Compressors are gain-riding devices that monitor the level of the incoming signal and then apply gain reduction in accordance with the user's control settings. Though this explanation is indeed simple, there are many types of hardware and software compressors available, and no two units will sound the same even with near-identical settings. Though the controls vary depending on the particular unit. Most compressors have controls for input level, output level, threshold, compression ratio, attack and release.
The input level and threshold control the level of incoming sound to be processed. The threshold controls the level at which attenuation will begin, and thus controls the sensitivity of the compressor. Since lowering the threshold and raising the input level essentially have the same effect on responsiveness, some compressors will not feature input controls.
Compression ratio determines the amount of attenuation to be applied to an incoming signal. Most units feature ratios as high as 8:1 or 10:1 and as low as 1:1 or sometimes even lower (ratios lower than 1:1 would cause the signal to be boosted instead of attenuated in most cases). The compression ratio is a constant value, as the ratio of the input to output change is always the same. Thus a ratio of 8:1 would mean that it would take an increase of 8 dB coming into the compressor to cause the output to only increase by 1 dB. Ratios at or beyond 10:1 are considered limiting as opposed to compression, in that the attenuation is so extreme that the signal is essentially “limited” from passing beyond a certain level.
Attack controls the time which it takes for the attenuation to begin once the input level surpasses the threshold. This can range from 1 millisecond to over a second depending on the unit. Fast attack times can keep transients from causing clipping, though they can also kill the impact and punch of a sound if used to excess.
Release controls the time which the compressor takes to end attenuation after the signal falls below the threshold. Fast release times are effective in increasing loudness though they can be destructive and cause much unwanted distortion if abused.
Most compressors are of either the Tube/Valve, FET, VCA, or Optical varieties, with each referring to the inherent gain control element.Traditional tube/valve compressors involve multiple stages of vacuum-tube audio amplification which provides what many describe as a “warm” sound with a slight roll-off in the lower frequencies. Many of the most expensive and sought after compressors are tube/valve compressors.
FET or Field Effect Transistor compressors were the first transistor to emulate tubes in the way they worked internally. Inherently a high-impedance device, the FET compressor sounds highly unique and can be extremely expensive to produce. FET compressors have fast reactions to input signals, and processing is achieved with little distortion. FET compressors are often used on vocals as they allow the engineer to achieve high levels of clean-sounding compression. The downside is FETs have a somewhat limited dynamic range, and very hot levels can cause unwanted pumping and other amplitude-modulation artifacts.
VCA or Voltage Controlled Amplifier compressors are likely the most flexible compressors. They can quickly change gain in response to multiple detectors looking at the same input signal. VCAs are typically associated with fast, clean compression with precise controls over attack and release. They are extremely popular due to their versatility and cost to manufacture.
Optical compressors contain a light source that gets brighter as your signal gets louder, and a light-sensitive resistor that reacts to the brightness of the light by decreasing its resistance. The resistor works in either a feedback loop or a voltage divider to change the level of your signal. In some older designs these were separate components, but in most modern designs the light and the resistor integrated. Optical compressors are smooth in response, and have slower release times, giving them a very natural sound.
It is important to understand that compression destroys many desirable sonic qualities of an audio signal. Many novice producers assume that everything must be compressed at least a little bit in order to achieve a professional sounding mix. This could not be further from the truth. Keeping this in mind, use only as much compression as is necessary and rely on your ears not your eyes. One should never attempt to compress or alter a waveform merely because it is uneven looking or not uniform.
Remember, flat rectangular waveforms are essentially just on their way to becoming white noise, so use compression sparingly, with moderate attack and release times, higher thresholds and low ratios. This will assure you do not end up with a distorted and unnatural sounding mix when all is said and done.