A Beginner's guide to Compression*Compressors are probably the most widely used signal processors in the audio industry.
A compressor can be thought of as an automatic volume control. Once the volume of
the signal exceeds a certain level (called the 'threshold'), the compressor reduces the
gain (in other words, 'turns the volume down'), causing the signal to be less loud than
it would otherwise have been.
The amount by which the compressor reduces the gain is determined by the 'ratio'.
The ratio is conventionally expressed as a numerical value, e.g. '4:1', which represents
the amount by which the gain is reduced when the volume of the signal rises above
the threshold.
Let's take an example with some real numbers. If the threshold is set to -10 dB and the
ratio is set to 4:1, any signal whose level exceeds -10 dB needs to rise in level by 4 dB
for the output of the compressor to rise by 1 dB. Therefore an input signal with a peak
at -6 dB (which is 4 dB above the threshold) would emerge from the compressor with
a peak at -9 dB (1 dB above the threshold). Signal levels below the threshold are
unaffected, so if the signal in the above example varied between -20 dB and -6 dB
before entering the compressor, it will vary between -20 and -9 dB after being
compressed. Its dynamic range (the difference between the quietest and loudest parts of
the signal in dB) is reduced from 14 dB to 11 dB.
Compression results in any variations in the volume of the signal (in other words, the
signal's dynamic range) being reduced - the amount of this reduction is determined by
the threshold (the level above which the gain is reduced) and the ratio (the amount by
which the gain is reduced.) Higher ratios are referred to as hard ratios; lower ratios are
called soft ratios.
Because compression causes a reduction in volume level of loud signals, gain must be
applied after the compressor to bring the overall volume level back up, so that the
maximum volume before the compressor is the same as that after the compressor. This
is called 'make-up gain', and is necessary so that the maximum level of the signal is
always the same, for correct level matching with any further processing or other
equipment.
Once 'make-up gain' has been applied, the part of the signal that was lower than the
threshold volume (and hence not compressed) will now be louder than it was before
the compressor. This will cause any compressed instrument to sound louder. One use
for this phenomenon is to give guitars more sustain.
In most pop music, the backing instruments (such as drums, bass guitars, rhythm
guitars etc) tend to be compressed heavily (using a fairly hard ratio and low threshold),
so that they remain at a consistent volume level throughout the track. This will
provide a solid backing, without occasional drum hits or bass notes poking through (or
disappearing from) the mix untidily.
A soft ratio tends to be used on instruments such as lead guitars or vocals that 'sit' on
top of the mix. In this situation it is often desirable to preserve more of the dynamics
of the original performance, to retain more expression. A reduction in variation of
volume level is still required (for the reasons mentioned above), but not to the same
extent.
The other controls included on most compressors are attack and release.
Attack determines the speed at which the compressor starts to reduce the gain once the
threshold has been exceeded. Think of it as the time taken to turn the volume down.
Very short attack times mean the compressor ‘kicks in’ very quickly – short attack
times are typically used for vocals in order to keep the levels under strict control.
Longer attack times mean more of the original signal’s attack dynamics are preserved –
this is a good way of keeping percussive and guitar sounds exciting and punchy.
Release determines the speed at which the compressor stops acting once the signal
drops below the threshold. Think of it as the time taken to turn the volume back up.
Short release times mean the compressor very quickly returns the signal to its normal
level. This can produce a ‘pumping’ sound, where the changes in volume are very
audible. Depending on the style of music, this can be undesirable, or a useful creative
effect.
Longer release times may mean that parts of the signal below the threshold end up
being compressed, or that the gain doesn’t have a chance to return to normal before
the next ‘above threshold’ sound – remember that the compressor works on the whole
signal. See the diagram below:
*Quoted from the Focusrite manual
Edited by user 15 years ago
| Reason: Not specified
Hai Lúa attached the following image(s):
