As Sanjay’s been working on his series of compression tutorials, I’ve been posting them on various production forums and the reception has been pretty well. However, someone made a comment that I hadn’t thought about and rang very true: there’s tons of articles on what the knobs and parameters of a compressor (none as good as Sanjays, though!), but no real articles about how and when to use compression. Hopefully this article is the first, and even if it’s not, hopefully it’s useful!

Before we get going, you really should read part 1 and part 2 of Sanjay’s articles.

Compression vs Limiting

For right now, it’s important that you understand that there’s a difference between compression and limiting. Limiting IS a form of compression, but not the one you’ll be using in your everyday productions (unless you’re a mastering engineer). I will also be talking about single-band compression and not multi-band or de-essing, both frequency specific forms of compression.

Compression is generally used in two major ways: 1) to control the dynamic range of an instrument and 2) as an effect like in sidechain compression seen in most popular dance music. Limiting, however, serves two slightly different purposes: 1) to make sure the levels between songs on an album/EP are equal to each other 2) and, sadly, limiters are often made to make stuff horribly loud.

If you’re doing everyday music production (whether it be with acoustic instruments or otherwise), you’d probably be best to stay away from limiting. Stick to using a compressor.

Compression the Utility

Like I said above, compression is both a utility and effect. The utilty being the ability to control a sound’s dynamic range, making it extremely dynamic or squashing it to make it loud.

When Should I Use It? In a recording situation, a great example would be drums: most drummers are not able to completely control the velocity at which they hit their different drums, which can make a drum recording sound uneven and sloppy (depending on the drummer). You can even run into issues of a drummer hitting the microphone and/or clipping the preamp. So in this case you would apply light compression on whichever drums you think need to be controlled: usually the snare, toms, and kick. When I say light, I mean around 4-5dB of gain reduction. Of course, there are no hard and fast rules, so compress it to taste.

Likewise with a guitar player. Sometimes their strums can be real quiet or clipping-ly loud. So treat it like any drums you’d compress. You can also compress an electric bass lightly depending on what you’d like to do. All this is also a technique (along with EQ) to get your instruments sitting right in your mix, that is, get them playing along with each other.

Throwing a compressor on a channel isn’t the only thing you can do to compress your mix. Here’s a few techniques to consider:

• Parallel Compression: This is a technique that blends an uncompressed and compressed signal together. I guess it’s pretty arguable as to whether or not this falls under the utility or effect category, but it’s a nice trick either way. When using this technique, you’ll typically want to buss a group of something (usually used with drums, but can be done with a synth, guitar, or vocal group) to a stereo aux track. Place a compressor on it and smash it pretty hard. Then, blend in the heavily compressed signal on the aux with the rest of the mix.
• Buss Compression: Very, very, very simple technique. Simply place a compressor on your master channel. You should be lightly compressing at around 2-3dB of gain reduction. This is often referred to as “mix glue,” meaning it helps glue your mix together a little bit better. Be careful with this, though. I would suggest not using this while you’re recording or even mixing. Save it for the end of your mix. It’s very easy to get carried away with this!

Compression the Effect

The only thing I can think of, off the top of my head, is sidechain compression. You know, that thing that makes synths pump in dance music? (If you’re still not sure, listen to Deadmau5).

Sidechaining is when the compressor listens for an external signal to trigger it. So in typical-House scenario, you’d set up a compressor on your synth track and have the compressor’s sidechain listen for the kick. So everytime the kick hits, the compressor effectively compresses your synth and, with the proper settings, you’ll get the desired pumping effect.

You can also get the pumping going on basses and pretty much anything else you can think of. I’ve even heard it on crash cymbals. The pumping effect usually isn’t desired with most acoustic instruments, though. So if you get a bunch of it while mixing your rock bands’ tracks, you’re over compressing.

Frequency-Specific Compression

I said I wouldn’t go into detail on frequency-specific compression, but here’s a quick overview of those technologies:

• Multi-Band Compression: Like limiting, this is typically a mastering engineer’s thing. Multi-band compressors will give you the ability to compress various frequency ranges. Some offer options like low, low-mid, mids, high-mids, and highs. Others might be three band and offer low, mid, and high. A good MBC will also have crossovers so that you, the engineer, can have some freedom.
• De-Essing: This is a single-band, frequency-specific compressor often used to remove sibilance from vocal takes. It has a sidechain and allows you to listen to offending frequenices.  You can then compress and hopefully remove and irritating frequencies from your vocals.

Hopefully this article will be of some help to you next time you’re mixing. If you have other questions, feel free to email us: askasoundguy[at]gmail(dot)com

7 Responses to “Practically Using Compression”