One of the first things any recording engineer or producer will invariably come across is compression. It seems like some sort of mythical beast, untameable and obscure. This article will break down what exactly compression does, cover some of its uses and applications within the audio world, and the 5 main functions you’ll find on just about every fully-featured modern compressor.
In simple terms, compression reduces the loudest points of your waveform and brings up the quietest parts at the same time (hence compression of the dynamic range). Why is this helpful? It increases ‘perceived’ loudness, making your instrument or track sound louder and more dynamically even without increasing the volume and taking up precious headroom in your mix.
We’re using Ableton Live’s standard Compressor effect as an example, but most compressors will have the following parameters:
Usually expressed as a –dB level, a compressor will not start applying gain reduction until this audio level is passed. That way gain reduction will only take effect when your track reaches a volume of your choosing.
If you have a compressor set to 4:1, this means that for every 4 decibels over the threshold, 1 is allowed through. The higher the ratio, the more the dynamics are squashed, with an infinite ratio turning your compressor into a limiter, which allows nothing past your threshold. Having a high ratio will cause your compressor to start ‘colouring’ the sound and imparting its character onto your track. This can sound great or it can sound shit, depending on the instrument and compressor being used.
The time it takes for the compressor to start applying gain reduction once the threshold level is passed. If you want to preserve the transients (ie. The ‘smack’ of a snare drum) of your track a longer attack is better as it will allow a small snippet of the audio to poke through before the compressor kicks in and tames the signal. A short/fast attack is better for vocals and anything that you want to be consistently compressed.
How quickly the compressor’s gain reduction backs off once the signal dips below the Threshold again. Ideally you’d use a fast release time with sounds that have a fast release (Snare, kick, Hi Hat) and slow release with (you guessed it) sounds that have a slow release for a smoother, more natural sound.
This allows you bring your overall track volume up after the dynamic range is compressed, thus giving your track a perceived increase in loudness without it actually being louder and messing with your mix.
There’s a number of other things we can talk about with compression, we haven’t mentioned Knees or Sidechaining or Parallel Compression, then there’s Peak compression vs RMS compression and Compression vs Expanding. But at this point, let’s keep it simple. Those five functions above are what you’ll adjust 99% of the time.
There is no set rule to compression, its application is entirely up to the user. You can use gentle compression in podcasts to keep everyone’s volume consistent, or you can crush your kick drum into oblivion to make it tougher than overcooked meat. Like all processing it’s up to you and your ears.