that matter—works best when it is recorded by

hand, in pencil. It is tempting, in this age of slick computer graphics, to transcribe your track sheetsinto some sort of computer generated format. After a session, you kindly take the track sheets intothe office and type them into the computer. Nifty. Cut and paste some graphics, select a cool font,and the print outs will look slick.

Please, don’t do it! The track sheet is a living document. At any point in the project, from the basicssession to the mastering session, the track sheet should welcome creative and free thinking. If themusic suggests you should erase the cello and track a triangle, then do it. If you’ve gone to the trou-ble to type all the tracks into the computer, you’ll hesitate an extra bit. Replacing the cello with a tri-angle means that tonight, after the session, you’ll have to type in the change and print out a new one.

That’s a chore. And it just isn’t necessary. Moreover, if it diminishes, in any way whatsoever,the creative energy of the project, then it is a mistake. The manual track sheet system is the pre-ferred approach.

In addition, a good track sheet has little scribbles and notes that, though meaningful to the engi-neer, may not seem important to the assistant transferring it into the computer. In computerizing it,some of that information is inevitably lost. Stick with hand written track sheets.

Some people, though talked out of using a computer for keeping track sheets, make a worse mis-take: they use ink. Ink doesn’t erase. Tape does. Use pencil. We record on tape or hard disk beca u s eit’s easy to erase and record new ideas. Erasing and re-recording is an everyday part of modern mul-t i t rack music production. The track sheet should follow. Consider it law: track sheets (and all studio doc-uments) should be done in pencil. Pens and laserjets are too permanent. They are strictly forbidden.

...and even if you don’t have to erase

them to make room on your disk, you’d

darned well better have a good way to

know which are the keepers, and fast.

Excerpted from the Febru a ry edition of RECORDING magazine.©2001 Music Maker Publications, Inc. Reprinted with permission.



of audio are , w h i ch tra cks to use,w h i ch not to use, what can bee ra s e d , and so on.

It is wise to allocate tracks as con-sistently as possible across a projectso that, for example, the snare isalways on track three and the leadvocal on track 24.Allocate the morevariable musical elements to othertracks. Not every song has piano.Some use clavinet, some just useguitar, etc.

Good habits laying out the tra cksheet consistently from song to songreduce the effort associated witha dvancing to the next song for then ext ove rd u b . With maybe drums,b a s s , and rhythm guitar alre a dy setup and sounding balanced for contro lroom and headphone monitoring, yo ucan tweak the didgereedoo and tro m-bone as re q u i red for this particularsong and get on with the ove rd u b .

And all of this goes double fo rthose nifty modern hard diskre c o rd e rs that let you save gazil-lions of takes per final tra ck! Somem a nu fa c t u re rs , l i ke Roland, p rov i d eyou with a certain number of alter-nate takes per tra ck; others , l i keA k a i , offer a genera l i zed pool ofava i l able edits and alternate tra ck s .Either way, yo u ’ re now talkingabout dozens—or h u n d re d s !— o ft a ke s , and even if you don’t have to

e rase them to make room on yo u rd i s k , yo u ’d darned well better havea good way to know which are theke e p e rs , and fa s t .

Push the decision make rs tod e c i d e . If you are the producer or ifi t ’s your mu s i c , step up to the plate.But even if you are just acting in an

Clear notes like TBE commu n i c a t eex a c t ly which tra cks can be nu ked ifn e c e s s a ry to accommodate add i t i o n-al ove rd u b s . The session losesmomentum if you have to pause theove rdub session and look for anava i l able tra ck .“ U m m , it says here ,

‘ t a m b o, t a ke 3.’ I think we ’ re goingwith take 2. Hang on a minute whilethe producer and I listen to all fivetambourine parts and figure outw h i ch one we can era s e . Oh! Yo u ’ resounding gre a t . L ove the energy inthat last take . G ive us five or tenm i nutes and we’ll do another.”

Good habits laying out the track sheet

consistently from song to song reduce

the effort associated with advancing to

the next song for the next overdub.

engineering capacity, help the ses-sion by coaxing these sorts of com-mitments out of the key playe rs .Hedging your cre a t ive bets by arch iv-ing countless mediocre takes willn e e d l e s s ly increase the studio time(a budget bre a ker) at the ve ry least.

Wo rs e , and more like ly, it will ro bthe project of its cre a t ive and perfo r-mance edge. Safe albums don’t usu-a l ly sell.

N ext month we discuss the rest ofthe studio documents: take sheets,setup sheets, and recall sheets. G o o dstudio documents are a session toolyou can have without parting withtoo mu ch money. S u re it would bem o re fun to buy another micro p h o n eor compre s s o r, but it’s worth thee f fort to develop and use these docu-ments thoughtfully.

Alex Case doesn’t stop at studio doc-u m e n t s .You should see his grocery list.Send questions and suggestions toc a s e @ re c o r d i n g m a g . c o m

Home Made DocumentsIf you’ve got a printer and some graphics skills, by all means go ahead and create your own studio documents. You’ll need

a track sheet—maybe both an eight-track and a twenty-four-track version, maybe others. You’ll also need a take sheet suit-able for any type of session, live to two or multitrack. You’ll want a setup sheet that outlines the basic studio setup for anysession you might encounter. You’ll create recall sheets for every piece of you own so you can document their use in any appli-cation. And—my favorite—design a professional looking invoice. That’s the itemized list of expenses for a session (studiotime, engineering time, tape costs, shipping, special gear rental, meals, limos, etc.). The cool part of the invoice it the part atthe bottom that says, “Please pay $X,XXX dollars. Make checks payable to the order of, “Me, the engineer.”

In creating these documents, you’ll naturally want to give them a professional look that supports your image and re p-utation. You’ll give them a consistent, integrated look, using the same font for example. Go to town. But here are someother things to consider:

SpaceLeave adequate space where re q u i red for the document to work. A large logo might look cool now, but it won’t leave

room for the engineer to write all the information legibly. On a track sheet, the space for the tracks should be as larg eas conveniently fits. I’ve seen track sheets where the number in each track space was so large I couldn’t write the word s ,“Lead Vocal.” When blank, the track sheet should be mostly open space for notes. When used, the document should wel-come, even inspire, clear note taking. In addition to the tracks themselves, I leave room to write particularly import a n titems like the name of the artist and the song titles in larger print.

For eight-track projects, I use a separate track sheet that leaves even more room for all kinds of notes (see FigureTwo). This month’s article discussed the wealth of information that must be re c o rded on the track sheet. As the attachedE i g h t - Track Sheet shows, even more is documented when possible. Scribble cues off to one side of the page during thecourse of the first playback of the song so that you can instantly find verse four when asked. Note when the harm o n yvocal is singing (e.g. choruses only). Note the times of problematic spots to go back and check. I think I heard a flubbednote on the Acoustic Guitar at about 6:43 and a fret buzz around 7:31.

PaperThese documents get pretty rough treatment. You’ll use them at the basics session, at every overdub session, and final-

ly during the mix sessions. They’ll get written on, erased, written on again, and erased, and written on again.... It’sinevitable that they’ll be used as coasters, scratch paper, and note paper. They’ll document audio tracks, phone messagesand food orders. If your track sheet does all these things, it’s a session asset. And it will better survive all this abuse ifyou print it on to heavy paper. Even card stock isn’t a bad idea.


Studio Documentat i o n ,Studio Documentat i o n ,BY ALEX CASE

Pa rt 2—The TAKE SHEET, SETUP SHEET, AND RECALL SHEET

difficult to distinguish ve rse one from ve rse two and takeone from take two. For a smooth session it is positive lyvital to keep a thorough and accurate take sheet, s t a r t i n gwith the song title and start time.

B eyond this basic bookke e p i n g , the take sheet serves ava l u able production function. During the course of a ses-

sion we use the take sheet to keep tra ckof which songs have been re c o rded andw h i ch songs have not. It identifies thosesongs that are re c o rded and those thata re n ’t .

C o n t i nuing last month’s session withthe band Scribe, we see the four songst h ey are wo rking on are scribbled dow nat the bottom of the take sheet (seeFi g u re 1). Ty p i c a l ly this is done on theb a ck of the tra ck sheet; we show it herefor illustra t ive purp o s e s . As the bandcompletes a take that eve ryone like s , t h eengineer ch e cks that tune off. For 12-song pro j e c t s , this sort of thing is ve ryu s e f u l .

The take sheet does more than just listthe tunes tra cked and their start times.We note the take nu m b e r, w h i ch as we ’ l lsee below is important information fo r

monitoring the health and pro d u c t ivity of a session. Wealso note the approximate end time for each take (ro u n d-ed off to the nearest 5-second increment) and calculatethe length of the song. Wa t ch and compare these nu m-b e rs to tra ck how the session is brew i n g .

Th e re is also a ‘Notes’ column on the take sheet.N a t u ra l ly, h e re you make notes of critical observa t i o n so f f e red by the producer or the band members . I t ’s impor-tant to keep tra ck of comments such as the pro d u c e rl i kes the bridge, the drummer loves the solo, the singerhates ve rse 3, e t c .

hough we are engineers and not banke rs , we ’ ve got alot of documentation to complete. Studio paperwo rk

essentials include the tra ck sheet, the take sheet, t h esetup sheet, and recall sheets. We talked at length ab o u tthe importance and application of tra ck sheets lastmonth; we now pick up where we left off and continu eour discussion with the take sheet.

L i ke the tra ck sheet, the take sheets e rves a misleadingly stra i g h t fo r wa rdfunction: it lists the takes re c o rd e d . O n egoal of this month’s N&B is to reve a lsome of the hidden benefits that comef rom giving the take sheet a little ex t rac a re and attention.

Fi g u re 1 shows the take sheet we useat my studio, Fe r m a t a . On the top wefind ve ry nearly the same info r m a t i o nthat capped the tra ck sheet. The pro j e c tis identified by artist name (written inl a rge print), the pro d u c e r, the engineer,the assistant, and the date the pro j e c tc o m m e n c e d . The reason for all thisi n formation is self-ev i d e n t . The heart ofthe take sheet is what comes nex t .

OdometerThe principal role of the take sheet is to identify the

p recise location of each and eve ry take of each and eve rysong re c o rd e d . I t ’s risky to re ly on memory. I t ’s foolish tore ly on the assistant engineer’s memory. I t ’s flat out dan-g e rous to re ly on the drummer. And ab ove all we must tryto avoid torturing the client with “ Wa i t , let me find it.H e re it is. N o, wa i t . Is that Ta ke two? Not sure . Hold on.”

I t ’s even wo rse on those sessions where the vocal does-n ’t get re c o rded until some future ove rdub session.Tonight you might just be looking for that killer take fo rthe rhythm section only. Without a vocal tra ck it will be

T

Instead of

panicking and

w i l d ly turning

k n o bs, yo u

can instead

ca l m ly tra c e

problems to

their source—

if you took

good notes.


Figure 1


Figure 2


In add i t i o n , eve ry take gets one oft h ree codes:

C: Complete —this is a completet a ke , top to bottom. It does notreflect an opinion whether or notthis is the pre f e r re d , s e l e c t e d , s u reto win a Gra m my take , just that itis complete.

I: Incomplete—the band aborts thet a ke somew h e re along the way.Th e re ’s enough good stuff in it,t h o u g h , that rather than erase it andlose it fo rever you save it.

FS: False Start—the tune didn’tstart cleanly, m aybe someone misseda cue. The band stops and immedi-a t e ly counts it off, l a u n ching right

b a ck into the tune again. We consid-er all of this part of the same takeand just note the start time of then ext down beat without stoppingtape and interrupting the gro ove .

When the band lays down a takethat eve ryone knows is the one, c i rc l ethe take number to designate it theselected take . At the end of the basics

s e s s i o n ,t h e re should be only one cir-cled take for each song title.Ti t l e swithout circled takes are n ’t done ye t .

BarometerA good engineer and producer will

wa t ch the take sheet for clues ab o u th ow the band is feeling. You mighte a s i ly need thre e , fo u r, or more take sof the first tune as the band wa r m su p, the engineer gets the soundsunder contro l , and eve ryone getsused to the studio and each other.

When the band starts gettingthings in one or two take s , t h ey arein ‘the zo n e .’ At this point theyshould not be interrupted for any-

thing short of a pending nu c l e a rdisaster or a re a l ly good episode ofThe Simpsons.

If the session persists with mu l t i-ple unsatisfa c t o ry takes (e.g. ‘ Th a ti s n ’t happening. L e t ’s move on to then ext song’) song after song, yo u rt a ke sheet is trying to tell you there ’sa pro b l e m . C h e ck to make sure theheadphone mix sounds gre a t , rev i ewthe studio setup and make sure thep l aye rs can see each other, and mosti m p o r t a n t ly manage the sessionmood to help people re l a x , fo rget thes t u d i o, and just play the mu s i c .

Looking at Fi g u re 1, you’ll see wewe re having trouble with the tune‘ N o t o r i e t y.’ Ta ke 1 was just okay; noone re a l ly liked it. Ta kes 2 and 3we re incomplete, with a false start inb e t we e n . The band keeps hitting asnag and aborting the take .“ N op ro b l e m ,” the producer said, “ t h a ti n t e rlude is a tricky section. L e t ’scome back to it later.” So the bandp roceeds to one-take the next twos o n g s . A good sign.

Back to ‘Notoriety’ Take 4, andproblems resume.A careful look atthe timing of the aborted takesreveals that the band keeps stoppingat the same point, the interludeabout 90 seconds into the tune. Theproducer and band have a musicalroad block to solve; time torehearse, rewrite, or remove thetrouble spot.

SpeedometerBy song number 14, ‘Notoriety’

Take 7, the band has progressedbeyond the musical train wrecksthat caused the whole take to stop.But the finished takes are gettingFigure 3

A good engineer and producer will

w atch the take sheet for clu e s

about how the band is feeling.


longer and longer. The tune is drag-ging. No one seems thrilled with thefeel of the take.

M aybe the song is too hard; may b ethe band is too tire d . The take sheetpoints out the pro b l e m . The pro d u c e rmust decode this and direct the ses-sion accord i n g ly. Sometimes the takesheet simply hints that the bandneeds a bre a k , and they should haveo n e — p re f e rab ly befo re they noticet h ey need it. The easy way to take a

b reak without undermining theb a n d ’s confidence is to announce“ P i z z a ’s here ” or “ H ey, we ’ ve got af resh pot of coffee.”

D o n ’t fail to notice the oppositet rend: speeding up. No matter howtalented they are , bands are pro n eto rushing the tempo as they fighttheir way through a complicateda r rangement while the studio clockt i cks and the adrenaline flow s . G o o dp ro d u c e rs alre a dy have a targ e tbeats per minute goal for each tune.Let the take sheet help you measurethe tempo of the tune so you knowwhen the band is sprinting insteadof gro ov i n g .

Setup sheetsFor any session other than a single

ove rd u b , it makes sense to documentthe general layout of the studio—theequipment used, the location of thep l aye rs , the placement of the micro-p h o n e s , e t c .

The heart of the setup sheet is sim-p ly a list of what microphone andsignal processing was used in there c o rding of each and eve ry tra ck .This sheet acts as an equipment ro a dmap both during and after the ses-s i o n . Fo l l ow along on Fi g u re 2.

During the session, when the stu-dio is crowded with micro p h o n e s ,buzzing with mu s i c i a n s , and tangledwith countless mic cables snakingtheir way around the studio, it canbe difficult indeed to find and fixp ro b l e m s . I f, for ex a m p l e , you hearthe dreaded cra ckle and crunch of afailing microphone cable on the floort o m , good luck replacing it.

U n l e s s , of cours e , yo u ’ ve got ana c c u rate and current setup sheet

that tells you that the one pluggedinto microphone receptacle inputnumber eight is the floor tom mic.Unplug it. L e ave it in the ra t ’s nestof cables and just add another. Th ep roblem is solved and little timewas wa s t e d . ( I d e a l ly you’ll alsom a rk the guilty cable with a pieceof tape so you know which one getsre p a i red later. )

If you are lucky enough to ow nfour identical compre s s o rs — s a m em a ke and model number—it can beh a rd to remember which Squish-o-matic Tormentor Mark IV was onthe snare . Setup sheet to the re s c u ea g a i n .

When you are sitting at a consolefull of twitching meters spitting outthe sound of the band re h e a rs i n gtheir first nu m b e r, it can get confus-ing and more than a little intimidat-i n g . You might find yo u rself unab l eto locate the fader that controls theelectric guitar signal. Yo u ’ re try i n gto send it to tra ck ten, but the meteron the mu l t i t ra ck doesn’t budge.

Instead of panicking and madlypushing fa d e rs , t h rowing sw i t ch e s ,and cranking microphone pre - a m p sup to their maximum gain settingshoping to hear some guitar, you caninstead calmly trace the pro b l e mf rom its sourc e . Did someone fo rg e tto plug in the EGT mike? It’s a sim-p l e , common mistake . The setupsheet makes it clear: the electric gui-tar microphone should be pluggedinto microphone input number 22.

After the session the setup sheetguides you through the many thingsyou accomplished. I n ev i t ab ly the

Figure 4

I n e v i ta b ly the producer says some-

thing like “Didn’t you use the mic

t h at looks like a giant Tylenol?”


p roducer says something like “ G iveus that killer guitar sound you hadduring tra ck i n g . D i d n ’t you use thebig microphone that looks like agiant Ty l e n o l ? ” And this will be agood ten days after the basics ses-s i o n . To have a fighting chance of sat-isfying this re q u e s t , you’ll need asetup sheet that arch ives the basicelements of the signal you put ontape: mic(s) and any compre s s i o n ,equalization or other effects.

The other side of the setup sheethas a floor plan of your studio fo ryou to make notes on where you set

the various playe rs up in the ro o m .H e re you note the approximate loca-tion of the drum kit in the big space,the location of gobos around the gui-tar amp, that the singer was in thebooth but his amp was out in theh a l lway, and so on.

For super twe a ky sessions yo umight even make measurements ofthe locations of key micro p h o n e s .Photos—made especially easy in thisage of digital photogra p hy — h e l p,but a ske t ch of the session layout onthe back of the setup sheet is a use-ful way to document what happened.

RecallsIf you’ve had the pleasure of work-

ing in a full-rate studio that assignsan assistant engineer to your pro-ject, you know the benefits thisbrings. There’s a team of runnersand assistants getting food, etc. Theassistant also takes care of theessential thorough note taking thatgoes on during your session.

P rofessional facilities will docu-ment eve ry setting of eve ry single

piece of gear you use. This documen-tation makes it possible—at leastt h e o re t i c a l ly—to recall at a laterdate any sound you re c o rd at anytime throughout the pro j e c t . If yo uselect a microphone and re c o rd at ra ck straight to tape withoute f f e c t s , that is noted. If you addsome compre s s i o n , e q u a l i z a t i o n , d e -

e s s i n g , phasing and/or any othere f f e c t s , this is also noted. The docu-mentation must also capture theexact settings of each piece of gear.

Two things help: recall sheets andre c a l l able consoles. For each piece ofgear you ow n , it is wise to create arecall sheet that lays out eve ry knob,sw i t ch , and editable window in thed ev i c e . Fi g u res 3 and 4 show a com-mon pair of recall sheets I use.

Fi g u re 3 is the recall sheet for amic pre/eq—the Geoffrey Dakingmodel re f e r red to in Fi g u re 2. Th epage visually shows the knobs. H e re

we added sparkle to the lead vo c a lby pushing up the super high fre-q u e n cy range at 15 kHz. P re s e n c ewas helped by a little boost at 3 kHz.The other bands of equalization andthe low pass and high pass filterswe re not used (noted ‘ o u t ’ ) .

Fi g u re 4 shows the philosophicalo p p o s i t e , a digital mu l t i e f f e c t s

d ev i c e , the ve n e rable Ya m a h aS P X 9 0 . The recall sheet for thisd evice changes form to accommo-date the lack of knobs but widerange of editable window s . Th i smu ch detail is only necessary in situ-ations where your patches can getoverwritten by someone else—andyo u ’ re not running some sort ofSysEx librarian on the studio com-puter to save your own patches awayf rom the effects unit.

During a basics or live-to-twosession, there will be a fair amountto document; the typical overdub is

generally much simpler. But duringmidown you may have to documentsettings on every piece of gear youown.

B eyond recall sheets, we have theability on many consoles and pre t t ymu ch all digital audio wo rk s t a t i o n sto store the many settings andeffects on the mixer. A vocal ove rd u bmight get some equalization fro myour DAW. Documenting those set-tings is as easy as a “ S ave A s . . .” c o m-m a n d . At the end of a project a sin-gle song might have well more than ad o zen saved ve rs i o n s .

Taking noteP rojects end not only with a stack

of master tapes, but also with at h i ck file full of documents. Th et ra ck sheet (discussed last month),the take sheet, the setup sheet andall those recall sheets are an essen-tial part of the re c o rding cra f t .These documents help you get moreout of your equipment and commu-nicate a higher level of pro f e s s i o n a l-ism to clients.

Alex Case reminds you that wh i l eyou can tra ck a take sheet, and youcan take a tra ck sheet, you can’t sheeta take tra ck .Send questions and sug-gestions for Nuts & Bolts toc a s e @ re c o r d i n g m a g . c o m .

Don’t temptfate

Notice that these studio documents always con-tain the same key information: artist, producer,engineer, etc. These documents should live with themultitrack tapes at all times. But if ever they getseparated, this information will help you re-attachthem to the correct project.

A multitrack tape without a track sheet isn’t easyto use. You’ll need a couple hours just to figure outwhat the heck is on tape, track by track. Ditto for amissing take sheet. I’ve attended sessions wherethey mixed the wrong take of a song. A solid tenhours of studio time wasted.

A d d i t i o n a l l y, these documents should always,always include your studio contact information. Yo uwant to get the call from the re c o rd company wheni t ’s time to do some more mixing.

(The phone and address are omitted from thef i g u res in this article only because this magazine isread by billions and billions of enthusiastic and tal-ented re c o rding musicians. If even 1% of you decid-ed to call, I’d never get anything done. Rest assure dthat the actual documents do in fact include thisi n f o rm a t i o n ! )

It’s risky to rely on memory. It’s foolish to rely

on the ass i s tant engineer’s memory. It’s flat out

dangerous to rely on the drummer...

...and above all we must try to avoid torturing

the client with “Wait, let me find it. Here it is.

No, wait. Is that Take two? Not sure. Hold on.”

Excerpted from the March edition of RECORDING magazine. ©2001 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel:(303) 516-9118 Fax: (303) 516-9119 For Subscription Information, call: 1-800-582-8326


M i s - a p p lying EffectsM i s - a p p lying EffectsBY ALEX CASE

CompressualizationWhen is a compressor not a compressor? When it’s an

e q u a l i ze r, of cours e . A de-esser to be ex a c t . D e - e s s e rsa t t e nuate the ‘ess’ sounds in a vocal tra ck made by theletter S. A loud, s t rong ess of a vocal can zap you withan ear ringing, pain inflicting burst of high fre q u e n cye n e rg y.

Using eq to attenuate the problematic high fre q u e n-cies associated with the esses will also rob the vocal ofits airy, s h i m m e ry, voice of the pop music gods qualitythat yo u ’ ve gone to so mu ch trouble to cre a t e . The fact is,the vocal pro b ab ly sounds gre a t , if not perfect, w h e n eve rthe singer isn’t singing wo rds with the dreaded letter S.To get an edgy, emotion filled vocal that cuts through amix crowded with fuzzy guitars , hissing cy m b a l s , a n dshimmering strings, yo u ’ ve got to go for a bright vo c a lsound from the start—it influences mic selection, m i cp l a c e m e n t , and of course the effects you add . These ove r-ly bright esses are an almost unavo i d able side effect ofotherwise good re c o rding pra c t i c e .

The solution is to use a compressor instead of ane q u a l i ze r. The goal is to run the vocal through a compre s-sor that attenuates the vocal only on the problematic esssounds; the rest of the time, the compressor should notchange the magic vocal one iota.

Trouble is, no amount of fiddling with the thre s h o l d ,a t t a ck , re l e a s e , and ratio controls will accomplish this.These ess sounds happen so quick ly that only anex t re m e ly fast compressor attack time could grab them.

M o re ove r, even though they are perc e p t u a l ly ve ry loudand annoy i n g , the typical changes in loudness that occurf rom ve rse to ch o r u s , line to line, and even wo rd to wo rda re mu ch greater swings in amplitude than a little ol’ let-ter S. As a re s u l t , the compressor reacts to the louderparts of the singing, not the indiv i d u a l ly, p e rc e p t u a l lylouder sizzling sounds of the esses. We need a way towarn the compressor that an S is happening, despite theex p re s s ive changes in dynamics of the vocal tra ck .

We accomplish this through clever use of the compre s-s o r ’s side-ch a i n . The side-chain offers an alternative inputinto the compressor—an input that wo n ’t have a corre-sponding output into the mix. This other input is just usedto tell the compressor when and when not to compre s s .

To get rid of the esses in the vo c a l , we route a copy ofthe vocal signal with the esses e m p h a s i z e dinto the side-chain input of the compre s s o r. As shown in Fi g u re 1, wesplit the Lead Vo c a l , send it to a parametric eq, and thento the compressor side-chain input. Set the eq to a nar-row (high Q) but large boost (+12 dB or maybe more) atthe problematic fre q u e n cy ra n g e .To find the exact fre-q u e n cy ra n g e , you can hunt around from about 2 kHz–8kHz until the compressor starts to react to the esses.

You’ll find you can ze ro in on other sibilant pro b l e m sthat might arise—it’s not just for esses. You can de-F, d e -X , d e - T, d e - C h , de-Sh...this basic signal flow structure ise f f e c t ive at re m oving many related pro b l e m s . A sharpboost enables the compressor to duck the signal in re a c-tion to a single spectral spot.

This month we apply some of our studio tools in ways that might seem

like some sort of trick. We review some unlikely, unbelievable, or at

least counterintuitive approaches to using effects. The beauty of

this April Nuts & Bolts column is, it ain’t no joke.


You can broaden the bandwidth of the side-chain para-metric eq to catch a range of sibilant sounds. This can bepushed to other applications: de-squeak an acoustic gui-t a r, de-thump a piano. . . If you find the de-Suck setting,email me.

Some way cool compre s s o rs have a sw i t ch that lets yo umonitor the side-ch a i n . This enables you to re a l ly finetune the triggering fre q u e n cy that gets the hy p e r- b o o s t .

Once you get the compressor to react to the esses, yo umust then use your good judgement to set the compre s-sion ratio just right. Too high, and the compressor ove r re-acts to each S, l i t e ra l ly giving the lead singer a lisp. To ol ow, and the esses continue to annoy. L i ke many mix

m ove s , it is sometimes useful to tweak it too far (wherethe de-esser is audible and unnatural) and then back offuntil you imagine that you can’t quite hear it wo rk i n g .

In the end you should be able to push the eq on theactual lead vocal hard , without fear of sibilant destruc-t i o n . Then your lead vocal can have all the gr i t , gasp andguts that pop music demands. Betht of luck .

Eqwahlization“ Wa h - wa h .” What an effect. Used tastefully, it can give

a tune that perfect ex t ra push towa rd , we l l , w h a t eve ryo u ’ re aiming fo r. H ow ’s it done? With a Cry Baby effectspedal (or one of its siblings), n a t u ra l ly. But what if yo ud o n ’t have one? What if you d o h ave one but the last ninevolt battery in the Tri-State area just pooped out?

This effect is nothing more than va r i able eq. If yo u ’ vea parametric equalizer handy, p a t ch the electric guitar orkey b o a rd tra ck through it. Dial in a pretty sharpm i d range boost (high-Q, 1 kHz, + 1 2 d B ) . As the tra ckp l ay s , m a nu a l ly sweep the fre q u e n cy knob with one handand salute Jimi and Stevie with the other.

Hip DAWs with automated equalize rs make it easy top rogram this sort of eq cra z i n e s s . Without automation,you just print your wa h - wah performance to a spare tra ck .

I t ’s also worth exploring other fre q u e n cy ra n g e s . Trycuts as well as boosts; use narrow and broad bandw i d t h s ;t ry sweeping a highpass or a lowpass filter or shelving eq.

And perhaps most importantly, a p p ly it to any tra ck .Piano offers a welcome wa h - wah opportunity. It seemsp e r f e c t ly appropriate to wa h - wah a cello or a snared r u m — ab s o l u t e ly any t h i n g .

StimulatorAmp simu l a t o rs have been a boon to the home re c o rd i s t .

Some (most, a c t u a l ly) guitar amps only sound good whent h ey are cra n ked up to ear splitting leve l s .S o m e t h i n gmusical happens as the amp re a ches its limits—electro n i-c a l ly, m e ch a n i c a l ly, p hy s i c a l ly, and metaphy s i c a l ly.

But what is an up-all-night home studio to do? Recordd i rect and ach i eve that guitar amp near death ex p e r i e n c ecourtesy of amp simulation hardwa re / s o f t wa re .N e a t o.

Perhaps you use DI boxes when re c o rding bass. That isp retty common practice these day s . G reat sounding bassamps re q u i re money, c a re , s t re n g t h , s p a c e , a good bassg u i t a r, an excellent bass playe r, and massive amounts ofacoustic isolation when tra cking (there ’s that “the amp’stoo loud’ problem again). The direct inject device make s

Figure 1: De-Essing a Lead Vocal relies on a side-chain input with Boosted Esses. Patch this up. Add Delays, Reverb, and Sundry. Win Grammy’s.

The fact is, the vocal probably sounds great, if not perfect,

whenever the singer isn’t singing words with the dreaded letter S.

You can de-F, de-X, de-T, de-Ch, de-Sh... this basic signal flow

structure is effective at removing many related problems.

it possible to use the signal comingout of the bass guitar itself fo rre c o rding onto tape or disk.

Think about it. We stick micro-phones in front of instruments to con-vert the noise they make in the airinto an electrical signal on a wire .Once the music is “ i n ” that mic cab l e ,we can run it through our ra cks ofaudio equipment. S u ch an appro a chm a kes a lot of sense for voice and

p i a n o. But electric basses are , u m ,e l e c t r i c . Why not skip that whole ‘ e l e c-trical signal to amp to acoustic noiseto microphone to electricity’ thang?This simple view motivates the DI.

A DI has to take care of some elec-tricity book-keeping: it lowe rs thevo l t a g e , l owe rs the impedance, a n dbalances the signal so that whatcomes out of the DI behaves ve rymu ch like the signal that comes out ofmost micro p h o n e s . Off it goes into the

rest of the re c o rding ch a i n — e q u a l i zei t ,c o m p ress it, and print it to tape.

The DI is quite effective on bass.The sound can be tight, c r i s p, a n dr i ch with low end wa r m t h . In fa c t ,even when you have the luxury ofre c o rding the bass through an amp,it is common practice to simu l t a n e-o u s ly re c o rd the bass with a DI ontoa separate tra ck . With both an ampsound and a direct sound on tape,

you have more options for creating ap owerful bass sound at mixdow n .

R e c o rd electric guitar through ad i rect box and—blip, b o i n k , f l i rp —o u ch . Sounds thin, p e rky, s i l ly, [o t h e rcolorful descriptions thoughtfully delet-e d— E d . ] . It wo n ’t wa ke the neigh-b o rs , but it wo n ’t sell any re c o rd seither; the guitar amp is too mu ch apart of the tone equation.

Perhaps yo u ’ ve tried to play a CDt h rough a guitar amp. Notice how bad

it always sounded? It’s honky, with noh i g h s , mu ddy low s , and ze ro dy n a m i cra n g e . That is what the amp does tothe signal coming from the electricg u i t a r. And the electric guitar justi s n ’t an electric guitar without it; Id o n ’t think Leo Fender ever wa n t e dus to hear the sound coming out ofthe guitar itself. So whenever a ses-sion fo rces us resort to re c o rding elec-tric guitar direct to let the neighborss l e e p, it is essential that we grab theamp simu l a t o r.

The amp simulator offers us a sin-gle stomp box , ra ck space or pull-d own menu that throws in a ton ofd i s t o r t i o n , c o m p re s s i o n , e q u a l i z a t i o n ,and god only knows what else. Th i seffect begs for ex p e r i m e n t a t i o n !

D o n ’t let anyone pull a fast one onyo u .You can use compression toe q u a l i ze , equalization to wa h - wa h -i ze , and amp simulation toi m p rov i ze . H a p py signal pro c e s s i n g .

Alex Case makes the “wah-wah” facewhenever he uses the electric pencils h a r p e n e r. Offer therapy viac a s e @ re c o r d i n g m a g . c o m .

Why not skip the whole ‘electric signal to amp to

acoustic noise to mic to electric signal’ thang?

Excerpted from the April edition of RECORDING magazine. ©2001 Music Maker Publications, Inc. Reprinted with permission. 5412 Idylwild Trail, Suite 100, Boulder, CO80301 Tel: (303) 516-9118 Fax: (303) 516-9119 For Subscription Information, call: 1-800-582-8326

RECORDING MAY 2001

The Channel StripPurchase priority By Alex Case

I t ’s time to re-ask that favorite re c o rd i s t ’s question:What do I buy nex t ?

L e t ’s org a n i ze the answer by looking closely at thewo rk h o rse of the studio: the channel strip. Almost eve rysession—be it live to two, b a s i c s , ove rd u b , or mixdow n —re q u i res that chain of signal processing gear we use mosto f t e n . It consists of a microphone pre a m p l i f i e r, e q u a l i ze r,and compre s s o r / l i m i t e r.

PreambleThe vast majority of music we hear on the radio or on

recordings enters the recording signal chain through amic preamp. So it’s important to make sure we have agood one.

Furthermore, mic signals are different from mostaudio signals in the studio. Compared to the signal thatcomes out of a compressor, an equalizer, a compact discplayer, or your console—to name just a few—the signalcoming out of a microphone is much, much lower in volt-age.A microphone generally outputs about 1/1000 thevoltage of most of our other studio gear, so mic signalsmust be pampered and protected; the mic pre is noplace to cut corners.

O bv i o u s ly, the number of mic preamps you have deter-mines the number of different microphones you canre c o rd at once. If you build up your mu l t i t ra ck pro d u c-tions one tra ck at a time, one mic at a time, you can cer-t a i n ly make do with one microphone pre a m p.

But if you have the sort of facility capable of re c o rd-ing drums, you might need to have the fo l l owing micsup and re c o rding all at once: hi-hat, k i ck drum, s n a red r u m , t wo or three tom toms, a floor tom, t wo ove r h e a dm i c s , and two ambient mics. That adds up to at least tenm i c s . And it can easily swell to many more if the drumkit gets bigger, or if you like to experiment with mu l t i-ple mics on the same drum; placing two or three micson the snare drum alone isn’t unusual in pop mu s i cre c o rd i n g . Your need for mic preamps grows right alongwith the mic count.

If you want to be able to handle the sort of music inw h i ch multiple playe rs are re c o rded simu l t a n e o u s ly,you’ll need ex t ra mic pre s . A classic example is thep ower trio, m aybe blues: drums, b a s s , and electric gui-t a r — t h ey all want to jam tog e t h e r. You’ll need enoughmic preamps to get them all to tape or disk simu l t a n e-o u s ly. Want to re c o rd a big band?

But another reason to go out and buy another mic pre-amp is just for variety. As is true for mics, loudspeakers,and compressors, no two mic preamps sound exactlyalike. They have their own signature or flavor that cansound exactly right—or exactly wrong—when pairedwith a certain mic on a certain instrument for a certainkind of tune.

One final factor pre s s u res us to acquire additional micp reamps: session flow. Keeping the session moving effi-c i e n t ly saves the band money and makes the studio am o re cre a t ive place to wo rk . A common appro a ch — eve nwhen the session is a string of single-mic ove rdubs—is tol e ave each signal chain up and unchanged as you moveon to the next ove rd u b .

When the piano ove rdub is complete and yo u ’ re mov-ing on to a few cymbal swe l l s , the mics and mic pre set-tings on the piano stay where they are . You use differe n tones on the cymbal ove rd u b . Not only does that avo i dstopping the session to set up for the cy m b a l , it meansyo u ’ re re a dy to go if someone wants to change the pianopart in the bridge.

Created equalEqualization is a fundamental part of the music re c o rd-

ing cra f t , so it too is a part of the channel strip. B u tplease don’t re a ch for the knobs of the equalizer toos o o n . Th e re is definitely no substitute for good mic selec-tion and placement. If yo u ’ re lucky, p a t i e n t , and smartenough to have a beautiful sounding instrument that yo ucan place cleve rly in a great sounding, we l l - c o n t ro l l e dre c o rding ro o m , using excellent mics placed in that eve r-e l u s ive ‘ sweet spot,’ you may never use eq.

RECORDING MAY 2001

The rest of the time, we get by witha little help from our equalize r. A ddp u n ch , re m ove shrillness, a dd spark l e ,re m ove mu dd i n e s s .E q u a l i ze rs are anessential part of getting our pro j e c t sre a dy for prime time.

But even more than mic pre a m p l i-f i e rs , e q u a l i ze rs of different typesf rom different manu fa c t u re rs cansound quite different from one anoth-e r. Using the same equalizer on mosteve ry ove rdub you do starts to giveeve rything the same sonic aftertaste.

We need a better strategy thanjust ra n d o m ly buying a few differe n te q u a l i ze rs . And snapping up the lat-est eq du jour wo n ’t guarantee we ’ l lend up with a coordinated set. I sug-gest dive rsifying your equalizer col-lection based on the tech n o l og ye m p l oyed and the functional type ofequalization: softwa re ve rsus hard-wa re , solid-state ve rsus tube, i n t e-grated circuit ve rsus all discre e t , d i g-ital ve rsus analog , among others .

O ver time you’ll learn to hear thesubtle sonic differences betwe e nt h e m . A session starts. You hear thes i n g e r ’s tone. And a bell goes off.I n s t a n t ly you intuit the right ch o i c eof equalizer for this ove rd u b .

B eyond tech n o l og y, it makes senseto enrich your equalizer collectionbased on functional capab i l i t i e s :p a ra m e t r i c , s e m i - p a ra m e t r i c , gra p h-i c , or plain old program eq.

The parametric equalizer offersthe most precise control for spectra lm a n i p u l a t i o n , with three differe n tp a ra m e t e rs (hence the name) fo ryour knob tweaking pleasure . All theother types of equalize rs (semi-para-m e t r i c , graphic and program equaliz-e rs) have some subset of these thre ep a ra m e t e rs ava i l able for adjustingon the front of the box or in the pull-d own menu; the missing para m e t e rsa re fixed by the manu fa c t u re r. Wh e nyou learn how to use a para m e t r i ce q u a l i ze r, you are learning how touse all types of equalize rs .

P ro b ab ly the most obvious para m e-ter needed on an equalizer is theone that selects the center fre q u e n cyyou wish to attack . In search of shim-m e r, we might dial up an eq shapefocused on 10 k H z . We ’ ve got to lis-ten care f u l ly, t h o u g h , because theshimmeriness may be better at1 2 kHz for today ’s particular tra ck .

In para l l e l , we must determineh ow mu ch to alter the fre q u e n cy wea re selecting. The addition(or sub-t raction of frequencies happens viaadjustment of a separate para m e t e r :c u t / b o o s t . It indicates the amount ofd e c rease or increase in amplitude atthe center fre q u e n cy you dialed inon the first parameter just discussedab ove . To take the mu ddiness out ofa piano sound, select a low-ish fre-q u e n cy (around 250 Hz maybe) andcut a small amount—maybe about 3

to 6 d B . To add warmth and punch i-n e s s , boost maybe 9 to 12 dB at thel ow fre q u e n cy that sounds best, p e r-haps somew h e re between 40 and1 2 0 H z . As you can see, these twop a ra m e t e rs alone, f re q u e n cy selectand cut/boost, g ive you a terrificamount of spectral flex i b i l i t y.

The final parameter available on aparametric eq, bandwidth (a.k.a. Q),determines the ‘width’ of the cut orboost. That is, as you boost the fre-quency selected by the amountshown on the cut/boost knob, howmuch are the neighboring frequen-cies affected? A narrow bandwidth(high Q) is very focused on the cen-ter frequency, and it introduces asharp spike or notch to the frequen-cy content of the signal being equal-ized. A wide bandwidth (low Q)takes a broader brush approach,pulling up a wide region of adjacentfrequencies along with the centerfrequency being tweaked. Obviously,different bandwidth settings havedifferent uses. During the course ofa project you’ll often find the needfor a range of bandwidth settings.

A 4-band parametric eq has 12c o n t rols on it, so you can select fo u rd i f f e rent spectral targets and shapee a ch of them. This gives us the ab i l i-ty to effect a tremendous amount ofchange to the fre q u e n cy response ofa tra ck . The terrific amount of sonicshaping power that four bands ofp a rametric equalization offer make s

it a favorite part of the channel strip.But other options ex i s t , o f f e r i n gtheir own benefits.

Some equalizers have fixed band-width; the bandwidth is determinedby the designers of the equipment.This type of equalizer gives therecordist the freedom only to adjustthe frequency and cut/boost parame-ters. Because of the downgrade fromthree parameters to two, this type ofeq is sometimes called a semi-para-metric equalizer. Alternatively, theyare often called ‘sweepable’ eq,highlighting the fact that the fre-quency you are cutting or boostingcan be adjusted.

This configuration in which onlyt wo para m e t e rs (fre q u e n cy andcut/boost) are adjustable is ve ryappealing because it’s perfectly intu-i t ive to use. M o re importantly, t h eswe e p able eq is still ve ry mu s i c a land useful in the creation of mu l t i-t ra ck re c o rd i n g s .

D own one more level in flex i b i l i-ty—though not intrinsically in soundquality—sometimes an equalize ro n ly allows control over the amountof cut or boost, and can adjust nei-ther the fre q u e n cy nor the band-width of the equalization shape.G e n e ra l ly called program eq, this isthe sort of equalizer found on homes t e reos (labeled ‘ t reble’ and ‘ b a s s ’ ) .You also see this type of eq on manyc o n s o l e s , vintage and new. It appearsmost often in a 2- or 3-band fo r m :h i g h , m i d , and low.

In the case of your console’s ch a n-nel strip, this same equalizer isrepeated over and over on eve rychannel of the console. If it costs anex t ra $20 to advance the functionalc a p ability of the equalizer from pro-gram eq to swe e p ab l e , that tra n s l a t e sinto a bump in price of more than$600 on a 32-channel mixer. The goodn ews is that well designed progra meq can sound ab s o l u t e ly gorg e o u s .And it often offers frequencies that

a re close enough to the ideal spectra llocation to get the job done on manyt ra cks; often you don’t even miss thef re q u e n cy select para m e t e r.

A slight twist on the idea ab oveleads us to the graphic equalize r.L i ke program eq, this device offersthe engineer only the cut/boost deci-s i o n , fixing bandwidth and fre q u e n cy.

We need a better strat e gy than just ra n d o m ly

buying a few different equalizers. And

snapping up the latest eq du jour won’t

g ua rantee we’ll end up with a coordinated set.

The sonic shaping power that pa ra m e t r i c

e q ua l i z ation offers makes it a favorite pa rt of

the channel strip. But other options exist,

offering their own benefits.

RECORDING MAY 2001

On a graphic eq, s eve ral fre q u e n cybands are presented as sliders ra t h e rthan knobs. The fa d e rs provide agood visual description of the fre-q u e n cy response modification that isbeing applied—hence the name‘ gra p h i c .’ Handy also is the fact thatthe fa d e rs can be made quite com-p a c t . It is not unusual for a gra p h i ce q u a l i zer to have from 10 to upwa rd sof 30 bands that fit into one or twora ck spaces.

G raphic eq is ex t re m e ly intuitiveand comfo r t able to wo rk with. B e i n gable to see an outline of what yo uhear will make it easier and quicke rto dial in the sound you are lookingfo r. Turning knobs on a 4-band para-metric equalizer is more of ana c q u i red taste, and that degree ofc o n t rol isn’t always necessary.

Compressor libertyWhen signals as va r i ab l e ,e m o t i o n-

a l , and dynamic as music signals mu s tbe squeezed into our audio electro n-i c s , t h ey often need to be bro u g h tunder contro l . Th ey don’t fit natura l lyinto the constraints associated withstoring a signal on tape or modulatingit for broadcast out into the ether( t h a t ’s radio or Internet bro a d c a s t ) .

This is bad news for common mu s i-cal elements like sax solos and drumf i l l s . When the signals get too quiet,the music is obliterated by the hiss,r u m b l e , h u m , and buzz of our re c o rd-ing system. Distortion will occur ift h ey get too loud. The ex t ra o rd i n a r i-ly delicate timbre of a glass harmoni-ca or the subtly rich decay of a pianorisks being lost entire ly.

We have no choice when re c o rd i n gve ry loud or ve ry soft tra ck s . Th eloud stuff needs to be turned dow nto avoid distortion, while the quietstuff needs to be turned up to avo i dthe noise floor.

The compressor/limiter automati-c a l ly tames music ever so slightly. I tt a kes up residence on the ch a n n e lstrip because of this fundamentalc a p ab i l i t y.

Nuts & Bolts has raved about thec re a t ive applications of compre s s i o n .It is used to sharpen the tra n s i e n ta t t a ck of a sound; to lengthen itsd e c ay; to ex t ract all those bre a t h s ,gr u n t s , and rattles that perfo r m e rsand instruments make .

Trouble is, c o m p ression genera l lyc a n ’t be taken away, o n ly add e d .This sort of compre s s i o n , t h e re fo re ,d o e s n ’t genera l ly happen duringre c o rd i n g . Radical compre s s i o nm o re typically happens during mix-d ow n , when there is time re s e rve dfor tweaking the compressor until itsounds just right.

Unless aggre s s ive compression is akey part of the sound—tra ck i n gpiano with fierce compression onp u rp o s e , for the timbre it cre a t e s , fo rexample (see ‘ The Nuts & Bolts of

C o m p re s s i o n ,’ in the 3/00 issue)—it’sbest to defer such an ex t reme tonea l t e ration until you are sure itsounds right for the whole tune.Think of cre a t ive compression as aspecial effect.

M o re conserva t ive (ratio of ab o u t4:1 or less) compression on the otherhand is a common part of the re c o rd-ing path. It is an important part ofyour channel path.

As with eq, i t ’s useful to have avariety of compre s s o rs around fo rd i f f e rent applications—they allsound different from one another.

Pursuit of happinessA few channel strips of ve ry good

quality that simu l t a n e o u s ly offer ad e gree of sonic variety can give asmall studio the re c o rding vo c ab u l a ryof the big studios.You don’t necessar-i ly need a humongonormous mixer.A c q u i re or improve your ch a n n e lstrip stra t e g i c a l ly.

Alex Case mistakenly got cable T V,misunderstanding wh at they meant by200+ ch a n n e l s .Help him find thefaders at case@re c o r d i n g m a g . c o m .

When signals as variable, emotional,

and dynamic as music signals must be

squeezed into our audio electronics,

they often need to be brought under control.

Excerpted from the May edition of RECORDING magazine.©2001 Music Maker Publications, Inc. Reprinted with permission.



RECORDING JUNE 2001

Vo lume, Pa rt 1: Fiddling with Fa d e r sBy Alex Case

It is an axiom of the ro ck and roll re c o rding craft thatlouder is better. The good old volume control is a pow-erful audio effect—and eve ry studio has at least one.

This and the next episode of Nuts & Boltswill ex p l o rethe many applications of this humble effect—some obv i-o u s , and some not so obv i o u s .

A sense of balanceConsider the first step in

building a mix. C a re f u l ly, s y s-t e m a t i c a l ly, and itera t ive ly yo uadjust and readjust the vo l-ume and pan position of eacht ra ck until the combinationstarts to make musical sense.At that point the mix is bal-anced—the song can stand onits ow n , and eve ry tra ck con-tributes to the music withouto b l i t e rating other parts.

In pop mu s i c , u s u a l ly thevocal and the snare sit pre t t yloud in the mix, dead center,with the other pieces of thea r rangement (tra cks andeffects) filling in around and underneath. If the guitaris louder than the vo c a l s , yo u ’ re pro b ab ly going to havet rouble selling re c o rd s . If you can’t hear the pianowhen the sax play s , the song loses musical impact. S oyou wo rk hard to find a balance that’s fun to listen to,supports the mu s i c , and reveals all the complexity andsubtlety of the song.

This first step of a mix session is re a l ly a part of e v e r ys e s s i o n . For tra cking and ove rd u bb i n g , the playe rs can’tp l ay, the engineer can’t hear, and the producer can’t pro-duce until the signals from all the live micro p h o n e s ,re c o rded tra ck s , and effects are brought into some kind

of balance. R e lying almost entire ly on volume contro l s ,balancing a mix is one of the most important skills anengineer must master.

On the levelIf music is picked up with a

m i c ro p h o n e , you’ll need am i c rophone pre a m p l i f i e r.Guess what? Mic preamps arenothing more than vo l u m ed ev i c e s . And we ’ ve got to setthe volume just right when were c o rd to tape or hard disk(see sideb a r ) .

Because all equipment hassome noise, we natura l ly try tore c o rd music at as high a leve las possible so that the mu s i c a lwave forms drown out thenoise floor. So it seems truethat louder is indeed better.The question is, h ow loud?

Th e re are two differe n ts t rategies for setting re c o rd-ing leve l s , depending on

whether the storage format is digital or analog .Yo u ’ ve undoubtedly heard that for digital re c o rd i n g ,

the goal is to “print the signal as hot as possible withoutgoing ove r.” L e t ’s think a little bit about what that means.

P re s s u re in the air becomes voltage on a wire (thanksto the micro p h o n e ) , w h i ch then becomes nu m b e rs ontape or disk (thanks to the analog-to-digital conve r t e r ) .As the music gets louder in the air, the corre s p o n d i n gvoltage gets higher on the mic cab l e . But at some pointthe nu m b e rs getting stored by the digital system can’tget any bigger—it maxes out in mu ch the same way thata child counting on his or her fingers runs out at ten.

In pop music, if the guitar is

louder than the vocals, yo u ’ r e

going to have trouble selling

records. You work to find a

balance that’s fun to listen to,

yet supports the music and

reveals the song’s subtleties.

RECORDING JUNE 2001

At that point the digital data no longer fo l l ows themusical wave form (see Fi g u re 1). This is a kind of distor-tion known as hard clipping. The peaks are clipped off,gone fo reve r.

O bv i o u s ly, the way to prevent this kind of distortion is tom a ke sure the analog levels going into the digital re c o rd e rn ever fo rce the system past its maximu m . The meters willhelp you here . Digital systems genera l ly have meters thatm e a s u re the amplitude of the signal in decibels below fulls c a l e ,w h i ch is the “ten fingers ” point at which the digitalsystem has re a ched its maximum digital va l u e .

If you are intrigued by the wave form shown in thel ower part of Fi g u re 1 and are wondering what it soundsl i ke , you might want to ove rd r ive the digital system onp u rp o s e . Be my guest, but be care f u l . Fi rs t , monitor at al ow leve l . This kind of distortion is full of high fre q u e n cye n e rgy that can melt twe e t e rs .

Second, listen carefully. This type ofdistortion is extremely harsh; it’s not aparticularly musical effect, so it’s bestused sparingly if at all. But of courseit’s not strictly forbidden—musictends to rebel.

On analog magnetic re c o rding sys-t e m s , you typically re c o rd as hot asp o s s i b l e , and occasionally go ove r.U n l i ke digital audio, a n a l og audiod o e s n ’t typically hit such a hard andfast limit; instead, it distorts gra d u a l-ly as you begin to exceed its comfo r tra n g e . This gradual distortion at thepeaks is called soft clipping, s h ow nin Fi g u re 2.

At lower amplitudes, the analogmagnetic storage medium tra cks ve rya c c u ra t e ly with the wave fo r m . As theaudio signal starts to get too loud,the analog storage format can’t ke e pu p. It starts to re c o rd a signal that’snot quite as loud. As it runs out ofs t e a m , it does so gra c e f u l ly. L o o kc a re f u l ly and you might notice thatove rd r iven analog tape looks a lotl i ke compre s s i o n .

A quick glance at my effects ra ckreminds me: compression is an effect.I ’ ve bought ra ck spaces and pulld own menus full of compre s s i o n . C a nyou ove rd r ive analog magneticre c o rd e rs for an effect? You betch a .So we find ours e l ves using volume asan effect simply by setting levels aswe re c o rd mu s i c .

A n a l og mach i n e s , with faint tapeh i s s , p refer audio wave forms withoutquiet passages (low vo l u m e ) . Wh i l edigital systems don’t have tape hiss,t h ey do introduce other sonic artifa c t sat low leve l s , as we’ll discuss in af u t u re Nuts and Bolts article.

S t i l l , this low noise floor was a dri-ving fo rce in the transition from ana-l og to digital audio. Classical and jazze n g i n e e rs have to re c o rd acoustic

music with a wide dynamic ra n g e — music that sometimeshas long, o p e n , quiet spaces. For this genre of re c o rd i n g ,the nearly silent noise floor of digital storage was ad ream come true.

R o ck and ro l l , on the other hand, tends to have a mu chm o re narrow dynamic ra n g e . The song kicks in and ra re lylets up; hiss can’t raise its ugly head over the scre a m i n gvocals and grinding guitars .

M o re ove r, as we know from listening to ra d i o, l i s t e n i n gto great mixes, and experimenting in our own studios,ro ck and roll also loves a bit of compre s s i o n . As a re s u l t ,even in this ve ry digital age, m a ny pop re c o rds are stillre c o rded onto analog tape.

A dding further iro ny, these days digital audio dev i c e sa re consistently less ex p e n s ive to own and operate thanp rofessional analog audio tape mach i n e s . To d ay, we essen-t i a l ly must pay ex t ra for the tape compression effect.

RECORDING JUNE 2001

G iven a ch o i c e , the sound qualityd i f f e rences between analog and digi-tal re c o rd e rs as they react to wherethe volume knob is set are a key fa c-tor in selecting which format to useon a re c o rding pro j e c t . And here ’sanother clear case of using the vo l-ume knob as an effect.

FlavorIn the re c o rding studio, we gener-

a l ly run into two types of analog vo l-ume control: the va r i able re s i s t o rand the voltage controlled amplifier.

Electrical resistance is a pro p e r t yof all materials describing how mu cht h ey restrict the flow of electricity.Materials with ve ry high re s i s t a n c ea re classified as i n s u l at o r s; they pre t-ty mu ch don’t conduct electricity ata l l . We appreciate this property whenwe handle things like power cord s .

At the other ex t re m e , d evices withve ry low resistance fall into the cate-g o ry of c o n d u c t o r s. Copper wire is ac o nvenient ex a m p l e . The copper

within that power cable conductselectricity from the wall outlet to thepiece of audio equipment, g e t t i n gthe LEDs to flicke r, m o t ivating them e t e rs to twitch , e n abling us tom a ke and re c o rd mu s i c .

The volume knob on a homes t e re o, electric guitar, or analog syn-t h e s i zer is (with a few model-specificexceptions) a va r i able re s i s t o r. Set toa high re s i s t a n c e , electricity hast rouble flowing and the volume isa t t e nu a t e d . To turn up the vo l u m e ,l ower the resistance and let the

audio wave form thro u g h . Also calledp o t e n t i o m e t e rs , we typically think ofthem as simple volume contro l s .

In the re c o rding studio, we have tolook more closely at our volume con-t rols because there is a second type:voltage controlled amplifiers . H e pcats resort to three letter acro ny m s —V CA . The idea behind them is simpleand cleve r.

If the fader on an analog console isa potentiometer, it makes sense to

B e cause all equipment has noise, we try to

record music at a high level so that the

music drowns out the noise floor. So it

seems true that louder is indeed bett e r . . . .

Excerpted from the June edition of RECORDING magazine.©2001 Music Maker Publications, Inc. Reprinted with permission.



RECORDING JUNE 2001

p i c t u re the fader as a va r i able re s i s-t o r. But in the case of a V CA , t h efader that sits on the console is sepa-rated from the audio by one laye r.Instead of having that slider on theconsole phy s i c a l ly adjust the re s i s-tance in a potentiometer, it adjusts ac o n t rol vo l t a g e . This control vo l t a g ein turn adjusts the amount of gainon an amplifier.

Most compre s s o rs use V CA s , w h i cha re capable of reacting to vo l t a g echanges ve ry quick ly. And for con-s o l e s , the only other way to havesomething other than the engineeradjust the level would be to stick amotor on the fa d e r. This is a pricey,complicated option, but motorize dfa d e rs are certainly ava i l ab l e — a n dat an eve r- d e c reasing price.

AutomationMix automation can do many

things these day s . If you have a hipdigital audio wo rkstation or digitalc o n s o l e , you can automate it so thatit wa kes you to music first thing inthe morning (noon), starts the coffeem a ke r, and draws a warm bath.

While this is all quite useful,automation is almost always justused for two ve ry simple pro c e s s e s :fader rides and mu t e s . The point ofpushing fa d e rs and pressing mu t ebuttons? Controlling vo l u m e .

Not too long ago even the fanciestconsoles offered the ability to auto-mate only the faders and the cut(mute) buttons. Studios spent a fewhundred thousand dollars on a topof the line, state-of-the-art consoleand still couldn’t automate pan pots,aux sends, equalizers, compressors,or reverbs.

...The question is,

how loud?

RECORDING JUNE 2001

But as we know from all the music released from thebeginning of time up to about 1995, ex t ra o rd i n a r i ly elab o-rate and complicated mixes we re built with this re l a t ive lylimited amount of automation capab i l i t y. C l ever vo l u m ee f f e c t s — m o s t ly using V CA-based automation—are the key.

For ex a m p l e , using the humble mute sw i t ch , the mixengineer controls the mu l t i t ra ck arra n g e m e n t . Cut thebass in the ex t ra bar befo re the ch o r u s , pull the flute outof the horn part until the last ch o r u s , e t c . This sort of mixm ove happens throughout pop mu s i c . But ch e ck out anex t reme example by listening to U2’s A chtung Baby. Th ealbum begins with some heavy cut activity as the drumsand bass enter at the top of the first tune of the album,“Zoo Station.”

Automating fader rides in support of the arra n g e m e n tis a natural application of automation. M aybe it make ssense to push the guitar up in the ch o r u s e s , pull theC h a m b e rlin down during the guitar solo, and such .

I d e a l ly, the band (maybe with the advice of a pro d u c e r )gets these dynamics right in their perfo r m a n c e . But in thes t u d i o, the full arrangement of the song may not comet ogether for seve ral months as ove rdubs are gra d u a l lya dded to the tune. Fader rides may be just the ticket tohelp this assembly of tra cks fall into a single piece of mu s i c .

Volume changes are automated just to keep the song inbalance as mu l t i t ra ck components of the song come andg o. But it’s usually a good idea to keep these moves quitesubtle; they ’ re aimed at the musical interp retation of them i x , t rying to make the song feel right. With few ex c e p-t i o n s , it should pretty mu ch never sound like a fader wa sm ove d . L i s t e n e rs want to hear the mu s i c , not the console.

Another automated volume effect is the Au t o m a t e dS e n d . Some ve ry sophisticated mix elements can be cre-ated this way sends. Automation is employed to add richand spacious reverb to the vocal in the bridge only, i n t ro-duce rhythmic delay to the back ground vocals on keywo rd s , i n c rease the chorus effect on the orch e s t ra lstrings in the ve rs e s , a dd distortion to the guitar in thefinal ch o r u s , e t c .

The automated send—just another volume effect—offersa way to layer in areas of more or less effects, using noth-ing more than straight fo r wa rd fa d e rs and cuts automation.

We’ll keep digging deeper into volume next month,m oving beyond fa d e rs and exploring the finer points ofc o m p re s s i o n , ex p a n s i o n , g a t i n g , and tre m o l o, and howvolume affects the eq curve . S t ay tuned.

Getting paid to play the volume control is why Alex Casebecame a recording engineer. He used to do it for fre e .S p e a kup to case@re c o r d i n g m a g . c o m .

Dynamic RangeMusical dynamics are so important to composition and perf o rmance that they are notat-

ed on every score and governed closely by every band leader, orchestra conductor, andmusic dire c t o r. Making clever use of loud parts and soft parts is a fundamental part of com-position and arr a n g i n g .

In the studio we must concern ourselves with a diff e rent sort of dynamics: AudioDynamics. Follow along in Figure 3 as we keep careful control over the range of ampli-tudes that we encounter when re c o rding audio signals.

Exploring the upper limit of dynamic range comes naturally to most of us. We turn itup—whatever ‘it’ is—until it hurts our ears, our equipment, or the music. Cranking it ’t i l lit distorts. It seems to be the sole determinant for the position of the volume knob on mostguitar amps (including mine), car radios (at least for the car in the lane next to me),p o rtable stereos (the jogger who just passed me), home stereos (my neighbor in my fre s h-man year college dorm)… Here we have encountered a basic pro p e rty of all audio equip-ment: turn it up too loud, and distortion re s u l t s .

At the other extreme (turning it down too much) lives a diff e rent audio challenge: wes t a rt to hear the inherent noise of the audio equipment we are using. All audio equipmenthas a noise floor—equalizers, compressors, microphones, and even patch cables. Yu p .Even a cable made of pure gold manufactured in zero gravity during the winter solstice ofa non-leap year will still have a noise floor, however faint.

A constant part of the re c o rding craft is using our equipment in the safe zone betweenthese two extremes. This is the dynamic range, and it’s quantified in decibels (dB). The tar-get nominal level is typically labeled 0 VU (that’s a zero, not an O). At 0 VU the music getst h rough well above the self-noise of the equipment, but safely under the point where its t a rts to distort.

If we re c o rded pure sine waves for a living, we’d turn the signal up right to the pointof distortion, back off a smidge in level, and hit Record. However, the amplitude of a re a llife musical waveform races wildly up and down due to both the character of the part i c u-lar musical instrument and the way it is being played.

Electric guitars amps cranked to the limit—at that much savored edge of becoming fireh a z a rds—have very little dynamic range. If you haven’t already witnessed this yourself,re c o rd a guitar the way Spinal Ta p ’s Nigel Tufnel does—with the amp set to eleven. Yo u ’ l lo b s e rve the meters on your console and multitrack zip up at the downbeat. And they bare-ly move until the end of the song.

P e rcussion, on the other hand, can be a complicated pattern of hard hits and delicate taps.Such an instrument is a challenge to re c o rd well. The musical dynamic range of the instru m e n tmust somehow be made to fit within the audio dynamic range of your studio’s equipment.

Accommodating the unpredictability of all musical events, we re c o rd at a level wellbelow the point where distortion begins. The amplitude ‘distance’ (expressed in decibels)between the target operating level—0 VU—and the onset of distortion is called h e a d-ro o m. This gives us a safety cushion to absorb the musical dynamics without exceeding theaudio dynamic range of the gear.

The relative level of the noise floor compared to 0 VU, again expressed in decibels, isthe signal-to-noise ratio. The trick, of course, is to send your audio signal through at a levelwell above the noise floor so that listeners won’t even hear that hiss, hum, grit and gunkthat might be lurking down in the depths of each piece of equipment.

Making effective use of dynamic range influences not just how we re c o rd to tape, buthow we use a compre s s o r, a de-esser, a reverb, or any other piece of gear.

Look ca r e f u l ly and you might

notice that overdriven analog

tape looks like compression.