1) Gain structure
This is fundamental and both technical and artistic, we will focus on technical reasoning for now. We want to avoid noise and distortion which occupy the extremes of the systems which will use for recording and mixing. Too quiet and we risk hiss, too loud and we risk the dreaded crunch. In analog gear 0Vu was the level to aim for, people are often surprised when I say this equates to approximately a peak level of -10dBFS. (though it depends on transient information and the frequency content of the source)
DAW manufacturers do not help us by making -10dBFS on the metering in their DAW’s look very low indeed. Fact is if you keep your levels around -10dBFS you will be mixing at around the same electrical level as those professionals who mix on a big analog consoles. I suggest thinking very carefully about your gain structure when recording and mixing and a great tip is to always record and mix at 24 bit resolution. At 24 bit there is absolutely no need whatsoever to hit the level meters hard in record or mix stages. It serves no purpose (with the caveat of saturating a specific device for subjective effect) other than to diminish your headroom and create a higher chance of distortion in either the analog or digital domain.
2) Gain structure... again !
So pointer 1 explains why good gain planning is good, now we need to consider how to practically achieve good gain structure. Arguably mixing starts in recording, thats what learned engineers know and teach. So when you record, record at 24 bit and leave headroom. Take level from a practice performance and peak at -12 to -14dBFS to allow for the enthusiasm of a “real take”.
This leaves your signals at just the right level to start mixing with plenty of headroom. Use a ‘peaky’ source in your mix down session like your snare and kick drum and use them as a reference to start your mixing. Let’s say you peak your kick at -12dBFS then add your snare then the rest of your kit, great…. except you are likely to want to process with EQ or compression at some point. This obviously changes level, so bear in mind you do not want to move your reference too far from where you started. So as you eq and compress try and bear in mind your -12dBFS ref and keep the peaky sources in this ball park. You do not have to do this religiously after all the goal is some head room not a bad sounding mix balance based on numbers ! It’s a balance, the right level to leave headroom and the juggling of level and balance that mixing by it’s very nature requires.
3) Throw up the faders
Mixing is an iterative process for many so it is difficult to describe “how to mix” exactly. However you must start somewhere and no better place than introducing the faders with sources on and obtaining a rough balance. A rough balance is important as it allows you to consider what problems exist and what ones can be rectified with your tool set. It starts the brain firing off in the right direction as to what sculpting might occur. Usually this is a process which happens extremely fast. You might get a quick succession of thoughts… like… “kick drum is muddy, snare needs some brightness, cymbals are harsh, a short reverb would be nice on snare from plug in “X”, overheads sound a touch wide, a touch of de-essing required on lead vocal, guitars have excessive hiss on them in the pauses, things like this. If you need note them do so, but often if they are enough of a problem they will stick in mind or be rectified within a couple of minutes. In this way I think software is an advantage as it is very quick to load a plug in and act.
So assuming technical ability a rough mix can be a prototype mix within the hour if you have your chops together. That first hour has always been one which invokes slight excitement and nervousness in myself as you fathom what the sources have the potential to be. The judgement on the the individual sources and shaping them to become something greater when well blended is very exciting. This is where you gauge the potential of the sources and how hard you will need to work to obtain a sonic vision be your own or a producers.
4) Group your instruments
Groups are useful, predominantly for globally adjusting the level of a set of like instruments on a single fader but also for global processing. This could be using send effects such as reverb delays, chorus etc. and of course insert processing like compression or equalization. It is definitely worth setting up a few groups for your drums, guitars and vocals even if you are not sure you will use them at the outset.
5) Color is quicker
Making your project easier to navigate is going to make things more efficient. It might be a good plan to create some colour coding of tracks and channels. You may wish to develop your own colour templates which relates either to your own music or other types of music if you mix professionally. An electronic music production may have differences to acoustic r rock music for example. Once you have decided on a colour scheme that makes sense in your own mind you will find, mix after mix this becomes embedded in your way of working. This can speed up project navigation very nicely and keep the thoughts, impressions and remedial actions in a flow which gets results. It can be a good idea to colour your groups something specific and as a whole as there tends not to be too many of them.
6) Masterful processing
Stereo master bus processing is very personal, some people like to mix with nothing on the master bus whereas some like to use their favourite eq or compressor. This can of course be digital or analog on the way to your monitoring. I recommend keeping things subtle and if you find yourself using anything that could be deemed as extreme the chances are you probably need to work a bit harder with your mix sources first.
Limiters are extreme processes, by and large they are not necessary to create a good mix. I can understand the reasoning behind them finding their way onto a master bus. A few reasons are that you are doing your own self finalizing so you want to hear how the mix responds when being driven into a limiter or you might want to get a rough idea of what the track might sound like when professionally mastered. Some people simply like the sound of a limiter, after all not all limiting is bad.
With limiters it is most important to consider not boxing yourself into a corner. Mixing into a limiter for whatever reason will effect your mixing decisions. Drums and bass/present vocals will especially will especially be effected. Novice mixers will crush their mix to hold things in place (even limiting on channels), this will stunt your mixing skills progress in my personal opinion. It will also hinder any self mastering or professional mastering procedures. One reason some mixes sound good is often because there is space, interaction, dynamic interplay and transient information that adds excitement and power. Extreme use of limiting reduces all of that.
I would say to use bus compression and equalization you need to have a very true monitoring environment. You should be confident of the accuracy of the response of your monitors in room otherwise you can easily compound problems that already exist. This may make them more difficult to rectify later should you realize there is a problem with translation.So use these processes with care and attention to subtle tweaks rather than extreme changes. If a compressor or equalizer is adding something subjectively euphonic to a mix then there is no problem whatsoever. However if they are there to increase perceived volume only they are best left bypassed.
If you use the right gain structure there will be no need to protect digital zero with a limiter as ample headroom will be built in to the stereo master bus.
Mixing is often a very personal technique but applying some well grounded technical information and continuity between projects can bring enhanced results. This makes the way you mix more efficient and can definitely assist in achieving more consistent results and keep the mic down work flow logical and efficient.