Compression 101

Compression is one of the most important tools in any producer or audio engineer’s playbook, but it’s often misunderstood. It’s often thought of as mysterious and hard to get, but in truth, it’s not that complicated. Here, we’ll start you off with a basic overview – compression 101, if you will. If you’re brand new, this should help you lay a solid foundation, and if you’re a salty old veteran, this review may clear a few things up or just remind you of the basics.

What is Compression?

Put simply, compression is the process of reducing the dynamic range of an audio signal. Dynamic range is simply the difference between the loudest parts and the quietest parts of a signal. For example, if you have a tone playing that stays exactly the same, that signal has no dynamic range – the loudest part is exactly the same volume as the quietest part. On the other hand, if you scream into the mic and then whisper, that signal has a huge dynamic range.

Compression makes that difference smaller. It does this by turning the loudest parts down. It’s as simple as that! But as simple as it is in concept, the devil’s in the details. How a compressor reduces dynamic range – how fast, at what level, and so on - makes a huge difference in how it sounds. At the end of the day, reducing the dynamic range of a signal in a variety of ways is incredibly powerful. That’s why compression is so important in production and live audio.

Compressor Settings

Rather than hemming and hawing, let’s get right to compressor settings and what each does.

Threshold

The level at which the compressor starts compressing. Anything above the threshold gets turned down. For example, if you set the threshold at -6 dB, the compressor will turn the volume down every time the signal passes -6 dB. Simple enough, but can drastically change what happens to a signal.

Ratio

How much the compressor turns the signal down. A higher ratio means it turns it down more. 1:1 means no compression. 8:1 is a lot. You can go all the way up to infinity:1. Let’s take a look at what that first number actually means, though.

A 2:1 ratio means for every 2 dB a signal exceeds the threshold, the output is lowered to 1 dB above the threshold. So, it doesn’t mean the signal is reduced under the threshold, just less over. If it were 5:1, the output would be 1 dB over for every 5 dB the input was over. That may be kind of confusing so let’s look at an example.

Let’s say the input signal exceeds the threshold by 20 dB. At 2:1, the output signal would be reduced to 10 dB over the threshold. At 5:1, the output would be reduced to 4 dB over the threshold. As you can see, the latter is more compressed. Moving right along!

Attack

The attack is how fast the compressor responds once the signal exceeds the threshold – it’s usually measured in milliseconds. It’s important to understand that the attack is the time before the compressor starts turning the volume down – not the time it takes to turn it down.

Knee

Not often understood, knee is a crucial setting. This is the amount of time it takes to turn down the volume once it starts being reduced. There are two types of knee settings – hard and soft. Hard knee means the volume is reduced all at once immediately – like a switch. A soft knee means that it’s reduced gradually – how gradually is determined by the knee setting, measured in decibels. The higher the knee, the more gradually the compressor engages.

Think of knee like how fast you turn the volume knob. A higher knee means you’re turning the knob slowly. A hard knee is like hitting a quiet button.

Release

Release is the time the compressor takes to turn the volume back up to normal. This is how long the compressor takes to finish turning it back up, so knee doesn’t affect release. Hence, you’ll notice release time settings are longer. They’re still measured in milliseconds, but in hundreds of milliseconds rather than 10, 20, etc.

Makeup Gain

Makeup gain is super important to compression, and it’s the simplest setting (other than input gain). Since compression turns the loudest parts of a signal down to reduce dynamic level, makeup gain allows you to turn the whole thing up to compensate.

The result is that compressed signals can sound louder or cut through a mix more, because the net of the process is that you’ve turned up the quieter parts.

Common Uses of Compression

Compression is used all the time for all kinds of reasons, and it can be a powerful creative tool, even to the point of shaping signals and creating wild, new sounds. Still, there are a few super common, straightforward uses of compression you’ll see time and time again:

• Vocals – Compression is almost always used on vocals, especially in dense mixes, but even in sparser material. Compression brings out the notes and words that get lost, makes the vocal sit in the sweet spot, and brings them forward so they can be the center of attention.
• Overall Mix – Compression is used on nearly every mix to glue the mix together and start the process of making the mix sound loud enough for commercial use.
• Mastering – Compression is applied again during mastering as part of a process of bringing perceived loudness up. Subtle compression may be applied in several layers (serial compression) and/or applied in parallel, to preserve the dynamic nature of a song while bringing up perceived loudness. In other cases, a special subset of compression, limiting, is used to finally make the mix sound as loud as possible.
• Instruments – Compression can be used to tame transient spikes, make instruments like drums or guitars sound more aggressive, or even add sustain and presence (especially common for guitars).

Compression is a lot like guitar playing: if you’ve never used it, it can be intimidating, but once you jump in, it’s actually pretty easy to get started. Then comes the lifelong journey of mastery. So, there’s a lot more ground we can cover talking about the subtleties of compression. We’ll do that in subsequent posts, so check back for more!

Also in Audio Info & Education