Compression is used to reduce the dynamic range of an audio signal. A compressor is like an automatic volume knob that turns down an audio signal’s level when it gets too loud. We’ll be taking a look at how to identify when to use compression in your mix.
Before trying to operate a compressor, it’s critical that you understand what the different parameters of a compressor do. I wrote an article called “The Ultimate Guide to Compression” that discusses all the parameters found on compressors in detail; if you’re new to compression, it’s a great place to start.
I want you to think of a compressor as a problem-solving tool. Don’t use one unless you have an identifiable issue that you know compression is going to correct. I’m going to provide you with 3 scenarios in which it makes sense to use a compressor, along with some recommended compressor settings that you can experiment with. Please take these compressor settings with a heavy grain of salt because each compressor is unique in its design and functionality.
The stock compressor that comes with your digital audio workstation (DAW) is probably excellent for general compression purposes. For example, Logic Pro X has a very powerful stock compressor that includes multiple different compression algorithms. If you’re looking for a versatile third-party compressor plugin, I recommend checking out FabFilter’s Pro-C 2 or Waves’ H-Comp Hybrid Compressor.
1. When Transients Are Sticking Out of Your Mix
The most basic use for a compressor involves taming transient material using downwards compression. Perhaps your snare track “bites” too hard, or the bass you’re processing has too much “slap.” In both of these situations, an element of your song is refusing to play nicely with the rest of your mix; when you turn the troublesome channel’s level up, the transients are too loud, and when you turn it down, the channel’s soft parts are too quiet.
Peak compression can be used to deal with overly present transients. The idea is that you set up your compressor so that it reduces the level of transients while leaving sustained material unaffected. This requires a fast attack and release time. You want the compressor to engage as soon as signal rises above the threshold, and you want it to stop compressing as soon as it falls below the threshold.
Peak compression has been applied to the second half of the following audio example. To clearly hear the effects of peak compression, listen using a pair of studio headphones or monitors, and turn your audio output up until the transients in the first half of the audio file become slightly uncomfortable to listen to. In comparison, the transients in the second half should sound much more tame and controlled. The tail end, or ring of the snare should sound roughly the same level in both the first and second half of the audio file; this is because the compressor quickly disengages as the signal falls below the threshold.
If you set your attack time too fast, you may suck all the punch out of your signal, but if you set the attack time too slow, the compressor won’t clamp down on the transient in time. Striking a balance is important here, but for a frame of reference, you’re probably going to be searching for an attack time between 1-10 ms.
I usually default to a release time of around 25-50 ms when applying peak compression; this is also very dependent upon the compressor being used. The main thing to look out for when using fast release times is top-end distortion. As you increase your release time be wary of undesirable pumping effects; slow release times are responsible for this.
Recommended Settings for Transient Compression
- Attack: 1-10 ms
- Release: 25-50 ms
- Ratio: 2:1 (Low) to 10:1 (High)
- Threshold: Set your threshold below the level of the peaks you want to attenuate. The exact level depends on how much gain reduction you want to apply.
If you’ve applied peak compression to your audio file, the sound will appear further away. Sound waves produced by a sound source far away lose their transient properties as they propagate towards your ears; this is a result of absorption, diffusion, and diffraction.
Someone striking a snare beside you will sound much more harsh than someone hitting the same snare 100 feet away. If you were to record these two signals and level match them, you’d be able to hear that the further recording appears to have a smaller dynamic range.
The alternative to using downwards compression is using upwards compression, which boosts the level of signal that falls below the threshold you’ve set. This will avoid reshaping your transients while still reducing the dynamic range of your signal.
Ableton’s stock Multiband Dynamics effect allows you to perform upwards compression, as does FabFilter’s Pro-MB if you increase the Range knob to positive values. The following tutorial by FabFilter does a great job of demonstrating how upwards compression sounds.
2. When Your Mix Isn’t Transient Enough
In addition to attenuating transients, compression allows you to emphasize transients. This is done by compressing the sustained material between transients so that in relation to the newly attenuated sustained content, the transients are more pronounced. I want to really nail it into your head that the transients don’t get louder. Instead, the signal between the transients gets quieter; this is where a lot of the confusion involving this compression technique comes from.
Perhaps you’re mastering a track in which the kick and snare, two of the most prominent elements in the mix, aren’t present and impactful enough. By attenuating everything but the kick and snare, you’ll be able to draw attention to these two elements and help them sit above other track elements like synths, pads, and vocals.
When applying peak compression, the goal is only to compress the transients. With inter-transient compression, the goal is to do the complete opposite, and only compress the non-transient material present between transients. The result is a dynamic signal that breaths; if you push your compressor hard enough, you can achieve desirable musical pumping effects.
You'll notice in the image above that compression isn’t applied to the signal until after the transient has peaked; this is due to a slow attack time. A slow release time causes the compressor to remain clamped down on the signal until well after it has dropped below the threshold. The result is an audio file that’s snappy and punchy.