ABOUT COMPRESSION

In nearly all recorded music that you listen to, compression is a very essential part! Not only is it applied to almost every instrument you hear in a commercial recording, it's often even applied to the entire recording (or the sum of all instruments together)! So, a compressor is a very powerful and necessary tool in music processing.

Essentially, what a compressor does is it limits the dynamic range of an audio signal. 'Dynamic Range' refers to the difference between the loudest point and the quietest point in a signal. That loosely means that if a compressor is applied to a signal that has some loud parts and some quiet parts, the compressor will make the loud parts be a more consistent volume with the quiet parts.

Here's a breakdown on our compressor and most others work:

Our compressor has a user-set threshold that is essentially a volume at which the compressor will begin to act. Once the signal goes above that threshold and as long as it is above the threshold, it will work to reduce the volume of the signal in what is referred to as 'gain reduction.'
Our compressor also has a user set ratio that controls how much work it will do to reduce the volume of the signal when it is above the threshold. It is called a ratio because it actually is a ratio of the volume of the input signal to the volume of the output signal. For example, if a user sets a ratio of 4:1, if an input signal goes above the threshold by 8 decibels, the compressor will reduce its volume to 2 decibels above the threshold.
The user may set an attack time, which is how long it takes the compressor to reach the desired ratio of gain reduction once a signal goes above the threshold. This can be thought of as how long it takes the compressor to 'clamp down' on the signal.
Lastly, the user may set a release time, which is how long it takes the compressor to stop gain reduction once the signal goes back below the threshold after being above it.

The following graphics should help explain each setting of a compressor:

This graphic uses a square pulse as an input signal. The ratio of the compressor is evidently 2:1.

Uses of Compression Settings

As mentioned before, compressors are used to affect almost every instrument in a commercial music recording. This is because the human ear favors sounds that are more consistent and evenly leveled. In addition, it is easier for several instruments to blend together if they all have consistent volume levels. Though, in using compression on any musical instrument, it is important to tailor the compressor's settings (threshold, ratio, attack, release) to the qualities of the instrument to get the most favorable result. Here are some examples:

Ratio and threshold can be used together to control how much compression is applied to the signal. It's important not to apply too much, because artifacts of compression may become audible and distracting.
The attack time can actually be used to make an audio signal sound more punchy. A longer attack time means the compressor will take longer to 'clamp down' on the signal, which means that an initial transient (a loud, short peak in a signal) can get through without being reduced in volume as much. This is in contrast to a short attack time, which would reduce the transients of a signal much more. Shorter attack times are much more favorable to use when compressing vocal signals, since that will ensure that any harsh consonants emitted by the singer (an unfavorable "P" or "T") will not get overemphasized. Longer attack times are more useful for bass guitar or percussive instruments, since these instruments by nature are transient-heavy and it is often necessary for them to be punchy to keep the pulse, or the groove, going in a commercial recording. A compressor with a long attack time is likely what made the drums on albums like 'American Idiot' by Green Day sound so big and punchy!
It is far harder to identify appropriate settings for the release time in general, as it depends more on the nature of the instrument in question. Loosely, for transient-heavy instruments like percussion, a shorter release time is often more favorable to avoid compressing each transient unevenly. For more melodic, sustained instruments like violin, a longer release is often desirable so that the sound of the compressor stopping gain reduction is less detectable.

Image courtesy of Iain Fergusson in his article here:

http://en.wikipedia.org/wiki/Dynamic_range_compression

Image courtesy of Iain Fergusson in his article here:

http://en.wikipedia.org/wiki/Dynamic_range_compression