Compression is an important audio effect in music production, and it’s used in various ways to improve the sound of individual instruments and vocals, and tracks overall.
You often hear that audio compression can reduce “dynamic range”, increase the the overall loudness of a track, and make the sound appear to be closer to you.
However, the type of compression that I wanted to explore here is the use of a compressor to “glue a mix together”. I want to find out more about what is meant by a “glue compressor”, and how it’s actually used in music production.
A glue compressor is an audio compressor through which several elements of a mix, sometimes all of them, are passed. By applying the same dynamic range reduction to the various elements they sound like they belong together, or are “glued” together. This is also know as “bus compression”, since large hardware mixing consoles often had a built-in bus compressor to help “glue” a mix together.
So how does a compressor process audio to “add glue” to a mix of individual tracks, and earn the name “glue compressor”.
First, What Is Audio Compression?
Before we start discussing glue compression and glue compressors, we’ll take a look at audio compression in general. So, what is a compressor and why would you use one in music production?
You sometimes hear an audio compressor described as an automatic volume control. Every time the volume of a sound goes above a pre-selected level the volume is turned down.
I’m using the word “volume” here to explain this, although it’s not strictly accurate, and you often see “gain” and other terms used instead.
The image below shows some of the main controls found on audio compressors: Threshold, Ratio, Attack, Release and Make-Up Gain. We will look at each one in turn in the following sections.
Compressor Threshold and Ratio
The term used for the level at which the volume is turned down on an audio compressor is the “threshold“. The term for the amount the volume is turned down is the “ratio“.
So, if the volume is turned right down to the threshold level every time it goes above it, the volume will never go above this level. In this case the ratio would be “infinite”, meaning that the level of the sound cannot rise above the threshold level. A compressor set up like this is often called a “limiter”.
In most cases, you won’t want the sound level limited in this way. You will just want it turned down a bit each time the level exceeds the threshold.
If you want the level turned down to half-way between the threshold and its actual level, the ratio would be set at 2:1.
If you want the level turned down to a third of the way between the threshold and it’s actual level, the ratio would be set at 3:1.
I’m sure you’ve got the idea, but here’s one more. If you want the level turned down to a quarter of the way between the threshold and the actual level, the ratio would be set at 4:1. And so on.
You can read more about how compressor threshold and ratio work in this article.
Compressor Attack and Release
Two other controls that are usually found on audio compressors are attack and release.
The attack control lets you adjust how quickly the compression is applied to the sound when the volume goes above the threshold level.
If you set the attack control to a very short time then compression is applied almost immediately as the sound goes above the threshold. If you set the attack time to be longer, then some of the sound gets through at the higher level before being compressed.
The release control does the opposite of the attack control. As the level of the sound drops below the threshold it stops being compressed. The release control lets you adjust how quickly this happens.
The diagram below illustrates this process. Input level refers to the incoming audio signal. Output level is the compressed audio signal leaving the compressor. Image from Wikimedia Commons.
Proper explanations of this release process are quite hard to find, but here’s my simple understanding of it. If a short release time is set, as the level of the signal drops below the threshold the compression stops very quickly.
However, with a longer release time the signal continues to be compressed (even though it’s now below the threshold) for a little while.
This means that a short release time can lead to a kind of abrupt, pumping sound, while a longer release time will result in the compression sounding more natural.
Compressor Make-Up Gain
The last compressor control that I wanted to look at here (although some compressors have more (or fewer) than this) is the make-up gain control.
Compressing the audio as described will usually result in a reduction in volume of the sound. Make-up gain allows you to turn the volume back up so that the audio sounds as loud as before.
This is the main way that a compressor can make a piece of audio sound more intense. The peak levels have been turned back up to the same level as before they were compressed, but the quieter parts get turned up too, as outlined below.
Dynamic Range Reduction
All the parts of the audio that were below the threshold wouldn’t have been compressed. This means that when you add make-up gain you bring the peak levels up to where they were prior to compression, but you bring the lower level sounds up with them.
This results in a more intense sound since these quieter parts are now louder, and the levels are closer to the peaks that were compressed.
What you have here is a “reduction in dynamic range”, which means the difference between the louder parts and the quieter parts of the audio is now smaller.
This is an important part of the sound of modern audio that we’re used to hearing. However, overdoing it can result in a lifeless sound due to the absence of dynamic range.
But What About Glue Compressors?
Having reviewed some basic aspects of audio compression, what is it that a glue compressor does, and is it different to what we’ve just been looking at?
When I hear about various audio effects I initially think that the name can’t possibly describe the ways they work.
For example, a plate reverb or a spring reverb can’t possibly used plates and springs, can they? Yep, the sound causes the plates and springs to vibrate producing a reverb effect.
It’s the same with hardware compression units. An hardware optical compressor can’t possibly have a light bulb inside can it? Yep, that’s what causes the signal to be compressed.
So could a glue compressor contain glue to make it work? Thankfully not!
Ableton Live (Standard and Suite versions) comes with a compressor plugin called the “Glue Compressor“. This seems like a good place to start. What might be special about this compressor compared the the standard Ableton compressor, or any other hardware or software compressor?
The first thing to note is that the Ableton Glue Compressor has the same controls we have just been talking about.
In the middle panel under the dial is a Threshold control and next to that is the Make-Up (spelled Makeup) gain control.
At the bottom of the left-hand panel is the Ratio control, and above that the Attack control (measured in milliseconds) and Release control (measured in seconds).
So far, no surprises.
The dial above the Threshold and Make-Up gain controls is the Gain Reduction meter. This shows you how much the gain (similar to volume) is being reduced by the compressor. The more you lower the Threshold or raise the ratio, the more gain reduction you will get.
In the right-hand panel there are some controls we haven’t looked at.
At the bottom is the Dry/Wet control. This lets you adjust how much of the Uncompressed (Dry) and Compressed (Wet) signal that you hear. You can use this to blend the Dry and Wet sound, and this is usually called “Parallel Compression” or “New York Compression”.
In the middle there is the Range control. This lets you control the maximum amount of gain reduction that is applied to the signal. This means you can restrict the compression that’s applied despite setting a low threshold level and/or a high ratio.
At the top of this section is the Soft clip control. This used to radically compress very loud transients (like very loud drum beats) and is like a kind-of safety control to help with this type of sound with very loud short bursts.
Ableton Glue Compressor Is Modelled on a Hardware Mix Bus Compressor
The Ableton documents say that the Glue Compressor is based on the classic bus compressor from a 1980s mixing console. It is designed mainly for use on the Master Track or Group Track of a mix to “glue multiple sources together into a coherent sounding mix”.
So here’s where the “glue” function comes in. It’s designed to make different sounds, like vocals and various instruments, sound as if they belong together in the audio mix.
Glue Compression Relates to Compressing a Blend of Sounds
Glue compression then, is where compression is applied to a mix of several different sounds.
This could be the entire mix of the production, or a group of tracks where you want the sounds to blend together. This is the type of thing producers mean when they talk about audio elements “sitting nicely in the mix”.
This is in contrast to compressing an individual sound, like a guitar or vocal. However, attempting to make a sound like a vocal “sit in the mix” rather than “sitting on top” of the other instruments can be achieved by compressing that individual sound too.
You can see the way the bus compressor is located on the SSL hardware mixing desk in the image below (highlighted in a white square on the right-hand side.
Either the entire mix can be routed through it, or just a selection of channels in the mixer. Either way, the compression is being applied to a mix of tracks, glueing them together.
Requirements for a Bus Compressor to Glue a Mix
There are different types of compressor, that work in different ways, which produce different sound characteristics.
Some compressors impart their own sound on the audio in terms of things like saturation. Others might have a very fast attack built in, which results in its own characteristic type of sound.
One of the main features required for a bus compressor seems to be “transparency”. At the stage when you are applying compression to a mix you probably don’t want to colour the sound any more than necessary.
Compression and other effects will have been applied to the individual sounds already, and now you are just trying to get them to sound like they belong together.
Also, the amount of compression applied with this type of bus compressor is usually fairly light. This means a fairly high threshold, so that only the highest peaks will be compressed, and a fairly low ratio so that those peaks aren’t being compressed very much.
Dynamic Range Reduction Helps to Glue the Audio Elements Together
But what is it that enables the bus compressor to provide this “glue” that helps to blend all the different sounds together?
It appears that the glue is produced by the same dynamic range reduction being applied to all of the sounds in a group track, or even to the whole mix, whichever way it’s being done.
We’ll focus on applying bus compression to the whole mix, but the principles are the same for “glue” compression applied to a smaller collection of sounds in group track or sub-mix.
The thing that makes bus compression, or glue compression, work is that every element in the mix is passed through the compressor.
The aim here isn’t really to make the mix louder overall, as described above, but to make the individual parts sound as if they belong together. That’s why this type of compression is sometimes called glue compression.
Differences in Dynamic Range Differentiate Different Sounds
A guitar, piano, bass, synth, vocal and any other sound source sounds different because of the timbre, or quality of the sound. However, in addition to that the dynamic range (difference between the loud parts and quiet parts of the sound) help to differentiate the sound.
The type of bus compression we’re looking at here doesn’t really change the timbre of the sound in the way that applying EQ or low-pass or high-pass filter might. However, applying the same compression to all of them affects the dynamic range of each element in a similar way, which makes them sound as if they belong together.
Apparently, this is because dynamic range is one of the ways that we can distinguish the various different sounds in a mix. Reducing dynamic range reduces the difference between them, and this is what is meant by “glueing a mix together”.
This is similar to the way that compressing a vocal or lead instrument reduces the dynamic range and makes it sound as if it’s “sitting in the mix”, rather than “sitting on top of it”.
Glue Compressor – Just a Regular Compressor Applied to a Number of Audio Tracks
So, a glue compressor isn’t really a special type of audio compressor. It’s just a regular compressor that is used to collectively compress a number of audio tracks to make them sound like they belong together.
This glue compression is sometimes referred to as “bus compression”, since the mix bus is routed through the compressor.
In addition, people often think of the bus compressors built into large hardware studio mixing desks, which could make you think that only one type of compressor can be used for this purpose. However, various types of compressor are used for this purpose.
Something I should add here is that the compressor settings (and the type of compressor used) should ensure that no (or very little) additional processing is added to the sound. The glue compression really needs to be transparent, except for the small reduction in dynamic range to help the sounds blend together.