A tasty recipe for mixes you’ll be proud to share with the world
By Marc Urselli
Mixing is just like cooking. That’s right, I said it.
First you need great ingredients, ideally organic. Next you need good tools with which to actually do the cooking. You can definitely use some creativity, good taste (doh!) and a palate for experimenting and trying out new things. But most importantly you need the knowledge and experience that can only come with lots of cooking—whether for discerning crowds at restaurants, entertaining guests with a kickass meal at parties, or at the very least, cooking for yourself and your family at home.
Now re-read the sentences above and replace the words “ingredients” with “tracks”, “cooking” with “mixing”, “meal” with “mix”… Enough said! There you have it!
I am pretty sure most cooks would agree with me on all of the above statements. Even though I am half Italian and I know a little something about cooking, I am definitely a much better mixer than I am a cook, so let’s talk about how to cook up a good mix.
The kind of mouthwatering creation we want to make is the kind your listeners will remember, talk about and crave more of… Obviously you need the main ingredient (good songs!), but you also need the right condiments, the salt, the spices, and all that good stuff that elevates a dish from plain-tasting to incredibly delicious and full of flavor, rich in subtleties and aftertaste. Hopefully the artist has the main ingredient covered, which leaves you with the responsibility to make the songs sing, whether there are vocals or not!
For the sake of this cooking class I will assume that we are dealing with a standard rock band with drums, bass, guitars and vocals, but rest assured that a lot of what you’ll learn here can also be applied, judiciously and tastefully, to other genres as well.
The basics: before we start
A good mix always starts with a good record-ing. For me, the easiest way to assure this is to do the tracking myself! If the material for an upcoming project hasn’t been recorded yet, I will always ask the artist or producer about using EastSide Sound (eastsidesound.com), the New York studio where I work. I am a fan of room miking for “air” in recordings, and also of hear-ing things clearly and with presence, so I make use of proximity, separation, iso booths, and all of the great tools that EastSide puts at my disposal. I am not a fan of “fixing it in the mix” or of skimping on the record-ing process. If you want a record to sound great, you need to record it right!
There are, of course, times when the record has been tracked somewhere else, by somebody else, out-side of my control. I do a lot of mixing for people that I never even meet in person or who live in remote parts of the world; it’s more common than ever for me to receive tracks for mixing that I did not record myself.
I am overcome with feelings of anxiety every time I double-click on a Pro Tools file somebody sends me. More often than not, there are missing audio files, dozens of tracks called “Audio 01”, “Audio 02”, “Audio 03” and so forth… The lack of organization and good housekeeping on the part of the engineer is the first red flag, even before I press the space bar and listen to how it sounds!
Then there is the issue of signal chain and gain structure. This seems like such a basic subject that one should understand completely before attempting any type of recording. Yet I still receive clipped waveforms and/or teeny-weeny waveforms with an amplitude of 10% of the full available bandwidth, or even things that sound saturated even though they don’t look like it.
Then there are more complex issues, such as badly managed leakage or total lack of proximity. Leakage is fine; it exists, it has a place in recordings, and it can be used creatively. But you need to know what you are doing, or you will put the mixing engineer in a situation where it is impossible to work and there is so much leakage on everything that achieving clarity is almost impossible—I remember something like that happening on an Ornette Coleman recording that I was once hired to mix.
Finally, there is just plain poor audio quality. In this day and age of highly accessible and affordable technology, bad-sounding tracks should be a thing of the past, but you can still find yourself hearing tracks with a muffled sound, a dark sound (even on bright things like vocals or strings), a faraway sound, a sound that has a veil over it, or a sound that sounds as if it was black and white when instead it should be full color HD. Often this is due to subpar preamps and microphones, equipment with a high noise floor, a low bit rate and sampling frequency, or bad A/D converters. Sometimes—often!—it’s simply operator error… poor or lacking skills in basic audio engineering are usually the underlying and overarching problems of a lot of recordings I receive for mixing.
But enough complaining about these recordings already! This article is primarily about mixing, so let’s hope and assume we have good-sounding tracks to mix and see what can be done to make them shine!
Let’s get the kitchen ready: routing and housekeeping
The first thing I do is to listen to a rough mix of the song, or just open the tracks I am sent, to listen to what’s going on with the song. You will be surprised by how many people will start working on the kick drum with-out even having heard the song once!
After I get a feel for the song, I set all the outputs of the DAW tracks to “no output” so that nothing plays when I hit the space bar. Unless the artist and I (or the recording engineer and I) have had a conversation in which it was pointed out to me that they are particularly enamored with a certain plug-in, volume ride, or effect automation that they want me to pre-serve, the next thing I would do is remove any plug-ins there might be, delete any volume automation there might be, delete any other automation (mutes, sends etc.) and basically create a blank slate for myself to work from… basically as if the DAW was just a good old tape machine, for those of you who remember such archaic objects.
At this point I look at the names of the tracks (which have hopefully been named correctly, or named at all!) and I drag and drop and reorder the DAW tracks in the order I want to see them on the screen. If I am mix-ing on an analog console as I do at EastSide Sound, the order on the screen is usually the same order as that of the faders on the console. If I am mixing “in the box” (DAW), as I do in my home studio, I usually arrange the tracks in groups: drums first, bass next, etc.—but that is really just a question of preference.
Bringing it to a boil: the drums
When everything is laid out for me to start listening and dialing in sounds, I start with the kick drum. It depends on the nature of the material, but usually the kick drum ends up needing to be a bit louder in a full mix than where it will sound and feel right when you are just listening to the drums alone. You’ll always want to go back and re-assess that, once you add bass and guitars. If there are two kick drum signals—inside and outside mics, or a kick mic plus an external Yamaha SubKick (a reverse-wired speaker which acts as a mic that picks up very low frequencies), for example—I work on blending the two sounds together so that the kick drum is both punchy and full, assuming that is what the material calls for.
The other very important thing I do is check the phase. Phase is the little known but supremely powerful dark monster that hides under every engineer’s bed, waiting to jump out and make their tracks sound thin, disembodied, or even disappear completely. I am not going to attempt to teach you what phase is—I happen to know that in the next issue or two, our Paul Stamler has a huge article on the subject coming up!—but I will tell you that I check for phase-coherence between the two kick drum signals and between snare top and snare bottom. Sometimes I have even time-aligned the drums to the overheads to ensure perfect phase coherence, even though in general I find that with a good-sounding and well-tuned drum set, that level of time alignment is usually not necessary to achieve a good drum sound. Again, it all starts with the good ingredients!
Once you have a good and healthy kick/snare/hi-hat relationship, it’s time to add the rest of the drums. Depending on how roomy a sound you want for the drums, it can be useful to dial in overheads right away. Some people dial them in at the end and aggressively highpass the overheads, using those signals only for the high end sizzle of the cymbals. However, one should never underestimate the added body and punch that overheads can lend to the entire kit, so more often than not I won’t aggressively filter the overheads. Instead, I’ll dial them in before I add the toms to the drum mix, so that the toms benefit from the attack and air that comes from the overheads. For the record (no pun intended!), if the material I am mix-ing is more jazz than rock, I usually rely more heavily on the overheads for the entire drum sound and only add the other mics in around the overheads.
If you have recorded room mics for your drums, it is now the time to work with those and decide how much back slap from the room you want to hear. Gentle EQing and ballsy compression with high ratios and fast release times can help you achieve a larger-than-life room size. You can trick the listener into thinking that the drums were recorded in a much big-ger space than they actually were… assuming you had positioned your room mics correctly to allow you to do that in the first place (good ingredients, remember?).
If I am mixing in the analog domain, for the room mics I sometimes use a pair of linked Empirical Labs Distressors. If I am mixing in the box, I will either use a McDSP Channel G compressor (one of my favorite plug-ins) or an 1176 compressor plug-in, usually the Pro Tools stock one from Bomb Factory. If you are using an 1176, whether a real one or a plug-in, it’s important to know that the attack and the release knobs get faster as you turn the knobs clockwise. Most compressors out there are the opposite (clockwise is slower and counter-clockwise is faster); a lot of people don’t know that about the 1176 and therefore miss out on some great sonic possibilities. Another great thing about the 1176 on room mics is its “British Mode” trick (also referred to as “all-buttons mode”), which is when you press down all four ratio buttons at the same time, giving you a ratio of somewhere between 12:1 and 20:1. Many compressors and plug-ins based on the 1176 design offer this mode, and it’s quite useful.
In the end, it’s all up to your taste to decide how much room sound you want to hear. In my experience, room mics work best with slow to midtempo material so that you have time to hear that room reverberation, but when the music is fast they could very well muddy up your drum mix. So use them with caution!
Let’s add the beef stock: the bass
Any vegetarians in the house? I hope not, because this is where we add the basic flavor.
The next step would be to bring in the bass. Needless to say the bass is the beef of your dish! It is not a coincidence that the adjective “beefy” is often used to describe full, bassy, powerful sounds. The bass is your foundation, and you must have a solid foundation to build anything.
I like to compress my bass—usually I do between 4:1 and 6:1 on a McDSP Channel G plug-in or a real Urei 1176 unit—and I always check for phase coherence between my Bass DI and my Bass amp signals before blending the two to my liking. When I record an upright bass instead, I usually forgo the DI completely and add a second microphone closer to the bridge, to pick up the string snap and the sound of the upright bass at the fingerboard. I then mix that sound in with the fuller sound of the upright as captured near its F-hole until I get the balance just right. I check my phase here, too!
Although many bass players are sometimes outright opposed to using a DI or don’t care for the sound of the DI alone (and in the case of upright bass players I agree), I find that the right amount of DI signal mixed in with the amp signal of an electric bass can really add a lot of clarity and punch to the sound of the instrument. If you have a bass player that plays a lot of notes, having that brittle DI signal in the blend can really clarify what notes are being played.
This reminds me of a mixing session with my late and beloved friend Lou Reed, who once walked in just as I was working on the bass sound and I had the bass soloed. Lou got furious because he hated the bass sound I was getting, and he asked me to change it. Of course I didn’t argue; I sim-ply muted the DI signal and played him the amp signal alone, which was miked with a beyerdynamic M88. This immediately pacified him. Then I un-soloed the bass and we listened to the entire mix, and the bass was kind of lost—you could feel its presence but you couldn’t really “hear” its notes.
Lou, the ever attentive and sonically aware listener that he was, noticed it and asked me to raise the bass a bit. I suggested that instead of raising the volume I would try something else. “Trying some-thing else” meant that I tuned the DI signal back on (although I didn’t tell him until later that that was what I did!). This immediately added that missing intelligibility to the bass signal without physically having to raise the bass fader on the console. That’s when I got that priceless Lou Reed grin that we all came to love and we always strove to achieve!
Bringing it to a simmer: volume control
At this point we need to bring the flame down and check the temperature. And by that, I mean it’s time to turn down the volume!
I find a louder volume really helps when you are working on your kick drum and bass relationship, because the two have to lock and it’s nice and helpful to “feel” that relationship when you are driving your speakers a bit harder. But once you get that down, it’s time to give your ears a rest and to embrace the Fletcher-Munson teachings: the human ear hears things differently at different sound pressure levels. Frequencies are perceived differently at different volumes and, believe it or not, our ear also acts as a natural compressor.
On top of that, you need to take into account ear fatigue and the longevity and health of your hearing, if you are planning to do this for years to come (as I am). I take frequent short breaks when mixing, so that I don’t fatigue my ears to the point where fatigue clouds my judgment. And believe me, if you mix for 10, 12, 14 or more hours as I sometimes do, this is really important.
What this also means is that when you are mixing you can readjust your focus to different volume settings. Things that don’t seem too loud when you are listening loud will suddenly jump out in the mix when you are listening softly. This is especially helpful when mixing vocals: I set my vocal levels at softer volume and then check them again at medium and loud volumes.
One of my final listening passes before printing the mix is at super soft levels, so soft that anyone talking in the room would be highly disruptive to the listening experience. Your clients will usually want to listen to everything loud and proud. Mine always ask me if I can turn it up, or turn on the soffit-mounted big speakers in my studio! You can do that and get them excited, but you owe it to yourself and to your mix to also listen to your work very softly to make sure that you are not being too aggressive with any particular instrument, EQ or automation move.
As I mentioned before, the ear is a natural compression device, which means that if you have some-thing that is too loud in the mix at loud levels, you will not necessarily notice it because the ear is reducing the dynamic range of your entire perception. At lower levels, however, you will hear these things right away!Keep that in mind and don’t get too excited with that stove top knob… ahem, I mean volume knob!
Add the meat and potatoes: vocals and guitar
If we were cooking a stew, it would now be the time to add the meat and the potatoes. Some would argue that drums and bass are the meat and potatoes… I won’t argue with that, they have a point. But for the sake of the cooking analogy I’ve got going, I’m going to assume that my guitars and my vocals are the main ingredients here.
Guitars are relatively easy to mix. Finding the right sonic space for them is the hardest part but, unless your mix is too cluttered with a lot of overdubs, that should really be a piece of cake (nope, it’s not time for dessert just yet!). Hopefully you will have two guitar tracks that you can pan left and right to enhance your stereo imaging; if you have that, half the battle is won. The impact that a carefully written two-part guitar arrangement, or even a single double-tracked guitar part, can have on your song is not to be underestimated!
The vocals always go in the center for me. I might spread background vocals apart a bit, but the lead vocal is in the middle in pretty much every mix I have ever done. Compression is key to make sure that the vocal is always up front and heard… and to stop myself from falling into the trap of obsessively riding the volume automation syllable by syllable—another thing I used to do in the beginning and have taught myself not to, unless strictly necessary.
With vocals, I like my compression to be as transparent as possible. When I track vocals, I either track without compression or I track with some very light compression through a real Urei 1176 or a Teletronix LA-2A. However, when it comes to mixdown, I err on the side of digital compression. The reason is simple: if I want to compress a vocal more aggressively, the sonic properties of a vintage analog limiter or compressors often become less desirable to me than they are on, say, drums or bass or guitars.
My personal preference for vocal compression is in the digital domain and specifically my go-to compression plugin is a McDSP Channel G, because its compression is less audible. Most of the times (not always!) I want to be able to hear the vocals without hearing the compression, and I feel that with that particular plug-in I can do just that—along with a lot more stuff!
Beside compression, reverb and EQ will help you seat vocals in the mix so that they fit. You don’t want them to sound like they are in a completely different space, because that might create a disconnect. You don’t want them to be buried, but you also don’t want them to overpower everything else. I usually highpass-filter the vocals: sometimes more, sometimes less, depending on the singer, the formant structure of the voice, and its other tonal qualities. If the singer has too many plosives, I might filter a bit more aggressively, or just automate a highpass filter to only intervene on the plosives. If a singer has too many sibilants, I will employ the services of a de-esser.
I usually do most of my EQ on the console if I am mixing analog, or with a McDSP Channel G plug-in if I am in the box. As far as reverbs go, my go-to analog reverb is an old vintage spring reverb tank unit called a Master Room II. I also often use the Lexicon 480. When mixing digitally I use the McDSP Revolver convolution reverb, which is very powerful and has great presets, and the IK Multimedia CSR Classik Studio Reverb, which offers Plate, Hall, Room and Inverse. I could write a whole other article on mixing vocals, but I will keep it short and sweet here, since we are talk-ing about the “whole” mix!
The last few spices: effects
Sprinkling the magic seasoning before you plate your creation and adding the finishing touches is obviously very important, but should not distract you from the things that matter most. There are two quite vulgar figures of speech that, despite temptation, I’ve successfully avoided this far in this article to keep my Editor happy… but it ain’t my fault if they are true and fit the context perfectly, so here I go, censoring myself with a few asterisks. One is: “you can’t polish a t*rd” and the other one is: “s**t in, s**t out.”
Make sense? What these bucolic pearls of wisdom are trying to teach us is that it really does matter where you start from. No amount of reverb, delay, flanger, or other effect you might have up your sleeve is going to fix a bad track or mix. What this means is that you should feel free to creatively use whatever effects you want to use to enrich your mix, but you shouldn’t be forced to use such effect to make up for the lack of something else. In other words, if you have messed up your meal because the salt dispenser fell out of your hand and now you have half a pound of salt in your dish, no amount of sugar or water will ever fix that. Throw it away and start over!
That is another reason why I don’t sub-scribe to the “fixing it in the mix” philosophy. Get it right from the start; that way you don’t have to try to hide things later. In fact, you should never try to hide things!Mixing is about exposing things, making them more clear, more audible. It should never be about obfuscating and conceal-ing. If you are in the lucky position of hav-ing great recorded tracks played by great musicians, you can deploy your reverb strategy and add any other crazy creative effect ideas you might have.
When I say “reverb strategy,” what I am referring to is this: some people argue, and they have a point, that if you are trying to create a realistic representation of what the band would have sound-ed like if they played in front of you, you need to choose a reverb that fits the con-text and stick to that reverb throughout your mix. I have used this method on some mixes and I like it in principle.
However, more often than not, I use my own personal approach: adding reverb only when I want to hear reverb and to what I want to hear with reverb, even if it isn’t spatially and physically realistic. After all it’s not a science project, it’s art!
If I want to hear a room reverb on the snare, a plate on the vocal, and a hall on something else, I will do just that… Or I might even use the same type of reverb for everything but with different amounts and with different returns. Sure, I could just send different amounts of audio to the same reverb unit, but often I use different reverb units (or even different instances of the same reverb plug-in) and I dedicate one hardware unit or plug-in to one specific instrument. This also turns out to be helpful when you do stems at the end of your mix; you can actually submix, say, all the drums plus the reverb of the drums into your drums stem without also hearing the return of the vocal reverb send into the drums, or vice versa.
Everything but the kitchen sink?
Have you ever wondered why food is so much better in Europe than it is in the US?Have you ever eaten a tomato in Italy? Have you ever tasted yogurt in Switzerland?Have you cooked with butter in France? Well, here’s the secret: Less is more! Without side-tracking too much, I will tell you that at the heart of the European cuisine that everyone tries to replicate Stateside lies true simplicity and goodness of ingredients. In the US all of our food is pumped with preservatives and the ever-present corn syrup.
How does that relate to mixing, you ask? Simple: don’t ruin your mix by adding things that you don’t need! That goes for everything: EQ, plug-ins, etc.
When I was an apprentice engineer and I was cutting my teeth on local punk bands in Europe, I used to sit there and EQ the crap out of every single track I recorded. Now my approach is the total opposite. The less, the better! I EQ things only when strictly necessary, and even then, I try to cut rather than boost. Instead of boosting all the frequencies of a certain instrument I try to carve out space for that instrument by EQing other instruments a little bit differently. For example, if you have loud distorted guitars that sprawl in your upper midrange, you might want to try not boosting those same upper mids in the bass tracks, and so on.
What is important to remember is that you can give any one particular instrument more presence and space by just knowledgeably EQing and panning it, rather than by making it louder. Some rock bands just want “everything louder than everything else” (as Motörhead famously said), but trust me, this approach will not work when you are mixing a song. Instead, try the “less is more” approach and listen to your mix come to life through simplicity!
Only add plug-ins where needed. Don’t EQ everything. Don’t add reverb to everything. This approach would serve well in songwriting as well, but we are not going there in this article. The bottom line is: choose with parsimony, think as if you didn’t have unlimited tracks, unlimited resources and unlimited CPU power. An overwhelming array of choices should not be a reason to use something that you wouldn’t otherwise use, unless you are going for a creative approach!
One final word of caution about your levels to tape (or DAW of choice): it is the job of a mastering engineer, not that of a mixing engineer, to make your mix com-pressed and loud. Mixing engineers don’t fight in the Loudness Wars! A little bit of compression on the mix bus is fine, but make sure you retain a good healthy dynamic range throughout the mix so that your music can breathe!
One trick I can recommend is this: if you are mixing in the box, instead of bouncing to disk, route the output of your mix bus back into another stereo track in your DAW and monitor the output of that stereo track instead. This will not only ensure that you are actually listening to your final mix (rather than assuming that the digital bounce to disk is successful and error-free), but it will also give you a visual feed-back on the amplitude of the final mix waveform that you are generating.
You want to see a waveform that doesn’t clip past 0 dB and that has valleys and peaks. You don’t want to see a thick straight horizontal ragged line. If you did that, you might end up with a louder mix, but you will also end up with a mix that has the life sucked out of it… a mix that a mastering engineer cannot work with, because you have left him no headroom to work!
Can you feed yourself with some off-the-shelf corn-syrup-injected pre-cooked pre-packaged factory shrink-wrapped foods, and not starve to death? Sure you can! But really, what’s the point of eating like that?
Wouldn’t you rather eat at a 5-star restaurant where an award-winning chef cooks a meal specifically with you in mind—your dietary restrictions, your personal taste—using the best ingredients and best tools money can buy, to give you a truly unique experience that you will remember forever?
When mixing your music, make sure that it can stand the test of time! You need to be proud of your mix and you need to know that you will be proud of it down the line. If you listen to your mix and it doesn’t sound as good as it does in your head, or it doesn’t sound as good as the music of your favorite artists does, something is missing. When you have to, go back to the drawing board. Continue working on your mixing skills until you can make a mix that you are 1000% proud of!
(SIDE BAR) When should you call a chef?
If we could afford it, most of us would probably have a chef on duty cooking every meal for us, day in and day out, but if you were able to live like that you wouldn’t be reading this article. So let’s talk about the rest of us, who would only afford themselves the services of a chef when it’s a really important event. You don’t want to mess up your important dinner just because you wanted to impress your friends by telling them you did everything yourself, do you? That could work if everything turns out well, but if you are not a professional chef, why risk it? If it’s that important, don’t skimp on the quality!
A good cook with bad ingredients will always do a better job than an inexperienced cook with great ingredients. That is why you should always put yourself in the hands of an experienced professional mix-ing engineer when the job at hand is vitally important. If you are a hobby musician and mixing is a hobby, that’s all fine and well, but if music is your life and you take it seriously, why would you cut corners when it comes to the final presentation of your cherished and carefully crafted art?
I will never understand musicians who spend countless hours on writing scores, arranging songs, perfecting their craft… and then, when it comes to recording or mixing their music, they choose the cheapest option they can find and end up with a product that’s only half as good as it could be. Very sad, if you ask me!
Sure, I understand that money’s tight, but just as I wouldn’t go out and buy a trumpet and expect to be a trumpet player, you shouldn’t go out and buy an audio inter-face and expect to be an engineer. Just as it took you years to get to where you are on your instrument, it takes an engineer years to get to where he or she is with his/her engineering skills. Both are serious skill sets that can only be acquired and mastered by spending a lot of time doing the work.
This stuff can be learned. I wrote this article to help you learn. But while you’re learning, remember that there’s no shame in enlisting the aid of someone who’s better than you are when the chips are down… and this article will hopefully have pointed out ways for you to make sure that the tracks you deliver to the mixing engineer will be their very best.—MU