The topic of the day is audio compression. Compression is a common tool found on digital mixers, dedicated outboard units and audio software. Compressors can be tremendously helpful, but they are intimidating at first. This is partly because they seem to be doing lots of math, hidden behind complicated terminology. I’m not going to lie to you: compressors are not simple. They are probably the most advanced tools you will ever use in a mix. But I’m confident you can learn to use them and that’s why we’re here. To start, we need to understand what compressors do:
Compressors reduce the dynamic range of a source to help it sit better in the mix.
“Dynamic range” is the difference between the softest and loudest sounds in a source. A compressor reduces this range, evening out the levels of a performance and often making it sound louder in the end.
How does it accomplish this? When a source reaches a certain volume, the compressor kicks in and actively reduces the volume for a certain period of time. It’s like turning down the volume knob when the signal gets loud, except compressors can do this extremely fast, acting and releasing in a matter of milliseconds. Pulling the loud parts down brings them closer to the the level of the quiet parts, which is what evens out the level. But it doesn’t explain how they help sources sound louder. This happens in the last stage of a compressor: the gain control. The compressor’s gain, often referred to as “makeup gain,” compensates for the volume reduction in the first stage. Compression reduced the volume on the loud parts only. Therefore, if we boost the signal to where the loud parts match their original volume, the quieter parts–now closer in level to the loud parts–end up louder than they were in the beginning. The final result is a source that seems louder than before (for audio nerds: the peak level may be the same, but the RMS level is now higher). More than that, the compressed sound has a more consistent level and the subtleties (read: quiet parts) of the instrument or voice should stand out better.
Here it is in even more layman-like terms:
Compressors squash only the loud parts, then make the whole thing louder
Three ways to use a compressor:
There are certainly more than three, but catch these and you’ll be well past the point where this blog can help you 😉
1. Make a performance more consistent – it is often necessary, especially in live production, to use compression to rein in an instrument or vocal whose volume is all over the map. Compression can help even it out the levels so you don’t have to spend all your time riding the fader. It’s a common fix for keyboard patches with inconsistent levels, but the better approach would be to fix the patches themselves. It’s often necessary with more dramatic vocalists, but the better approach would be to improve vocal or microphone technique. This kind of use requires heavy compression settings which inevitably changes the tone of the instrument as well.
2. Make an instrument stand out or stay put – Lead vocals are the prime example of a source you want to stand out. If you listened to a singer up close in a quiet room you would hear all kinds of inflections, breath sounds, etc. These quiet sounds are quickly lost in a big room full band mix. Compression can make these quiet sounds louder, giving the impression that you’re hearing the voice up close and personal, even in a loud mix. This, along with a higher overall level, is what makes a lead vocal sound like it’s “out in front” of a mix. The same technique can be used to thicken the sound of quieter instruments to keep them from getting lost in the mix. A well compressed acoustic guitar, for example, can be mixed very quiet but still be present in the mix. I might help think of it like this: whereas EQ can help each instrument find its place in the frequency spectrum, compression can help each instrument find its place in the dynamic spectrum.
3. Change the envelope of a sound – this one just sounds advanced, doesn’t it? Throw the word “envelope” into a conversation and you’re instantly an expert! A sound’s envelope is the way it changes over time. For example, a snare drum has very loud “crack” when struck, then quickly decays to a quiet “ring”. But a synth pad builds slowly, sustains, then decays slowly even after the keys are let up. Snares and pads behave very differently relative to time, so we say they have different envelopes. Because a compressor affects gain over time, it always affects the envelope of a sound. A compressor’s time controls (attack and release) determine when the compressor starts and stops reducing gain. By carefully manipulating these time controls, a snare drum can be made to have less “crack” and more “ring”, or vice versa.
Think of it like this:
In the frequency domain, the “crack” is a higher frequency than the “ring” and we can emphasize one or the other with EQ. In the time domain, the “crack” takes place before the “ring” and we can emphasize one or the other with a compressor.
[for more on snare EQ, see this post from our series “Taking It Apart”]
Compressors have great potential to improve your mix, but this potential to help is directly proportional to its potential to wreak havoc on it. They are not the easiest tools in your toolbag. Although plugins and digital consoles now provide graphs to visualize what the compressor is doing, much of the sound-shaping action happens far too fast for our eyes to see. Once again we have rely on our ears. Mastering audio compression, then, is the work of translating a compressor’s settings into sounds we can identify, manipulate and reproduce. Its takes practice. To start, we need to identify what the compressor’s controls do.
Threshold – probably the most important setting on a compressor. The threshold is the amplitude (volume) at which the compressor starts acting. The threshold has more to do with your source than the sound you’re trying to create, so this parameter tends to get adjusted more often than the others.
Ratio – determines how much gain reduction will be applied once the threshold is reached. It is expressed as a ratio and ranges from 1:1 (no reduction) to ∞:1 (full limiting). Ratios 10:1 and higher can be referred to as limiting, whereas an ∞:1 ratio is called a “brickwall” limiter. If you’re new to camp compression, 2:1 to 4:1 is a safe range for testing the waters.
Gain – boosts the volume at the end of the signal path to make up for the gain reduction.
Two settings that may not be available depending on the compressor or setting. If your compressor has an “auto” mode, it probably locks these two settings to safe, predetermined values.
Attack – sets the amount of time the compressor will wait, once the threshold is reached, before reducing gain.
Release – sets the amount of time the compressor will continue to reduce gain, after being activated.
Other settings you may or may not see on your compressor:
Hard Knee/Soft Knee – changes the response of the ratio at the threshold. Hard knee compression acts on the signal according to the set ratio as soon as the threshold is reached. Soft knee compression starts acting at a lower ratio before the threshold is reached, then eases into the set ratio gradually as the volume reaches the threshold setting. Depending on the device you may see a Hard/Soft switch, or even a variable control over the knee size.
The good news is compressors all use pretty much the same controls. If you learn one, you can figure out how to use any of them. Go ahead, put your ears to work and have some fun!