By John Mills

What is a compressor used for? Why do we need compressors? What do all those knobs do and how do I set them? Ask 5 sound guys and you'll likely get 5 similar, but different answers.

Here are my thoughts on basic compression. I'm not going to get into advanced compression used for effects, but will try to offer some basic advice on how to tame a signal.

What is a compressor?
Compressors are devices used to reduce the dynamic range of a sound. The dynamic range is basically the db change from an instrument, or vocalist's, quietest to its loudest signal.

Since most things in real life exhibit very wide dynamic ranges, compression just doesn't sound natural to our ears. Natural sounding compression is hard to hear. I say it's hard to hear because when it's wrong it's noticeable, when it's right you shouldn't even know it's there. Things will just seem a lot more �in the mix� and in control.

Do I need a compressor?
The best answer to that question is no. I'm a bit of a purist when it comes to sounds. In a perfect world the instrumentalist or vocalist would be in complete control of their dynamic range. I remember reading a story once about a very accomplished singer. She hated the way compression sounded on her voice, but the recording engineer said they had to use it. Long story short, she sat with the engineer for hours, singing different lines and watching his needles on the soundboard. She learned how to work her mic, and vary her singing techniques to deliver a very even vocal performance. But in the world we live in and the varied talent of the musicians we work with, compression is often required to make things sit in the mix.

What do all those knobs mean?
All good compressors will have the controls illustrated in figure 1. Basic definitions and tips of the controls are as follows:
Gain Reduction Meter: Shows the amount of the input signal that is being reduced. If you see the �6 light lit up you are decreasing the input signal by 6db. You should probably reach for the Gain knob and add 6 db back to the output.
Threshold : Compression will occur when the input signal exceeds the setting on this control. I generally set this so that when the vocal or instrument is playing a quiet note, the gain reduction meter is just about to light up the first light. Remember most compressors gain reduction meters light up from left to right.
Ratio: This is the amount of change a signal undergoes after the threshold has been exceeded. A 4:1 ratio would mean that for every 4 db of change in the original signal there would be only 1 db of change on the output.
Attack: This is the delay between detection of a signal over the threshold and when the compression kicks in. A fast attack could be used to squash the �crack� of the snare drum, while a slow attack may be used to let the percussion of the bass guitar pluck be heard, while at the same time taming the overall note being played.
Release: The release is the time taken to release the signal from compression. This is the trickiest one to set and also easy to miss when it's wrong as the other instruments in the mix often mask the sound of the release. When in doubt put on your headphones, hit the solo button and check it out. The general rule of thumb here is that it should decay at about the same rate as the real sound decays. If it's too fast the sound will get louder at the end of its decay.
Output Level Meter: Indicates the level of signal leaving the compressor after the compression has been applied. You will want to get the meter back up to between zero and +5 to make up any gain you lost when applying the compression.
Bypass: With bypass engaged the input signal passes through the compressor unchanged.

What settings should I use?
While there are a few magic numbers and settings, I'm not going to give them to you. I want you to learn to HEAR compression. Because when you hear it you can fix it. When you look at the dials and think, �That's where John said to set it,� but it doesn't sound right, you may feel inclined to not try and fix it.

Ok, ok� before I get a bunch of emails, I'll give you a few guidelines. Remember though, these are only guidelines. Every person sings a little different, every bass player plucks a little (ok a lot) different, every drummer hits loud, but different; so use your ears to tame the dynamics of whatever signal it is.

First Steps.
•  Unplug (or bypass) the compressor.
•  Set your gain correctly on the channel. That was back in Volume 1, issue 2 of the magazine, if you need a refresher on gain structure.
•  Now put the compressor back in the line and keep reading.

Here's what I usually do to end up with a basic setting for just about anything. At first this is going to sound pretty bad, but bear with me; I'm going to show you how to really hear what's going on with compression.

For this example I'll be referring to a bass guitar, but the concept it the same for any other instrument or voice.

Set the Gain to zero, the ratio to 10:1, the attack to the fastest (all the way counter clockwise) and the release to about 1 second. Now play the instrument, grab the threshold knob and turn it so that you are seeing about �15 on the gain reduction meter. You are now hearing a very squashed and unnatural version of the original bass guitar.

Take the attack control and head clockwise with it. You'll start to notice that the beginning of the note starts to sound more defined. This is because the compressor is letting the original �attack� of the bass through because we are delaying when the compressor squashes the sound. A good setting will usually be somewhere between 5 and 20 milliseconds. You'll want to set the attack so that you hear the initial transient, but the remaining part of the note is not affected. If it is set too long you'll hear the note pump as the compressor kicks in too late.

Now go for the release control and watch the gain reduction meter. The meter should be letting go of the signal as the note decays. If you hear the instrument decaying and the meter is still hanging down in some serious compression you'll want to make the release faster. It should flow with the music.

Now don't be afraid to crank those attack and release knobs, feel them out and hear what they do.

Next, back off the ratio to 2:1, listen for a bit, and now crank it to 20:1. Do you hear the difference? The 2:1 setting is only creating a soft effect. The bass will sound pretty natural, but you won't be correcting much dynamics. The 20:1 setting will literally make every note the same volume, but that's no good as you would be sucking the life out of the music. What you are looking for is that the loud notes are now tame, but the soft ones still sound softer.

Every instrument is a little different so play around. What you are going for is manipulating the dynamics of their playing so that their quiet notes are no quieter than the other players and obviously that their loudest notes aren't louder than the others.