By Jeff Strong

The compressor’s job is to compress the dynamic range of the sound being affected. The compressor not only limits how loud a note can be, but also reduces the difference between the loudest and softest notes (it compresses the dynamic range).

The limiter works much like the compressor except it limits the highest level of a sound source. Any signal above the threshold is chopped off rather than compressed, as with the compressor.

Compressors/limiters are used for the following three main purposes (although other purposes certainly exist as well):

  • To keep transients from creating digital distortion during tracking: This is common with drums that have a very fast attack (initial signal) that can easily overload the recorder (or converters or preamps).

  • To even out a performance that shows a high degree of unwanted dynamic variation: You do this during either the mixing or tracking stage. An example is a singer who has poor mic control (moves constantly in front of the mic), and as a result, some recorded passages become loud while others are very quiet.

  • To raise the overall apparent level of the music during mastering: For example, listen to a CD recorded ten years ago and one from the last year or so, and you’ll notice that the newer CD sounds louder than the older one.

The purpose of the various parameters of a compressor are:

  • Threshold: The threshold setting dictates the level that the compressor starts to act on the signal. This is listed in dB (decibels). This setting is often listed as dB below peak (0dB). In other words, a setting of –6dB means that the compressor starts to act when the signal is 6 decibels below its calibrated 0dB mark. (In digital systems, 0dB is the highest level a signal can go before clipping.)

  • Ratio: The ratio is the amount that the compressor affects the signal. The ratio, such as 2:1 for instance, means that for every decibel that the signal goes over the threshold setting, the signal is reduced by a factor of 2. In other words, if a signal goes 1dB over the threshold setting, its output from the compressor is only 1/2dB louder.

    With ratios above 10:1, your compressor starts to act like a limiter.

  • Attack: The attack knob controls how soon the compressor kicks in. The attack is defined in milliseconds (ms), and the lower the number, the faster the attack.

  • Release: The release parameter controls how long the compressor continues affecting the signal after it has dropped back below the threshold setting. Like the attack, the release is defined in milliseconds. The lower the number, the faster the release time.

  • Gain: The gain knob allows you to adjust the level (volume) of the signal going out of the compressor. This is listed in decibels. Because adding compression generally reduces the overall level of the sound, you use this control to raise the level back to where it was going in.

  • Hard knee or soft knee: Most compressors give you the option of choosing between a hard knee and a soft knee (or they do it for you based on the setting that you’ve chosen). Hard knee and soft knee refer to how the compressor behaves as the input signal passes the threshold. They are defined as follows:

  • Hard knee applies the compression at an even rate regardless of the level present over the threshold. So if you choose a compression setting of 4:1, the compressor applies this ratio for any signal over the threshold limit. Hard-knee compression is used for instruments like drums, where you need to clamp down on any transients quickly.

  • Soft knee, by contrast, applies the compression at a varying rate, depending on the amount the signal is over the threshold setting. A soft knee setting gradually increases the ratio of the compression as the signal crosses the threshold until it hits the level that you set. Soft-knee compression is used on vocals and other instruments where the signal doesn’t have fast peaks.

Some two-channel compressors have a link function that allows you to connect the two separate channels and control them from one set of controls. Having the two channels linked ensures that you end up with the same settings on both channels. This is useful when you want to compress a stereo signal.