Q. What does 'parallel compression' mean?

Sound Advice : Mixing

As a beginner, it seems that the more I explore, the less I understand. Certain techniques are referred to a lot but go right over my head. Parallel compression is something I read about that is a complete mystery to me! Could you explain what it means?

Eliza Monterosso via email

SOS Reviews Editor Matt Houghton replies: 

Essentially, this is a technique whereby you use the compressor as a send effect (you’d traditionally use it as an insert), so that the dry and compressed signals are running in parallel — just as is shown in the diagram on this page. You’d normally achieve this by taking the source track and then using an aux send control to ‘send’ some of the signal to another channel with your compressor on it. You might then bus both the source and compressed channels to a group channel. In a modern DAW, you could achieve a similar thing by duplicating (or ‘multing’) a track, but I prefer using sends, as even after edits you’ll know that you’re working from the same source audio. Increasingly, new compressor designs include a wet/dry blend control, which avoids all this routing and allows you to perform parallel compression on a single channel. However, you might or might not process the compressed version differently from the dry version, and a simple wet/dry control doesn’t allow you to do this.

Parallel compression is often used on drums, and the technique is frequently referred to as the ‘New York drum trick’. The effect can help to keep a part solid and ‘anchored’ in the mix, while retaining some of the dynamics of the original — and you can easily determine how dynamic the part is in different sections of the song by balancing the two channel faders. A common extension to this trick is to put a slight ‘smile’ EQ on the compressed drums.

The clue to parallel compression is in the name: two signals — one ‘dry’, and one compressed — run in parallel to one another. This is often used with drums, as in the diagram, to blend wet and dry signals, thereby maintaining some of the original dynamics.

Some engineers take the parallel compression principle much further than this, though, and you don’t have to stop at a single compressor. Michael Brauer, for example, reportedly likes to send vocal parts to several different compressors and blends the results to taste. You can read more about that in a Cubase technique feature we ran back in SOS April 2009 (www.soundonsound.com/sos/apr09/articles/cubasetech_0409.htm).

One thing you should bear in mind is that you need to have plug‑in delay compensation switched on in your DAW to use this technique, particularly when working with transient‑rich sources like drums. Otherwise, when the parts are summed back together on your drum or master bus, the result might phase nastily, robbing your drums of power.

Published in SOS August 2011