Fine-tune compression

[KingRazor]

So I have a pretty thorough knowledge of how a compressor works. Now I'm trying to fine tune my compression settings.

I wanted to ask:

What is considered a "fast" attack? 1-5ms? When is it no longer considered "fast"? Same question for release.

Output gain - what do you use to base your level of output gain?

[projectMalamute]

In an objective sense I guess a 'fast' attack is something like the 1176, which is like 300 micro-seconds or something absurd like that. The dbx 166 has a program-dependent attack time that varies from a fast-ish 3 ms to a slow-ish 15 ms.

Really it depends on what you are trying to do. Do you want to even out the transient or fatten up the sustain? And are you using this on a snare drum or a vocalist?

I tend to run the makeup gain ball park around the same as what I'm seeing in terms of gain reduction. If I'm taking 6 or 8 dB off of most peaks I run 6 or 8 dB of makeup gain. The compressor is always before the fader anyway, so I just get it in the range where I'm in a comfortable part of the fader's travel and mix from the console.

[KingRazor]

I just wanted to know what is considered a "fast" attack. On the compressor I'm using you can set the attack anywhere from 0ms to about 100ms. With the Release you can set anywhere from 0ms to 8000ms.

[projectMalamute]

What is considered a fast car? When is it no longer considered fast?

Use your ears. Maybe try loading some files in to an audio editor and look at the envelopes of different sounds, try and imagine how different attack times are going to work on different sounds. Mostly use your ears.

[KingRazor]

My ears can't discern the difference, really. That's why I was asking. A lot of times when reading up advice on how to compress various inputs they'll say "use a fast attack" or "use a slow attack", I have no idea how fast they mean though. I was hoping to find some sort of standard, like "1-5ms is considered fast".

[ChasBass]

Check the presets section in this Presonus Bluemax manual Page 10

http://www.presonus.com/media/manual...inalManual.pdf

Has suggested settings and brief explanations.

[projectMalamute]

You will notice the difference most easily with transients. Try this:

Figure out some way to get loops of a few very simple sounds. Audacity, Hydrogen, whatever. Free software is available to do this if you do not already have something working. Start with a snare drum, quarter notes at like 80 bpm or so. crack ... crack ... crack ... etc. If you don't have the means to patch a computer in to the system you are using burn it on to a CD and bring that with you.

Turn the ratio up pretty high, higher than you would ever actually use on a snare. We are not looking for something pretty, we just want to hear the results of different attack times. Start with the attack as long as it will go. On a sound like a snare you will probably not be hearing anything, by the time the compressor has clamped down the sound is over. Now start moving the time down in small increments, comparing the bypassed to compressed sound. As you start seeing some gain reduction happening compensate with the same amount of makeup gain. At first the initial attack will still be unaffected but you will start to hear more sustain in the body of the drum. That's a slower attack time letting the initial transient through and then clamping down on the body. Eventually you will get to a point where you hear a marked difference in the sound of the initial crack. At that point you are using an attack fast enough to clamp down on the initial transient.

Try the same thing with a kick drum. With a low pitched source and a fast enough attack you can eventually get to the point where you start hearing distortion as the compressor clamps down on each cycle of the wave. Listen for that.

That's just a couple things to get you started. Again, the key is listening. Mix with your ears, not your eyes.

[projectMalamute]

Originally Posted by ChasBass

Check the presets section in this Presonus Bluemax manual Page 10

http://www.presonus.com/media/manual...inalManual.pdf

Has suggested settings and brief explanations.

Those setting look really, really screwy. Like maybe they are jumping back and forth between microseconds and milliseconds while using the same abbreviation for each. Or something, I don't even know.

They are calling the 47ms attack on bass guitar 'fast' (by almost any estimation that is pretty slow) and recommending a 188ms attack time for acoustic guitar, despite the fact that the unit only goes to 100ms!