Some thoughts on master buss compression.
As far as I’m concerned, the best preparation for a great master is…great stereo buss compression!
That’s quite a bold statement, since many people think compressing the 2-buss will overly complicate the ME’s job by reducing the dynamic range he can work with.
This may be true when bus compression is used for the wrong reasons: either because „that’s just how it’s done“, or because the mix sounds wimpy without it (hint: if it does, the problem is with the mix, not the missing compressor), or in a way that works against the tune’s groove or dynamics.
But great, judicious mix compression will enhance a great mix and let the ME work more and better magic.
Currently, my favourite stereo compressor for this job is the API 2500.
I use it when mastering other people’s mixes, pretty early in the signal chain, to give their mix a bit more punch, bounce, colour. Generally to add something that’s missing, rather than to enhance what’s there.
So there’s a limit to what it can do to some mixes (which can’t be changed anymore): if I apply more than a few dB of gain reduction, or too high a ratio, or both, it starts to work against the mix. The track will sound squashed and/or start pumping, losing life and groove.
Even worse: the balance of sounds will begin to shift unpredictably, changing the mixer’s original intent.
Later, when the mastering limiter is applied for loudness, there’s also less maneuvering space, because the sweet spot between fast limiting (loud, but risk of distortion), slow limiting (clean, but risk of pumping and less loudness) and the available loudness (before it starts affecting the balance and feel of the mix) is very small.
The usual compromise in this case is having it quite loud, but with a slow release, smearing the transients, losing punch and further changing the balance.
Personally, I prefer to use it on the mix bus. But I don’t „mix into it“ from the start.
First I make sure that my mix has impact and groove without the compressor.
The individual tracks and submixes (often heavily compressed themselves) must have found their respective place and dynamics, and the fader rides be programmed.
Only then do I insert the 2500 across the mix buss.
And not just a bit: I start with the settings suggested by its designer Paul Wolff: soft knee, old mode, med thrust,10:1 ratio (!), 200ms release, 70% of BPF link, autogain.
I’ll shorten the attack until the snare starts moving back in the mix, and then set it 1 click slower. The input/threshold will have the meter shows 10-20 dB of gain reduction (it’s actually less).
The mix will get denser, with punchy drum hits followed by a dense, solid tail of sound.
Now, the mastering limiter can do what it’s supposed to: make things LOUD!!!
It’s not busy glueing things together with a slow release, so it can be set fast (= loud).
With its fast release, it’s also backed off before the next drum hit (= more punch)
It can be applied harder before it starts affecting the balance and dynamics (= still louder).
So, there you have it, maybe it’ll inspire someone to experiment more with compression on their 2-buss…