So, I've decided that [for most rock] I like doing about 4-6dB total of bus compression. It sounds like a lot. Now let me explain.
Seeing as the consoles had headroom above 0.0dB and sometimes all the way up to something like +24dB, we have to consider a few things.
1) How hot are the tracks going into the master?
2) How much headroom are we leaving for the Mastering engineer?
3) What's my process? Why?
4) Why does it sound this way?
Let's address these in order.
1.) If you bypass all of your master fader plugs, your sound better not be clipping. Why? I'll tell you why. If your tracks set off the OVER (clip) meter, putting a limiter on the master doesn't mean it's not distorting digitally, it just means it's not triggering the meter. Digital busses aren't mind readers, they are mathematical devices only doing what they are told. If you trigger an OVER meter on ANY tracks, you are clipping. Just because you put a limiter on doesn't mean you're protecting the sound from clipping. HOWEVER...when it comes to a compressor, it's possible to elude this.
A compressor works by gain reduction, so, it's actually preventing clipping in a way. However...to be honest, if your tracks are hot enough where they are even CLOSE to 0.0dB, you're not leaving yourself nearly enough headroom per track. If 65 tracks combine at -6dBFS, your master is going to be SMOKING loud. I'd go for about -12 to -18dBFS per track on a large production. A
2.) You need to make sure and leave a good 3-6dB for the mastering engineer to work with. The more the better. If you're the last link in the chain (like many project studio owners), you STILL Should be mixing at about -6 or -4dBFS. It'll allow you to control your mix better. Don't mix with a limiter on the bus!!!!!!!!
3.) So, we've got our tracks in, not mixed yet. What are the levels on the bus? If I've gain staged the thing correctly, my master seems to hit around -6dBFS on a good day. Usually what I do is add Tritone Digital's Colortone Pro (for saturation) in the first slot. After that, I add Waves RenComp doing at MOST 1.5 - 2dB of compression with a very fast attack and a medium release (200-300ms). This helps control some of the transients. To set this, I go to the loudest part of the song and adjust from that. After this, I add the Waves SSL bus compressor, which I love. In this, I set the attack just a bit slower, usually 3ms, and the release a bit slower, maybe even 600ms, and on this compressor I do a max of 4dB of compression in the loudest parts of the song. It actually ends up to be more like 3dB, sometimes even 2dB.
So why nearly 6dB of compression on the master? A few reasons.
1) It helps the mix come together quickly.
2) I don't have to do nearly as much automation.
3) I don't have to use so many compressors on individual tracks, and this is the main reason. Not only is it more DSP efficient, it also just SOUNDS better. If you let the tracks just be naturally dynamic, the bus compressor can take care of a lot.
4.) Stay with me on these comparisons...Analog consoles and and analog tape will compress more per input signal in dB/ratio increments, as in, the compression was NOT linear. For example, at 0.0dB, the console might be compressing something "like" 2dB at 2:1. However, at +6dB, the console might compress a bit harder, something "like" 5dB at 4:1. That is to say, the compression CHANGES in a nonlinear way as you get hotter and hotter. That's not to say that +24 is simply a limited signal, because usually, +24 didn't sound too great.
SO, this is one of the reasons why I use two separate compressors on the mix bus. One is to control the transients, (as you can imagine, things -6dB to 0dB on a console) and the other is to control more of the overall mix punch (0dB to +12dB on a console - just made up numbers, just trying to illustrate a point). So basically, it's like saying two heads are better than one.
You don't just want to compress 6dB with one compressor, because you have a fixed attack and release. Instead, you can compress with separate multi-stage compressors for different parts of the sound. You could use SIX compressors only doing 1dB of compression and you might get a great sound doing that. Set each one for different ratios or attack/release times for different parts. A very pulsating tune might need more compression on the attacks to control them, however, a sparse tune might call for some more mild compression.
Moral of the story...MIX WITH YOUR EARS, NOT WITH YOUR EYES!!!!!