What is Mixing?
What is mixing? A dictionary definition (if there is one) would be something along the lines of ‘To balance all the sounds of a multitrack recording to give a coherent whole’. That’s pretty accurate but also pretty clinical. For me, it is also incidental. When you’re talking about a finished piece of music, or song, what is a coherent whole? What is a finished mix and how is this determined?
The art of mixing audio is really the art of compromise and although we really only have four tools at our disposal, those four tools mixed with our sense of taste and our skill level can give a single piece of music infinite possibilities when mixing. To become an accomplished mix engineer takes time, effort and commitment. It’s not easy or quick and touches on many surrounding factors such as monitors and acoustic environments. It is a science and an art and no accomplished mix engineer really proceeds without knowledge of both the art and the science. I’m not saying you need a PhD in Physics to be a mix engineer, you just need to have a grasp of both. Do you really need to know what a half normalled balanced patch bay is? Not necessarily, but you should. One day you may find yourself in a studio where the client is running everything to outboard and you want to gate a hi-hat with a string sound whilst leaving the hi-hat intact in the track. There’s your science right there. Your art will be how you apply it and what you do to it. Learn both, don’t be lazy.
The Four Tools of the Apocalypse
So what are these four tools? Surely there are way more? For me, and some may rightly argue otherwise, there are but four basic tools and they are:
- Level (with our fader)
- E.Q.
- Pan
- Effects (reverb etc)
Effects, granted, is a huge category and some may say that ‘effects’ is too broad a term and could quite legitimately be split up into a few categories such as time based (reverb, delay etc), dynamics (compressors, limiters, gates etc), distortion, modulation (phaser, flanger, chorus etc) and pitching effects. I am covering mixing as an overview and as such, I don’t feel the need to go into depth on any of our four tools despite how huge some of them may be.
The tools themselves are pretty self explanatory. Level is the volume of that particular sound on that particular channel whilst E.Q. will allow us to tailor the frequencies of that sound. Pan will give us a left to right (for stereo) positioning and our effects such as reverb can be used (among many other things) to give us depth. So we instantly have volume, left right positioning and depth controls to ‘position’ each of the tracks, everything we need to balance and produce a good mix.
When we start to use multiples of our four basic tools our palette really begins to open up. Parallel compression, for example, will use multiple tracks, faders, effects (dynamics) and quite possibly E.Q. Plus, ever put a little filthed up ambience on your paralleled channel? Very quickly we can seemingly have way more than four tools. In reality we haven’t. We have, by application and imagination, used multiples of our four basic tools. Add to this creative routing, VCA groups and automation and your imagination really is the limit.
Did You Say Compromise?
Yes, I did say compromise. If you don’t like compromise you’re not going to like being a mix engineer. No matter how many tracks we put down, compile or multiply there comes a time when you need to make decisions, and decisions will always mean compromise to some extent. This is not a bad thing. It relates back to the artistic side of mixing and also has a good foothold in the scientific. In a lot of ways, being able to have hundreds of tracks in our DAWs has spoiled us. The quicker you make your decisions the smoother your mix will go. The more you make these decisions the better you will get at making them. Basically, stop putting it off.
So what are your key decisions? What is more important, more prominent, aids the song in it’s entirety? How does the energy flow through the song and how do you keep that energy throughout? These are all things that relate to your taste and to your ears. We all hear music differently, and, as a mix engineer; you have the responsibility to bring out the song. To make it the best it can possibly be for the artists vision. So some things you have to compromise. Let’s have an example.
Lower mids are always tricky to get right. Knock too much out and your mix will sound thin. Add too much or leave too much in and your mix will sound muddy. So when is a compromise not a compromise?
Easy, when you’re a mix engineer. Confused?
A couple of years ago I mixed a song that was very heavy in the lower mids. It seemed like everything important for the first three quarters of the song happened down there in those lower mids. It was a tricky mix. When I played the finished mix to the client they really liked it and said ‘How did you get all that low stuff to all come out and sound full?’.
The answer was, I didn’t. Remember, when your audience is listening to your finished mix that is all they are hearing. They have no clue, and neither do they want any clue, of what was on the multitrack. There is no other point of reference than the final mix they hear. I gave the illusion of everything being fully there by compromising. If I pulled up the multi and took all the E.Q.’s etc off it would be a muddy mess, which was how I received the track.
I decided what was more important, what should sit on top and what should be underneath. I moved certain things out of the way by panning and pushed other things back with reverbs or delays. Often it was a combination of all of the above.
The bass part was seriously overlapping the bass drum. The bass drum was deeper, overall, than the bass part so I knocked some low and low mid out of the bass part and added some high end into it. Just a touch. A db or so. Remember, this was still a bass part. There was plenty of bass left in the sound. We have separation, we have compromise. We have a good sounding track. This is how compromise becomes a positive because, as all things should, it served the song, not the part.
I went on to do this with all the instruments that were overlapping until it felt right. When it feels right it usually sounds right. This song had a very dark and ambient mood and I don’t want you to think I separated all those sounds out because I didn’t. That would sound clinical and wrong. There was still plenty of overlap but it was quite surprising how much I could cut and still keep it deep and growling whilst leaving the sounds melting together. Despite all the cuts I made, the illusion was of no cuts at all.
What matters is how it sounds to the audience not what you needed to do to get it there.
That’s the Art, Where’s the Science?
The science begins with your room and monitoring. Spend (nearly) all of your money on these two things. Get it sounding flat. If your listening position and monitoring does not allow you to hear the lower mid you’re just going to crank it up until you can hear it. The result is that all your mixes will have too much lower mid once they leave your room. I’m picking on the poor lower mid, of course, this will apply to any frequency or group of frequencies that your room emphasises or cancels.
Remember, you’re not mixing it to sound wicked on your monitors, you’re mixing it so that it will translate to every monitor and headphone system on the planet. Ok, exaggeration… but that should be your goal. So your monitors can take +8db at 60hz? No ones ear buds will. You choose as you’re the mix engineer. You make the compromise.
If lots of sounds overlap at one frequency it will be hard to make out what is actually going on in that particular area. That IS just physics. Make decisions, compromise, use your taste and fix it.
So, is this Mix Finished?
Someone once asked Frank Zappa when he deemed a mix to be finished. He replied “When I can’t stand it any more.” It’s a good answer. A lot of mixes are finished when the deadline runs out and seeing as we should all be learning all of the time I don’t think a mix is ever really finished. It’s just another decision and another compromise we make. We’ll listen back six months later and tell ourselves what we would do now because we have gained so much more knowledge. That is something you learn to live with as well. Knowing when to walk away from your mix.
A big tip is to take frequent breaks. At least every hour and for 10 minutes. Go outside, get a drink. You’ll be amazed how much quicker your mix will grow.
Get a good, solid static mix before you start tweaking with automation and keep a note pad so that when you get that great idea of automating that final verse word into a reverb that will feed into a compressed delay you can write it down and not get distracted from getting your decent, solid, static mix.
Don’t mix drunk or stoned. It will sound great at the time but the next day it will sound drunk and stoned and you’ll just have to start from the beginning. This will depress you and you’ll get stoned again. To counter it you’ll have a drink or two and the whole sorry mess will circle resulting in you never ever finishing a record.
Work hard, learn your craft, listen to others, ignore others, embrace the art, embrace the science, make decisions, stick to decisions, change decisions but always always always… serve the song.
And finally: Be brave! Try everything, try something wild if you think that it will still serve the song. Innovation favours the brave.
Paul Smith 