Propellerhead Record: The Channel Strip Compressor, Demystified

Line 6's Propellerhead Product Specialist Matt Piper talks about the Channel Strip Compressor included on each audio channel in Propellerhead Record, and explains in plain English some fundamental concepts of audio compression.

What is compression?

A relatively large number of people who make music actually have a rather vague idea about what compression is. Is the compressor the "good" button that makes everything sound better? Warmer? Louder? Smoother? Is it like a delicious barbeque sauce that you can put on everything, the more the better? Not really! In fact, if you are using a lot of compression in your mix, from time to time it's a good practice to take all the compression off and check the mix that way. It may suddenly sound like you've taken cotton balls out of your ears and your mix has come back to life! Yes, you can actually choke the life out of your mix with too much compression.

In audio signal processing, a compressor reduces the difference between the loudest and softest sounds in a mix. In other words, it decreases (compresses) the dynamic range. With fast attack times (the time it takes the compressor to reduce the signal level), it can soften the attack of drums and other instruments. Compression can be used more or less transparently as a utility to remove peaks in the audio signal that might otherwise cause "clipping" (unwanted distortion due to overload), or it can be used as a noticeably audible effect to achieve a desired sonic outcome.

Threshold and Ratio

The Threshold setting  is measured in decibels (dB). When you set this knob, you are telling the compressor to reduce the level (or gain) of any signals that are louder than the level set with the Threshold knob. You are choosing the incoming signal level necessary to activate compression.

The Ratio setting determines how much compression will be applied to signals that exceed the Threshold setting. A ratio of 2:1 is relatively light compression, while 8:1 is pretty heavy. Beyond 10:1, compression is often referred to as "limiting." The Channel Strip Compressor's Ratio knob goes up to 16,129:1 before it actually reads Infinite:1!

What does the Peak button do?

Just like the famous SSL 9000K SuperAnalogue Mixing Desk on which it is based, the compressor on each channel of the mixer included with Propellerhead Record has a Peak button. The Peak button changes the detection type used by the compressor. If the Peak button is off, the compressor uses its RMS detection mode. Are your eyes rolling back in your head yet? No worries: I'll explain.

As I mentioned in the Threshold and Ratio section, when you set the Threshold knob, you are choosing how loud an incoming signal must be in order to trigger the compressor. The detection mode is how the compressor detects the level of those incoming signals.

In RMS mode (Peak button off), an average of several milliseconds of audio signal is used to calculate the incoming signal level. In Peak mode, the detection is instantaneous. While RMS mode relates more closely to how our ears and brains perceive loudness, it can completely ignore very loud but very short noises. The attack of a snare drum or kick drum may be very loud, but it may also be of such a short duration that when averaged with the rest of the sample period of the RMS detection, the compressor is not activated at all, even though the Threshold may have been exceeded for a fraction of a second. RMS mode can be good for applying more transparent or "natural" compression for loud sounds of longer duration, but for controlling peaks with sharp sounds such as drum attacks, Peak mode is better. To see the difference between how these work, try compressing both ways and watch the Gain Reduction LEDs on the right of the compressor.

What does the Fast button do?

You may notice that the Channel Strip Compressor has no Attack knob. By default, this compressor has an automatic adaptive attack time, based on the audio program material and available headroom. When you engage the Fast button, the Attack time is fixed at 3 milliseconds for 20dB of gain reduction. This is good for clamping down on peaks as quickly as possible. However, like the Peak button, when using the Fast button, you may noticeably change the sound of the signal you are processing, especially at high Ratio settings.

If you need to use some really strong, fast compression, but you notice you are losing some of the character of the original sound in the process, you may want to check into using Parallel Compression. To do this, you will use two copies of the same audio signal. On one channel, use the compression as heavily as you want, and on the other, use no compression at all. The idea is to mix in some of the unprocessed signal to re-introduce some of the transients (content in the initial attack of a sound) that are lost through heavy, fast compression.

Setting the Release time

The Release setting determines how much time it takes the compressor to "let go" of the signal once it has fallen back down below the Threshold setting. On Record's Channel Strip Compressor, this setting is variable between 100ms and 1 second. For a snare drum or kick drum, I often set the Release time to its lowest (fastest) value of 100ms. That way, I control the attack of the drum (usually in Peak mode, maybe with the Fast button activated as well), but then the compressor "lets go" quickly and allows the rest of the drum sound to ring out without having its gain reduced by the compressor. For many other types of sounds, more gradual release times may sound more natural in the mix.

Where is the Make-up Gain or Output Level knob?

Many compressors have output level knobs or make-up gain knobs (essentially the same thing). Since the overall signal level is reduced by compression, and output level or make-up gain control is necessary to bring the signal level back up to an appropriate level when it exits the compressor. You are still decreasing or compressing the dynamic range (the difference between the loudest and softest sounds), but you are afterward increasing the overall output level to something that's easy to mix with.

The Channel Strip Compressor does not have an output level or make-up gain control because it does automatic make-up gain behind the scenes. After the compressor reduces the gain of signals louder than the Threshold setting, automatic make-up gain raises the overall volume of the compressed signal to an appropriate level, similar in perceived loudness to the original signal. (However, signals compressed with extremely high Ratio settings and very low Threshold settings may exit the compressor rather quietly in spite of the automatic make-up gain.)

Ok, what is this Key button all about?

The Key button is automatically activated whenever something is connected to the Sidechain input. The way Sidechain works is: An audio output from some other source is connected to the Sidechain input. The audio on the original channel is still compressed, but the compressor will be triggered by the external source instead of being triggered by the audio on its own channel.

Sidechaining can be used to "cut a hole" in the mix to make room for another sound. For instance: If you have a legato synth bass line playing on the compressed channel, and connect the output of a kick drum to the sidechain input of the synth bass channel, the synth bass will have its gain reduced each time the kick drum strikes. The bass will "pump" (inversely) along with the kick drum. By the way, if you do have something connected to the Sidechain input, but do not want the compression to be externally triggered any more, simply turn off the Key button.

There is far more to learn about the artistic and effective use of audio compression than can be contained in this blog post. I encourage you to research, experiment, and use your ears. And most of all: Enjoy making music!