Compression - ectoplasma

Set threshold, ratio, attack, and release, then switch between transparent control and audible pumping.

Sound has a dynamic range - the distance between its quietest and loudest moments. A whisper and a shout in the same sentence. A snare crack next to a sustained pad. A compressor narrows that gap. It watches the signal level and turns down the loud parts automatically, hundreds of times per second, faster than any hand on a fader could manage.

The result is a more even signal. The quiet details become audible. The loud peaks stop overwhelming everything else. This is why every recorded song you have ever heard uses compression somewhere in its chain. Without it, recordings sound jagged and unpredictable through speakers.

The machine below gives you the four core controls of a compressor. Press play, then move each slider and listen to what changes.

What compression looks like

Follow the three stages left to right. The original has wild swings. Compression squashes the peaks down - the signal gets quieter and flatter. Then make-up gain turns everything back up, filling the headroom. The result is louder than the original without clipping.

1. Uncompressed

2. Compressed (quieter)

3. + Make-up gain (louder)

The four controls

Threshold sets the level above which compression begins. Any signal louder than the threshold gets reduced. Pull it low and the compressor catches almost everything. Push it high and only the loudest peaks get touched.

Ratio controls how much reduction happens. A ratio of 4:1 means that for every 4 dB the signal goes above the threshold, only 1 dB comes out. At 1:1 there is no compression at all. At 20:1 you are close to a limiter - a brick wall that lets almost nothing through above the threshold line.

The diagram above shows how a compressor maps input level to output level. Below the threshold, it is a straight diagonal - what goes in comes out unchanged. Above the threshold, the slope flattens. The steeper the original signal, the more it gets squashed down.

Attack is how quickly the compressor reacts once the signal crosses the threshold. A slow attack (50 ms or more) lets the initial transient through before clamping down. This keeps drums punchy and percussive, because you hear the sharp click of the stick before the compressor grabs the sustain. A fast attack (under 10 ms) catches the transient itself, producing a more controlled, even sound.

Release is how long the compressor takes to let go once the signal drops below the threshold. Short release means it snaps back quickly. Long release means it holds on, smoothing out the signal over time.

Make-up gain: where loudness comes from

A compressor only turns things down. After compression, the signal is dramatically quieter. The peaks have been squashed, the overall level has dropped. On its own, compression makes things sound worse.

But the dynamic range is now squished flat. The peaks that used to stick out above everything else have been pushed down level with the rest. That means there is empty space above the waveform - headroom - that wasn't there before.

Make-up gain fills that space. You turn the entire signal back up. Because the peaks no longer stick out, you can push the volume far higher than you could before compression - without clipping. Everything comes up together. The quiet parts, the medium parts, and the squashed peaks all rise as one. The average level is now much higher than the original signal. That is perceived loudness.

This is the entire basis of the loudness war in commercial music. Compress harder, make up more gain, and the track sounds louder on the radio. The cost is dynamic range - the music stops breathing.

Pumping

Try this: Toggle Pumping Mode and listen. When the attack and release create a rhythmic breathing pattern - the volume ducking and surging in time - that is called pumping. Daft Punk built entire tracks around this effect. In sidechain compression (lesson 16), the pumping is triggered by a kick drum, so the rest of the mix ducks on every beat. It sounds like the music is breathing.

Why ASR, not ADSR?

The pad in this machine uses an ASR envelope (attack, sustain, release) with no decay stage. A pad that holds a steady tone does not need the decay dip. ASR is simpler: the sound fades in, stays at full level, then fades out. No peak, no dip. A continuous drone.

This makes the compressor's effect clearer - any volume movement you hear is the compressor, not the envelope.

Transparent vs aggressive

Compression is one of the most transparent tools in audio. When it is done well, you cannot hear it working at all - you just notice the mix sounds "finished" and solid. When it is done aggressively, you hear pumping, breathing, or a squashed, airless quality.

The loudness war: from the 1990s onward, mastering engineers compressed entire albums harder and harder to make them sound louder on the radio. The dynamic range shrank from around 12 dB to as little as 4 dB on some releases. Modern streaming platforms now normalise volume automatically, so the loudness war has mostly ended - but its flattened recordings are still everywhere.

Exercises:
1. Set ratio to 1:1. Sweep the threshold. Nothing changes - because the compressor is not reducing anything.
2. Set ratio to 8:1 and threshold to -24 dB. Listen to how the peaks are tamed.
3. Set attack to 1 ms and release to 150 ms. Toggle Pumping Mode. Hear the rhythmic breathing.
4. Try a very long release (600 ms+). The compressor holds on and the sound feels squashed flat.