Hey y'all,
I just made a vid on compression (https://youtu.be/TUOBcCa0Wfw), as it's so important (and so often misunderstood).
I go through the controls and what they do. I also go through the 4 main reasons for using compression.
Here's a quick summary if you can't be bothered to watch the vid (shows audio examples, so recommend it if this is new to you):
-
Compressors are just volume controllers. A bit like having an elf on the mixer, riding the volume fader up and down in response to the audio signal entering it. Filthy little elf.
-
All compressors have threshold, ratio, attack, release knobs. Decent ones also have soft/hard knee and output gain. Here's what they do:
Threshold: Sets the level above which the signal is compressed. so if it's set to -4dB, only audio louder than -4dB will be compressed.
Ratio: The extent to which the audio above the threshold is compressed. A 1:1 ratio won't compress it at all (i.e reduce it's volume). 1:2 will halve the volume. 1:4 will quarter it, etc.
Attack: This determines the speed at which the compression is applied. So a very quick attack will compress the signal straight away, whereas a slow attack will let the first part through without compressing it.
Release: As you'd expect. A quick release will stop compressing the signal quickly (i.e. stop reducing the volume of anything above the threshold), and a long release won't.
Soft/Hard Knee: Determines how smoothly the signal is compressed above the threshold. If a hard knee, the full compression ratio is applied to EVERYTHING above the threshold. The softer the knee, the more gentle the transition from non-compressed to fully compressed. Visually, it would be kind of like two colours either having a sharp line between them, or a smooth gradient.
Output Gain: When you compress a signal, it reduces it's volume. The output gain is just used to make-up that volume at the end of the compression process.
Here are the 4 main reasons to compress:
-
Taming the peaks of a signal. If you have a vocal with some loud parts and some soft, you can pull the loudest parts into check to avoid clipping. Generally gentle settings on the threshold and ratio. Season with attack and release to taste.
-
Fattening a sound. Basically making it more sustained and squashed, which allows it to sound louder and fatter without clipping. Short attack, long release. Tip: Don't do this on everything, and do it for a good reason! Can kill a mix's dynamics if abused.
-
Adding energy. Great for percussive stuff. Longer attack (to let the transients through, like a drum hit) and longer release.
-
Glueing sounds together. If got a layered sound, some light compression can "gel" the layers together and make them sound more cohesive.
That's it! Here's the vid: https://youtu.be/TUOBcCa0Wfw
Other uses would be sidechain compression, buss compression, parallel, mastering, etc. but all based on those 4 ideas.
Hope you enjoy, and let me know if I've missed anything / been unclear.
Cheers!
Submitted January 25, 2018 at 02:45PM by willdarling https://www.reddit.com/r/edmproduction/comments/7sxtg1/how_and_when_to_use_compression/?utm_source=ifttt