Sooner or later you’re going to get to a point where you need to use a compressor. If you don’t really know what a compressor does, or even if you know what it does, but don’t have the slightest idea how to use one, this article is for you.
I often read articles on compression because it’s one of the darker arts of recording, and it’s interesting to pick up on the various different methods of dialling in a compressor. I’ve tried to break it down to some key ideas you can remember for the majority of compressor types.
What is a compressor?
It’s basically an automated volume slider. A compressor listens to the audio being fed to it and based on parameters you set, it turns the volume of the track down whenever it exceeds a given threshold.
So an example… if you have a recorded snare drum track that has been hit fairly inconsistently by a drummer, you’re going to get some hits at the volume level you want, and some will either be too loud or too quiet. You can’t leave the volume fader in one position as you’d like to because in order for the hits to be at the right level, you’d have to keep adjusting the fader. A compressor takes care of this by reducing the peaks automatically for you.
In fact what it does is reduce the dynamic range of your audio content so the difference between the loudest and the quietest part of your track is smaller. This enables you to set the fader in one position and have all the snare hits sound at roughly the same level. Not only are the peaks squashed, but because the dynamic range is now smaller, you can turn up the track and the quieter hits will also be a lot more audible.
Compressors are also prized for the side effects they create. But more on that later.
The Basic Controls
The basic controls on most compressors are as follows: Threshold, Ratio, Attack, Release, and Make-up Gain.
Let’s concentrate on Threshold, Ratio and Make-up Gain for starters.
Threshold tells the compressor at what decibel level to start compressing your audio track.
This knob kinda functions backwards to how you’d expect. Rather than starting all the way down and going clockwise, you start all the way up and turn anti-clockwise. At the maximum setting of the threshold control, no audio is being compressed. As you lower that threshold, waveform peaks begin to cross it and the compressor is triggered to compress those peaks. How much it compresses them is based on the next control, ratio.
Setting a threshold control: most compressors will have a Gain Reduction meter that shows how many decibels the audio is being compressed by. As you lower the threshold control, you’ll see the meter showing more and more of the signal being compressed. How much you lower the threshold depends on what your audio needs. So if you just want to control the loudest peaks, you’d turn the threshold down so only those peaks were crossing it. To reduce more of the audio’s dynamic range, you could lower the threshold further so more of the waveform would be affected by the compressor.
Ratio tells the compressor how much to squash the audio that has crossed the threshold.
A ratio of 1:1 doesn’t affect the audio at all. A ratio of 2:1 will halve the amplitude of the audio that crosses the threshold. For a ratio of 3:1, for every 3db increase above the threshold, only 1db will be output, and so on. Eventually, as you increase the ratio the compressor would start to function as a limiter – effectively creating a barrier that no peak would be able to cross.
Some compressor plugins give you a handy graph that shows the threshold and ratio in action. They often look a bit like the image below.
Setting the Ratio Control – It’s a good idea to keep an eye on your gain reduction meter as well as the track’s input meter as you increase the ratio. You’ll know when to stop when the track’s input meter is no longer peaking as it was and the signal is within a comfortable range.
Make Up Gain
This control is so-called, because rather than just using it to make your track sound nice and loud once it’s been compressed, it’s for compensating for the reduction in volume that the compression has resulted in. It’s good practice to A/B your track by bypassing the compressor to ensure that the amount of make up gain you have applied doesn’t make the track louder than it was before compression. Since we perceive that anything louder sounds better, it would be difficult to tell whether the compression we have applied is in fact any better at all.
Attack, Release and Knee
These controls help you further tailor the compressors action to the task at hand. They’re important because they control how quickly the compressor responds to audio peaks crossing the threshold. Attack is how quickly the compressor clamps down, and release is how quickly it lets go. They are usually measured in milliseconds but can also extend to microseconds or whole seconds in some models.
Setting the attack and release at different speeds allows you to customise the compressor’s action for different material. So for a snare drum, you’d probably want much faster attack and release times than for an acoustic guitar or vocal. To illustrate the different effects attack and release can have even on one sound like a snare drum, check out these examples:
Fast attack, fast release
The compressor kicks in quickly compressing the snare’s initial transient hit then lets go. Because the tail of the snare signal is uncompressed, it sounds louder in relation to the initial transient hit.
Fast attack, slow release
With these settings, because the release is set slower, the compressor is effecting the whole hit of the snare, which results in the overall hit sounding the same but quieter.
Slow attack, slow release
With a slow attack, the initial snare transient can sneak past the compressor but because of a slow release, the rest of the tail is compressed resulting in a louder first hit and a quieter tail.
Which leaves us one further option: Slow attack, fast release. In fact, this isn’t a useful setting as what it would give you is an uncompressed transient then a very short section of signal that was compressed and then released resulting in a rush back to normal volume. Kinda unnatural sounding. Alternatively it may just do nothing much at all.
Setting your attack and release values – first, work out what you want to do to your audio. What does it need? Do you need to control individual peaks or do you want to just control the average level of the content, reducing it’s dynamic range to hear quieter parts of the track better? If you want to control level changes in a much more general way, then you’d go for a fast attack and slow release. To give something more punch, use a slower attack and release so more of the transients poke through.
So called because it’s the hinge point on that graph above where the compression kicks in. A hard knee is an instant transition between no compression and compression. A soft knee is a gradual application of compression as it kicks in. This can help to make things a bit more transparent or smooth. Not all compressors give you this control.
There are solo many ways that compressors are organised in terms of the controls, and some features are given completely different names, so it’s good to get familiarised with the variations. I’ll tackle these variations in the next post.
I hope this has cleared up a few compressor mysteries for you. Now go compress with confidence! (remembering – ONLY IF IT NEEDS IT!)