DavePensado's impressive resume includes top pop and R&B artists suchas Christina Aguilera, Pink and Boyz II Men. Tim Palmerhas mixed for a range of artists from David Bowie to U2 to OzzyOsbourne to Faith Hill. Veteran engineer RogerNichols's client list runs the gamut from Steely Dan to Take 6 to BelaFleck and the Flecktones.FIG. 1: TimPalmer observes that too much rhythmic correction can eliminate thenatural flamming produced when an ensamble hits a simultaneous beat.Slicing it up and putting it "on the grid" can actually make it soundthinner. FIG.2: One way Roger Nichols adds variety to a vocal track is to split thevarious sections of the song onto separate tracks and then usedifferent reverbs and even different EQ and compression oneach.FIG. 3: ThisPro Tools screen shot from Dave Pensado's mix of "Lady Marmalade"(performed by Christina Aguilera, Pink, Mya, and Lil' Kim on theMoulin Rouge soundtrack) demonstrates how you can add dynamicsinto programmed music. Notice how the volume levels (as depicted by theblack lines) have been altered for the song's various sections.
Mixing a song is like driving on a very busy and confusing freeway.There are lots of choices and decisions to make, and an incorrect onewill send you off in the wrong direction. You're surrounded by baddrivers (unfocused producers and egocentric musicians), who candefinitely delay your arrival. Bad monitors are like potholes, anddistractions abound, all threatening to destroy your concentration andset back your timetable. Misleading signs are everywhere, and amoment's hesitation can lead to a nasty accident. Getting to yourdestination will require a good road map; fast, accurate decisions; anda will to win.
So how do the seasoned vets of the mixing world survive this chaosday in and day out? I corralled three of the best and asked them. Ofcourse they have great equipment, but take it from me: if RogerNichols, Tim Palmer, or Dave Pensado were working in your home studio,they would still get a great mix. Using their ideas will get you whereyou're going much faster, and the ride will be a whole lotsmoother.
Tim Palmer is a modern-day scion of the British sound who works on,as he says, “everything from Goth and metal to pop.” Hisextensive credits include U2, Tears for Fears, Pearl Jam, OzzyOsbourne, Faith Hill, and David Bowie, among many others. When wespoke, he'd just finished mixing a live album for singer-songwriterJason Mraz and a project for the subversively glam Kill Hannah, and waspacking for a trip to Finland to work with HiM.
The perennially busy Dave “Hard Drive” Pensado iscontinually in demand by R&B and pop royalty. If you watch musicvideos or listen to the radio, you've likely heard some of hischart-topping mixes such as Pink's “Get the Party Started,”Christina Aguilera's “Beautiful” and “Dirrty,”and the diva rendition of “Lady Marmalade” from theMoulin Rouge soundtrack, which featured Christina Aguilera,Pink, Mya, and Lil' Kim.
Pragmatic and outspoken, Roger Nichols is an iconoclastic Grammywinner who's legendary for his work with the pristine madness of SteelyDan (13 albums) as well as John Denver (27 albums), Rickie Lee Jones,Take 6, Béla Fleck and the Flecktones, and Roseanne Cash, amongmany others. On the day I visited him in the studio, he was completingmixes for a Rogers and Hammerstein tribute album that featuredperformances by such artists as Phoebe Snow and Toots Thielman.
These multiplatinum mixers have very different styles, but theyshare a lot of the same philosophies. I found out how they approachtheir mixes, and the mental processes that enable them to deal with allthe inherent obstacles.
HAVE A PLAN, STAN
All three mixers were emphatic in stating that diving blindly into amix, trusting that the muse will lead you to inspiration, is a recipefor disaster. Before you begin, they agreed, you must take the time tofigure out what it is that makes the song work best.
“Maybe it's the vocal, maybe it's the lyrics, maybe it's thegroove or a hook,” says Palmer. “If it's the lyrics, youhave to think from a vocal perspective: is the music working to helpthe story to come across? In another song, the key thing might be thatamazing and memorable guitar lick. Or, sometimes a song has a greatgroove. Then you've got to make the drums and bass really solid so theydrive the whole thing along. It's your job to figure out what thatspecial something is, and then carve your mix to bring it out. You mustplay to the strengths of the recorded material.”
“What's important is that you have a concrete vision of whereyou want to go,” Pensado agrees. “Then you'll find a way toget there. One of the main differences between me and a lot of theproducers I work with is that I have the skill to do the mix in a fewhours. They might be able to do a similar mix, but it would take themat least a week. I think the lesson there is to mix for the forest andnot the trees. When you start mixing for the trees, you paint yourselfinto a lot of corners that it's difficult to get out of. You want tokeep yourself on line with the vision you started out with.
TIPS FORBETTER MIXING
• Beforeyou start moving faders and twisting knobs, have a clear vision of whatyou want the mix to sound like when it's finished, and stick toit.
• There'sno hard-and-fast rule as to whether to build your mix one element at atime or start with all tracks up. Be flexible, and let the needs of thesong determine your approach. Avoid doing detailed tweaking onindividual elements until you have a good idea of what the rest of thetracks sound like.
• Don'tdehumanize your tracks with overly aggressive pitch and timecorrection. Remember that subtle irregularities in rhythm and pitch canoften make the overall song sound better.
• Whenequalizing or adding an effect to a particular element, always takeinto account how that change will affect the mix as a whole.
• Too muchprocessing can negatively affect the overall sonic quality, so bejudicious. When equalizing, subtract rather than add whenpossible.
• Preserveyour dynamics. Don't go overboard with compression. When working withprogrammed tracks, add in enough dynamics to make the songexciting.
• Monitoron a variety of speakers at various levels, and try to check your mixin different acoustic environments.
• Don'tstay on one mix for too long. Take breaks, and even switch to adifferent song for awhile if possible. When you come back to a songafter a prolonged break, listen through it and create a list of thechanges you want to make.
• Don'tfeel constrained by convention. Experiment when you can, and if itsounds good, do it!
“Here's something to think about. When the old radio shows ofthe '40s wanted to create the sound of horse's hooves, they used acouple of coconut shells. A real horse walking through the recordingstudio wouldn't have sounded as much like a horse to the audience asthe sounds that they created. That tells us that the original soundyou're given is not as important as the image you want to convey in themix.
“For me, and I think for a lot of mixers, it's very visual:you tend to want to see the horse, or, in the case of a great rock mix,the color of the guitar someone is playing. It actually becomes like apainting. Some of the early Nirvana mixes, for example, were veryimpressionistic. There are impressionistic mixes, abstract mixes,realistic mixes, paint by numbers mixes, where you just fill in thecolors, and, of course, [laughs] Jackson Pollackmixes.”
Rough mixes are what you throw down at the end of a session to giveyourself a working road map. However, they often take on a life oftheir own. Everyone has a horror story about the artist (ask BruceSpringsteen's engineer about the trauma of Nebraska) or A&Rperson who got attached to a seriously flawed rough mix and could neverbe convinced that there might be a better way to go. In those cases,trying a different direction is just a waste of time. But most peoplewant to be open to creative possibilities and to the hope that the mixwill take their song to a whole new level. So, isn't it best to startwith a completely clean slate? Well, no, not according to thesepros.
“It's very helpful to listen to the rough mixes beforebeginning,” says Palmer. “You can definitely get off on thewrong foot if you go straight in and start pushing up faders withoutknowing what your goal is. If you have engineering ears, when you pushthe bass drum up you can tend to think, ‘That bass drum is dull,I need to do this and that to it.’ But maybe, to make the songwork well, the drums need to be dull and not too defined. Here, if youdidn't work out what was important before you started, you can make allthe wrong decisions. You get to the end and realize that you've wastedyour time. The song was much better with the low-fi drums.”
“In a rough mix I'm listening for the elements,” saysPensado. “The most obvious question would be, Are there a lot ofeffects like reverb and delay on the instruments, or was the mix moreup-front, dry, and in-your-face? Where is the vocal relative to thetrack? Because if the A&R guy has been listening to a loud-assvocal for six months and you put it ‘right,’ you could havesome problems. The other thing I listen for is — I hate to usethe word, but — vibe. When you sit down at a console you'resitting down at a blank canvas, and the first stroke you put on thatcanvas doesn't come easy. The first thing you sit down to do is thehardest on any creative project. The rough mix gives you a startingspot.”
Nichols also finds roughs useful, but for different reasons.“I think the whole process of making a record simply revolvesaround listening and paying attention,” he says. “That maysound obvious, but in my opinion, a lot of people aren't doing it. Forexample, I get stuff to mix that other people have recorded where,especially if it's a Pro Tools session, they automatically put alimiter on every channel. Just because they can, or because they thinkthey should. It doesn't matter what it is. It could be an emptychannel, but it's got a limiter on it!
“Hopefully they recorded in Pro Tools and they can send me thesession file so I can take the compressors off and start over. If Ihave the rough mixes, I'll listen to what they did first to get theconcept of what they were trying to do overall. Did they like thevocals or the percussion loud? Then I'll start over, without theircompression, to get to that platform of the loud vocal or percussion,or whatever it was that defined the whole feel of the song.”
WHAT'S UP FIRST?
Whether you're laying out a 100-input console or mixing from acomputer-based workstation, you've got to begin with some piece of thesong's instrumentation. Some engineers will throw up all the fadersright off the bat for a quick reference balance. Traditionally, manyother people begin with drums. Pensado tries to let the song determinehis first moves.
“For some songs you might start with the rhythm track, otherswith vocals, some with just the kick drum,” he explains. “Agood friend of mine who's a mixer always starts with vocals. SometimesI do too, especially if it's a group like Take 6 or Boyz II Men that'sknown for their vocals. I'll bring the music in around them. Somethingto keep in mind is that, unless you have an unlimited amount of time— which you never do — with a heavily laden vocal track youdon't want to be working on your vocals at midnight. If I only have oneday to do the mix, and the vocals are important, I'll get to the vocalsearly on, if not first. Then, when it's 11 o'clock at night, and I'mworking more on instinct than creativity, I have confidence that myvocals are sounding right.”
Nichols takes a holistic approach. “I've seen people start amix by pushing up the kick drum before they've listened to thesong,” he says. “They've never heard the song, but they'retrying to get a good sound on the kick. Then they work on the snare,then the bass. When they finally turn up all the instruments, itdoesn't fit. Then they have to mess with this, mess with that, and theyend up wasting a lot of time.
“The first thing I do is turn up all the faders and get a feelfor what the song is and what the mood is supposed to be. Then I'll geta rough sort of balance between the instruments — this is howloud the strings should be, and the piano and the vocal. Then, if Ihear that the vocal kind of sounds muddy against something else, I'llstart equalizing a little bit, and fixing things so they meld together.Then I'll go back and listen to the drums by themselves, and the pianoby itself, and refine that. I'll go around a couple of times in a bigcircle and then, usually, it's close to done, except for vocalrides.”
EVERYTHING IN ITS PLACE
Mix engineers must routinely deal with multitudes of tracks, lots ofgood parts, a few bad ones, and often a band full of musicians who allwant their individual parts turned louder. But beyond just finding theoptimal levels, the engineer must figure out the best stereo placement,fitting each piece into the overall puzzle. The task is complicated bythe fact that vocals, keyboards, and guitars often use a lot of thesame frequencies to assert their presence. So when you're facing thatdaunting wall of midrange, skillful placement along with judicious EQand filtering will really pay off.
Q&A: USING CDS FOR REFERENCE
Do you reference your mixesagainst finished CDs during the mixing process?
Palmer: If I'm not familiarwith a room, yes, I'll listen to some other things I'vedone.”
Nichols: If I am in a studioI have not mixed in before, I will play back something I did somewhereelse that came out well so I can hear what the speakers are doing inthe room. Once I have it scoped out I use what I heard to modify themixes so they match when I leave. I do not play the CD any more duringthe mixing.
Part of it is the Meyer HD-1speakers I use. I either rent or bring my own. I have used the HD-1s oneverything I've done since 1989. Most studios where I have used themnow own a pair so I don't have to cart mine.
Pensado: I like to referenceto CDs, and I like for my clients to hear it. It gives people in theroom who aren't accustomed to my monitors a frame of reference.Actually, I've been told that I listen to way more music during my mixprocess than most engineers. I always have a music television stationon and a couple of CDs in the player. I'll just randomly hit it. Ifwhat's in there sounds better than what I'm doing, I keep working. Ifit doesn't, I'll print!
“Everything can't be in the middle,” says Nichols, asurvivor of major track wars. “Unless you want to make a monorecord, which is okay. But that's another challenge. If there are twothings going on during the song — say, both a guitar and pianoplay throughout — then my first inclination is that maybe thepiano should be a little bit on one side. Even though it's in stereo.So I'd cheat it over to one side, and I'd cheat the acoustic guitarover to the other side, so you can sort of hear what they're doing allthe way through. They're the bed you're going to use.
“Then I'll probably put the bass in the middle. But not 100percent of the time! Sometimes whatever the bass is doing gets muddybecause of what the piano is doing. Then, maybe the piano should be alittle to the right, and maybe the bass should be off center just alittle bit. If that's a mistake, you put it back in the middle. You'realways trying little refinements of placement to ensure that everythingadds to the song instead of subtracting from it.”
“It's not just about creating a nice balance and makingeverything sound good,” says Palmer. “Maybe some thingsshouldn't sound so good; maybe something should be way out of balance.When you're first learning, because you have pride in your work, youtend to think of every sound as an individual item that should soundgreat. But as you move along through the years, you realize thatsometimes, for one thing to sound good, something else may have tosuffer.”
So too much use of the solo button can steer you wrong?“That's right,” Palmer says. “You have to look at theoverall picture. It's not always about perfection. Sometimes you cantake a part out of context, and it's not so good, but within contextit's amazing.
“You listen to a sound and think ‘That bass playing isrough, I don't know about that.’ Then you put it with the drummerand it sounds great. Here's your plan helping again. Because if you goin and start fixing that bass up, putting it in time with the bass drumand lining it all up, you may have just ruined the wholegroove.”
IN AND OUT OF TUNE
That leads us to a theory put forth by Palmer. On the great oldrecords we all seem to love, the bands were tuning mostly by ear,especially during the basic tracks. Everyone may have started outreferenced to a tuner, but as the overdubs progressed, things shiftedback and forth a bit. Tuning was a bit more about taste than exactscience.
“Think of the perfect tuning in a song as a thin, straightline,” he says. “In those days, after the basic track waslaid down and a few overdubs were added, some of the tuning mayactually had drifted, but not to the detriment of the sound. Inreality, some of the parts may be a little sharp (above the line ofperfect tuning) and some players may be a little flat (below the lineof the perfect pitch). The tuning line is now a lot wider and thicker,and so is the sound!
“When the vocalist went to sing, he obviously aimed for thecenter of the pitch, but had a lot of space either way. He could go alittle sharp with emotion, or flat a touch, and it still sounded good;it could actually add to the performance. Now, with each overdubeverybody retunes. You get an overall pitch ‘line’ that'svery thin, and that doesn't give much room for expression. When thesinger performs, it's not as much fun! If he goes a little sharp, itdoesn't sound good anymore. So what do you do? You tune him or her aswell. They join the line and now the whole thing sounds small!
Pensado thinks that although pitch correction can be helpful, it'soften overused and can detract from the feel of a song. “A greatexample is blues,” he points out. “When you go up a minorthird, sometimes it sounds better when you don't quite get there.Violin players, when they're going up the scale, they play a differentinterval between B and C then when they're coming down the scale. Sothe human ear readily accepts imperfections in pitch, and it's a very,very personal thing how you interpret it.” He counsels aselective approach to pitch correction, rather than just strapping itacross an entire track: “Find those notes that just could neverbe resung [or replayed] again, correct those, and move on.”
Rhythmic feel can also be damaged by too much correction, saysPalmer. “If you chop the whole thing up into enough pieces soit's all bang on a grid, the ear doesn't hear any natural flamming ofinstruments any more. In many cases a bit of that flamming makes themusic sound bigger and more appealing. With a great band — LedZeppelin or whoever — you have the drummer who might play reallylaid back, the bass player who's slightly ahead of the beat, and theguitar player who's slightly behind. When they hit that big chordtogether, you've got ‘one, two, three,’ all hitting yourear at slightly different times. It sounds huge, with just enoughnatural offset [see Fig. 1]. Think of an orchestra; one of themain reasons it sounds good is because they are a bit out of time andout of tune!”
LESS CAN BE MORE
Most recordists know that the more processing you use, the morephase shift and artifacts you introduce into your signal path, and thisis generally detrimental to a good sound. Yet it's so tempting, and soeasy, to reach for those powerful knobs and that brand new plug-in.Nichols deals with this conundrum by taking a subtractive approach.
“Say I'm working on a vocal,” he says. “Mainly Iwant it to sound the way the singer really sounds in the room. The wayI do that — instead of trying to add something to make it better— is to figure out what's bad and get rid of it. If you start outsaying, ‘This needs to be a little brighter, and that needs tohave more bottom end,’ pretty soon you've got EQ adding 57 littlethings, when maybe all it really needed was for you to find the badpart and remove it. All of a sudden it's better because you took outthe sound of the room where the vocal was done that wasn't a very goodroom. Or you took out some woofy part of the piano, where you had toput the lid on a short stick with a blanket over it. If you look forthe bad thing and remove it, you're about 60 percent there. Thenyou can say, ‘Oh, a little brighter here…”
Q&A: LEAVING ROOM FORMASTERING
How much room should you leavefor the mastering engineer to add the finishing touches?
Palmer: I do like to leavesome room for mastering. I don't like to kill the mix by adding toomuch compression too soon. I'd rather keep an album sounding as dynamicas possible. If more compression is needed for the radio, the masteringguy can squish it a little further.
Pensado: I'm fortunate thatI work with the same few mastering engineers I trust, and I prefer notto put too much compression on my stereo bus if I know they are goingto work on it. These guys are great; they can do it as good or better,and I've saved myself going through those electronics, which they'regoing to go through anyway.
Nichols: Because I actuallyhave a mastering business and have been mastering since themid-‘70s at ABC/Dunhill, I know when to quit! The monitors andthe monitoring environment are the important things. If you are notexactly sure about how things are going to translate, then giveyourself a little more room for mastering. If the monitors andenvironment are perfect, you can cut it a lot closer.
You can also leave yourselfflexibility with alternate mixes. In Nashville, they usually printabout 20 versions of each mix. Vocal +1 [dB], vocal -1, backgrounds +1,vocal 0, backgrounds -1 — all the permutations you can imagine.When the songs get to mastering, if the vocal is too low in the chorus,they just edit in the version with the vocal up louder.
Palmer agrees. “It's very easy to over EQ,” he says,“it can be better to look for frequencies that are unnecessaryand remove them. If I'm looking for a frequency to remove, I'll boostthe EQ radically and sweep it until the offending frequency sticks outlike a sore thumb. Then I know exactly where to cut.”
“You should never put something on the mix because youperceive that's what you're supposed to do,” says Pensado.“For instance, If I'm doing a hip-hop mix I probably won't putany compression across the stereo mix. If I'm doing a ‘LadyMarmalade,’ I've got it crunched. Why? I knew it was going to bea pop song getting a lot of airplay on a wide variety of stations, andI wanted my compressor to be controlling things more than theradio's.
“But say on ‘Beautiful,’ by Christina Aguilera,there's nothing on the stereo bus — no EQ or compression. Wewanted an old-fashioned sound. There are some songs I mix, say onPink's ‘Get the Party Started,’ where I wanted you to hearthe mix. I wanted it to be part of what was hooky about the song. Butwith ‘Beautiful,’ if anything at all made you think ofanything besides Christina's voice, it was a failure. So I‘anti-mixed’ it. I really worked on subtracting mypersonality from the mix.”
DON'T DISTURB THE GROOVE
“Groove is a very personal experience,” says Palmer.“It's really the amount of ‘out of timeness’ that youpersonally enjoy and accept. Sometimes I'm working with an artist whowill constantly want to replay a part because they feel it is out oftime, while I'll be telling them it sounds great. Their personalreference point is just different.
“It's important to keep as much of that feel as you can. If Ihave to move a few parts around during a mix I will move them by ear.Remember, it's important to listen to music rather than look at it. Alot of people now just look at the screen and make decisions based onthat. But if you listen, it might sound good even if it looks wrong.Using the screen can, of course, be helpful at times, especially ifyou're wondering why something's not working. It's convenient to have alook around and see what's going on. But if you're finding problemsbefore you've heard them, that's when you get into trouble.”
“To me, the groove is the mood of everything taken as awhole,” says Nichols. “The internal rhythms, the delay onthe guitar, the little space between the vocal and the reverb, the waythe kick and bass connect, the sustain of the acoustic piano, thebounce of the congas. It's all the little nuances that make everythingwork together.
“There doesn't have to be a lot of stuff on the record, itjust has to feel right. I have gone into the studio intent on doingoverdubs on a song, and when I put up the mix and listened, it didn'tneed anything else. We'd try anyway, but at the end of the day, wedidn't use anything we added. The song was done, but we hadn't realizedit.”
THE EFFECT OF EFFECTS
“I like to find one reverb and use it as a main reverb for thewhole song, sending everything to it a little bit,” says Nichols.“Then for the vocal, I'll use three or four different reverbs,splitting out the vocal to different tracks [or channels]. When thevocal gets louder, the quality of the reverb changes because it'ssending to a different unit. When the vocal is soft, maybe the reverbhas a long decay time so it's nice and moody and has all this bigambience. That long, three-second reverb doesn't work when the person'sscreaming into it; it will be two choruses before the thing finallydecays.
“Taking the vocal apart and splitting it out to differenttracks allows you to give the loud parts of the vocal a completelydifferent reverb send and completely different compressor or EQsettings than the soft parts [see Fig. 2]. Of course, thosethings are easy to do using Pro Tools. You can clone tracks so you'vegot two vocals or two snare drums — one for the cross stick,another when he's playing the snare — and make them different.Doing those things will really make the mix easier.”
Pensado says, “I look at reverb in two ways. First, it's afront to rear panner. If you want something to go to the back of themix, put reverb on it. This is the most fundamental psychoacoustictrick we have. Of course, sometimes I use reverb just because it soundsgood. Who cares if it sounds like it's in a sewer pipe, as long as it'sa great effect?
“I don't like to pan my effects returns hard left and right.I'd rather get identical units or plug-ins, then pan one hard left(both returns) and the other hard right (both returns). Select thesame, or similar, programs on each unit and tweak them so they'reslightly different.
“I like to use 30 to 40 delays, from a 128th note to a halfnote, panned all over the place. I'll intentionally make some of thedelays out of time, creating what I call asymmetrical delays. Just movethem until they sound good. Plug-ins are perfect for this, because somehave 6 to 12 delays built in.”
Some engineers like to precisely time their delays to multiples orfractions of a song's tempo. Palmer says, “I do that if I'm goingfor a specific effect, but often I think delays are better when they'reout of time and create a bit of rubbing. Especially with music being soprecise these days, it can be good to loosen things up a bit and relaxthem. Maybe a delay that's bang on will just disappear into the snaredrum sound, and you'll find yourself turning it up louder and louder tohear it! If it's a little bit out of time, you'll get a bit of rub andyou might hear it more clearly.
“Of course,” he notes, “all this is specific toeach mix and artist. When I work with U2 they don't have these tuningor timing issues. They play as a band and don't need to be loosened up.They don't cut things up in Pro Tools and they don't overtune. With U2,maybe I want the delays sitting just right in the pocket becausethere's already enough movement within their instruments.”
THE CASE FOR DYNAMICS
“The current trend for overcompression doesn't do it forme,” says Palmer. “It makes things sound very, very loud,but also very undynamic and small. Lately, even the mastering engineersseem to be bored with the cheap thrill of overcompression! If you'remastering specifically for the radio, you obviously have to make sureyour songs aren't going to dip compared to your competition. But acommercial CD for home use should have some depth. Lately every CD isprepared as though it's on the radio already. I want my CDs to soundmore dynamic and have more variation in sound and level.”
“Usually by the second verse, if you've got a drum machine atpretty constant volume or parts that aren't dynamic, you're notnoticing them anymore,” says Pensado. “But if you just kindof yank a drum up here, or something up there, your ear finds it andyou'll remember it for another 32 bars or so. Something I've learnedfrom the visual analogy is that it's okay to make things loud for oneor two bars, then tuck them back where they should be. It's okay totake the kick drum and at the beginning of every eight bars turn it up8 dB. Make it stupid loud. The engineers might say, ‘Ooh, did youhear that?’, but 99.9 percent of the people who buy the recordare going to go, ‘That's cool!’”
“It takes me about an hour, but sometimes I'll put every kickand snare at the level I feel it would have been if it was played live[see Fig. 3]. And almost invariably, you feel like the drumsshould come down a bit at the second verse. And then there's the kindof bridge that should subtract energy from a song, so that when youhear that last chorus come in on the radio, you're going to drivestraight to Tower to buy the record.”
THE FINAL FRONTIER
Should you find one set of monitors you like and stick with it, orlisten on every different speaker you can find? Sanity is elusive inthe domain of psychoacoustics, especially when you've been at the mixfor a while.
Q&A: DEVELOPING YOUR EARS
How did you develop the abilityto know what a good mix should sound like?
Palmer: You're not likely tohave it right out of the box. It's something that you develop overtime, something that you definitely get better at. When I first startedout as an assistant, I'd see people spending five or six hoursequalizing bass drums and think, “How can they tell anythingafter all that?” But the longer I've done it, the easier it'sbecome. You have a lot more clarity because you know where you've madeall your previous mistakes.
Nichols: I don't call myselfa musician, but I'm a guitar owner, and a keyboard owner, and I canmuddle through. That ability has helped me to know what I liked, andwhy. Knowing what to do came partly from listening to the balance ofinstruments and the timbre of their sounds. The other part of it is afeel thing, as basic as when you hear a song and something in you goes“Wow, that's cool!” When I'm mixing, I'm always trying tocreate — besides a sound — something emotional. The way thebass sustains under a piano lick, or the way some other parts happentogether… that's what gives the song its mood. To me, that's atleast as important as getting an individual guitar to soundgood.
Pensado: If you're a goodmixer, you mostly have the canvas in your head when you start. It'sjust a question of visualizing the sound as you work. That sound comesfrom the records you heard as a child, the experiences in your life,all sorts of things that give you references. Most of us who do thisrely heavily on those references. That's why it's so important to beopen to new music, new things, and to force yourself to change yourreferences. It's incredibly easy to have them stay solidified, andthat's not good.
“I switch between several different speakers for differentpurposes,” says Palmer. “If you're trying to get a tightsound with the bass and the bass drum knitting together, you might putit on the big speakers where you'll get the best bass response, thencrank it up a bit to make sure it's solid. When you're balancing vocallevels and you're thinking about what the emotions of the music are andif they speak well, you're probably going to be listening more quietly.At that point you don't need to be blown away by fidelity, you're justlistening for a balance you think is correct. I'll usually do that on aYamaha NS10 or an Auratone — whatever I have that's fairly quiet.For fidelity I'd go more for my Genelecs or the big speakers. And ofcourse my car is a good point of reference for me because I listen to alot of music there.”
“If everybody in the world had the same speakers and poweramps I could do a mix in five minutes,” laughs Pensado. “Tobe really good at understanding what happens to your mix at differentvolumes and on different speakers, you need to listen in a variety ofenvironments. I use four sets. I love the old Yamaha NS10s, I've got aset of Augsbergers that are dual 15 TADs with a TAD driver, I've got aset of old Auratones that are on their last legs, and I use the littlemono 3-inch speaker that comes with the ½-inch Studer machine.Even after all that, I'll try to listen in the car, or go to the loungeand listen on a boom box. I also have a little college-radio-stationtransmitter I've put together so I can send FM out to the radio in mycar from the control room — under FCC limits, I want to pointout. I also really like it when we have the budget for clients to takethe mix to their home studio to check it out.”
Okay, lots of different speakers. But what exactly are you listeningfor? “A lot of the music I do requires the kick drum to be asimportant as the guitars are in a Led Zeppelin song,” saysPensado. “If you're listening on the big speakers, you can bethinking ‘Man, it's Grammy time!’ then you go down to theAuratones and there's nothing there. That tells you the frequenciesbelow 100 cycles are right where they need to be, but the frequenciesabove there are pretty much nonexistent. This gets into the area ofpsychoacoustics. You're not ever going to get the low end from the kickdrum to come out of the set of Auratones. But what you can get are thefrequencies that make you think you're hearing the lower frequencies.That could be anything from 200 to 3K.
“A lot of times you'll add top end to a kick just so your earcan find it. You're not making it louder or fatter, but the perceptionis that the kick drum got louder. Actually, it just gave your ear theopportunity to find it. Think of it this way: if you have a power ampon your system, it takes several hundred watts to move the low end, butonly 20 or 30 watts to move the high end. Your ear hears highfrequencies much more efficiently than low frequencies; what you'redoing is playing on the ear's efficiency to find the low end.
“Here's an example. Put on the mix and listen to the level ofthe vocals relative to the music. Pick a song that has a lot offull-range frequency, not one that's thin on the bottom, and when yougo to the smaller speakers it's going to sound like the vocals are tooloud. When you go to the big speakers they'll sound like they're notloud enough. What's right is probably to have your vocals a little loudon the smaller system and a little — just a little — moredifficult to hear on the bigger system.
“As an engineer,” continues Pensado, “if I makethe decision that most of the people who will buy this song are goingto be blasting this as loud as they can in a car, I won't pan thingsquite as wide. And I'll use less reverb, because they're going to begetting the reverb from their environment. It's like when you go to abig dance club where the music is reverberating off the walls; if youhave too much reverb on your mix you're in serious trouble. That's whywe do dance mixes pretty dry, except for maybe an effect on somethingyou want to sound big and wide and nebulous.”
Mix engineers face more than just technical challenges; they alsohave to learn to deal successfully with their clients. “Sometimesa band wants something you think is crazy, just nuts,” saysPalmer. “Then you try it, and it's pretty good! You can't letyour ego get in the way. On the other hand, I recently met with a majorartist who wanted to do things in a way I didn't think was right forthe project. I thought about compromising and doing things his way. Butultimately I realized I would be very unhappy. I told him I wasn't theright person and left. If you don't enjoy your job, you're not going todo it very well.”
“One of the neat things about success,” says Pensado,“is that the people who are telling you what to do have probablybeen successful. That helps with trust, and that bond you definitelyneed to have between an engineer and a client. But I believe conflictis always a necessary ingredient in creativity. [Laughs.] Showme a totally happy environment and I'll show you some crap coming outof it! It's not coincidence that Lennon and McCartney or Jagger andRichards weren't speaking half the time. Or that Van Gogh cut off hisear!
“I like it when a client disagrees with me and can back itup,” continues Pensado. It makes you do things and think aboutthings differently than you have in the past. Most people you work withhave great suggestions; sometimes they don't know how to articulatethem. I'm thinking Fahrenheit and they're describing in centigrade. IfI can get the formula that can translate that, we're doing some greatstuff.
“I think the ideal balance of working with a client is to givethem what they want, but 15 or 20 percent more. With some clients,maybe only 10 percent. Then they're going to feel, and rightly so, thatwhen they're coming to you, they're getting new stuff. Not just whatthey want, but also a sound that's special and really works. It's afine balance. You've got to understand your client psychologically toknow just how far ahead of the curve you want to take them. Rarely isit more than 20 per cent. Give them more than that, and they'll leavestudio thinking you're kind of cool and hip, but a week later you'llprobably recall it and get all that stuff out of there. Eventuallythey're going to take you back in the direction of the rough mixbecause that's their comfort zone.”
KNOW WHEN TO STOP
“I generally work from my instincts and have a feel for whenthe mix is right,” says Palmer. “Some mixers will work for12 hours, then pull the faders down and start again. I've never beenlike that, I would rather work fast and then have another go later if Iam not happy. Also, I find it helps to keep the hours sensible, and toattempt to keep some distance. Taking breaks instead of spending hourson end in the control room helps to keep perspective. Listening througha door or in your car is useful. I sometimes take a break and put theTV on, check out some MTV. [Laughs.] There's always the chanceyou might hear something you can use. But seriously, keepingobjectivity is half the battle.
“I take a mix up to a point where I think it sounds prettygood, but I don't push it too far,” Palmer continues. “ThenI go home — hopefully not too late at night! In the morning, thefirst thing I do is listen to the mix and make my notes with a freshbrain and fresh ears. Morning is probably when I'm the most critical.I'll get a list — whether it be 2 or 20 things to do — andgo straight to the studio and do those changes right off. Then I'mpretty much ready for the band to arrive.
“I don't want to be part of this factory-line sensibility thatsome modern-day mixers employ: ‘Get it done; next!’Occasionally, though, it can turn out really well to have to work fast.When you don't get too much time to overthink, you're working purelyoff your instincts, which can be good. On the Tin Machine records— that people either love or despise — I would start to mixand David Bowie would often say, ‘I love the rough mix, don'tspend too long and take the edge away.’ I'd tell him there were afew things I could get better, so he'd give me one hour to mix. Thatwas it ! You really had to work from your gut. That's all well andgood, but when you're unhappy, and the budget doesn't allow for asecond look it's very frustrating!”
“I like systems with total reset, so that when you open it upit's exactly where you left off,” says Nichols. “Especiallywhen you're working on a whole album. When you don't want to hear thefirst tune anymore you can save it and go to something fresh. You canget it 50 percent done, then go through the whole ten tunes and do thesame, then go back. That saves you from getting tired. You don't wantto spend four hours riding the vocal, then you ride something else andnow the vocals are all too low.
“You bring the vocal up more, and some of the rides you did nolonger mean anything. That's chasing your tail. A break and going toanother song can really help. When are you done? When there's nothingleft to do! If you can listen to the whole record, and not just focuson one thing, if nothing bothers you, and you can hear everything, thenit's okay. That's it.”
On the other hand, Pensado says, “I have a saying that you'renever through with a mix, you just run out of time. I'm not sure thatyou're ever really done. I don't get to do a lot of recall mixes, butunlike a lot of engineers, I actually like doing them. Even when you'venailed a mix, when you hear it again you'll have other ideas. That'sjust a sign that you're growing. A mix is never really finished, but itis definitely possible to overmix something. There's a point in theprocess where another two hours can very easily make it worse. You haveto look for those signs. You ask yourself, ‘Was I more excitedabout this an hour or two ago?’ The minute you start feeling it'sgetting worse instead of better, you were done an hour ago and youshould probably backtrack!”
Finding your own voice is one of the most important and rewardingthings you can do, in life and in mixing. But in our increasinglyhomogenized world, it can be a risky business that requires hard workand a high level of self-awareness. On the other hand, great art, andlasting records, don't come without pushing the envelope.
“I think these days a lot of people are affected by marketingwhen it comes to how they're working,” says Nichols. “It'sabout what equipment they're using, and what somebody else who'ssuccessful did with that equipment. They're not really paying attentionto the specifics of what they themselves are recording or mixing. Theythink they have to use a certain drum machine, or a certain microphone,or a certain compressor that's set in a certain way because that's whatthey've heard they're supposed to do. To me, what's important is toexperiment, some little bit, every single time so that you can come upwith your own rules.”
“If people in our industry thought more about the correlationbetween the visual elements of mixing and the audio part, we could takemixing to another level,” says Pensado. “We have to step upthe music to where a kid wants to buy our record as opposed to a pieceof software or a PlayStation game or any of the other temptations thatare out there for 20 or 40 bucks. Obviously we all would like to havebetter music, but those of us who do what I do can contribute a littlebit by making the music we have better. That doesn't require replayingthe parts, it requires understanding what the original vision of theproducer or the writer or the artist was and trying to piggyback yourvision onto that rather than having a contradictory vision, ornone.”
Palmer states simply: “The song is, and will always be, king.It's funny that as we increase sampling rates and bit rates in therecording side of music, the public is moving the other way anddowngrading from CD to MP3. They are showing us they really care aboutthe songs, artists, and performances. That is not an excuse for poorproduction and mixing, but a reminder about what makes someone want toown a piece of music. Sometimes I feel we miss the point. Don't forgetthat the best cure for a bad mix is a great song!”
PENSADOPICKS PLUG-INS OVER HARDWARE
Dave Pensado hashad plenty of experience working with all types of processors,including many classic analog units. But surprisingly, if you gave hima choice between using the original hardware or the softwareequivalents, he would choose the latter in most cases. Why? “Whatthese plug-ins do is give you the opportunity to take a classic soundand tailor it for today,” he says. “The concept is to walka very fine line between giving listeners something that reminds themof something classic, but that also sounds new. So if you're into retroeverything, just use the originals, you'll be fine. But if you wantthat plus more, the plug-ins give it to you. It's just a more modernsound.”
One of theplug-ins that Pensado likes is the McDSP Compressor Bank CB2, a TDMplug-in that he uses to emulate the sound of a classic Fairchildcompressor. Among the artists whose vocal tracks he's used this plug-inon are Brian McKnight and Beyoncé Knowles. “What I likeabout it over the analog Fairchild is that I have more flexibility overthe controls. I can control attack and release, I can control theamount of compression [see Fig. A]. With the real Fairchild, Ican't get in the cracks enough to do what I want with Brian's andBeyoncé's vocals. Some of the older effects are almost aone-size-fits-all type of approach. But the CB2 plug-in allows me totailor it for each song and each vocal.”
FIG. A: For the vocals on BrianMcKnight's recent CD U Turn (Motown, 2003), Dave Pensado usedthis Fairchild compressor-emulation setting on the McDSP CB2compressor. He used a similar setting for Beyonce Knowles's vocals onthe Destiny's Child album Survivor (Columbia, 2001)
Another examplecomes from Eventide. Pensado prefers the plug-in version of thatcompany's Harmonizer H910 (one of the components in Eventide'sClockworks Legacy Bundle for TDM) to the original hardware unit.“The Harmonizer plug-in is better because it's stable,”Pensado says. “A lot of the old first-generation orsecond-generation digital equipment had stability problems. ThisHarmonizer plug-in gives you that classic guitar sound, that classicinstrument sound, without the stability problems. It's just anall-around improved, but still sonically identical, version of theoriginal.”
He says that whenhe's done blind listening tests to compare the plug-ins he uses and thehardware processors they emulate, the plug-ins usually win. “Whenyou don't know which is which, you'll probably pick the plug-in 70percent of the time,” he says, “if you're being honest withyourself.” (For more of Pensado's observations about plug-ins,and to see more of the settings he uses, refer to “Dave Pensado'sPlug-In Secrets” at www.emusician.com. You can also contact Pensadodirectly by e-mail at email@example.com.)
Maureen Droney,whose engineering credits include projectsfor Carlos Santana, George Benson, John Hiatt, Whitney Houston, andAretha Franklin, among many others, is the Los Angeles editor forMix.