[Only registered and activated users can see links. Click here to register]
And this is something I wrote myself:
I like the Michael Stavrou approach, explained in his book 'Mixing with your Mind' and it goes a little like this:
Setting up a compressor is like cracking a vault: every tumble has to fall in place and they have a particular order. Don't go to the next tumbler unless you are sure the previous setting is in the right place. Otherwise you will be attempted to keep changing previous settings, ending up in a loophole, chasing your tale.
Here is the right order to set up your compressor:
- Attack at minimum, shortest attack
- Release at minimum, shortest release
- Ratio at maximum (100:1, infinity, whatever the highest ratio of your compressor can be)
- Threshold to the lowest (-40 or -60 dB, so you get maximum compression on the whole song)
Now your music sounds completely crap, but you will hear the effect of each function clearly. You probably have to turn up the volume to hear the instrument/mix.
1. Determine your attack time. A short attack makes the sound thin, longer attack makes the sound fuller. Choose what you like. Ignore the distortion because of the short release time. Take a snare drum for instance; with a short attack time, it will sound like the snare gets hit with a thin stick. A longer attack will give the impression that the drum stick is thicker. A guitar with a short attack will give the impression that the strings are just touched by the fingernails, but with a longer attack time it seems like the guitarist is really hitting those strings. Just listen to what sounds best.
2. Setup your release time. The sound is compressed after the attack and will be uncompressed after the release. You will hear some kind of swing (loud-soft-loud). Control the swing by changing the release time. It doesn't have to be in time (don't set it on 1/16th or 1/8th note), just listen until it grooves. You shouldn't ask yourself you short the release time has to be, but 'how long can I get the release time without destroying the dynamics'. Don't get mathematically on this (I have to get a release at xx ms because the next sound will get crushed) but listen to the swing (loud-soft-loud) and how it goes with the music. Let your ears and your gut feeling decide, even if the release time looks insane. (I have been using release times of 800 ms on 140 BPM techno songs, just because it sounded right!) 'Groovy' doesn't have to be 'in time'. I have heard lots of 'groovy swings' that aren't in time with the music but it sounded good.
3. The ratio. You have the ratio at maximum, so you hear maximum 'swing'. Now start lower the ratio until you still hear the swing, but not too overdriven. Compare with the uncompressed version (use makeup gain to adjust levels). Uncompressed music is wide, compressed music sounds narrow but focused. Determine how wide or how focused you want your music to sound. You can compare the ratio with the zoom function of a photo camera; how much do you want to zoom in on the subject (high ratio)? And how much of the surroundings do you want on the picture? (= zooming out, making music wider) Compromise on this.
4. Threshold. Music has to breathe. Not all parts have to be compressed. That's called overcompression. On the quietest parts, compression shouldn't occur (let the music get some 'air' in the quietest parts instead of keeping it compressed). If your music is compressed all the time, you can as well lower the overall volume level, because the whole instrument/mix will sound quieter with compression). The purpose of a compressor is to only compress the loudest parts. So put your threshold in a way that not everything is compressed, and the music can breathe in between. This makes the compression subtle. If it's too subtle for you, just put a second compressor after the first one, and do the procedure again (serial compression). That way you will get a less subtle compression but still acceptable. Remember, if the gain reduction meter is pumping during the whole mix, you are overcompressing...
Now make it louder with the makeup gain and you're done.
This is a 'compressed' version (pun intended) of what's mentioned in the book 'Mixing with your Mind'. But it's worth to read the 8 pages because they are written much more colorful than this little part.
Thanks to this approach, I sometimes get results that I never would have gotten otherwise. Like that release time of 800 ms on a fast-paced 140 bpm techno song. Pure mathematical it's not right, because the release doesn't end before the next beat kicks in, but somehow it has that groove.
Thanks to the comparison of wide and narrow music (while setting the ratio) I learned to not overdo compression. I seldom use a ratio of more than 2:1 while at first 2:1 was unnoticeable for my deaf ears.
This method learned me that compression is about feeling, not mathematics.