Back to blog
ReferenceMay 9, 202613 min read

How to Use a Compressor: Threshold, Ratio, Attack, Release Explained With Real Examples

Compression is the most misunderstood tool in mixing. Beginners use it to "make things louder" without understanding what it actually does, end up squashing the life out of every track, and wonder why their mixes feel small. Understanding what each compressor control actually does — and what specific musical problem it solves — is the difference between compression that helps and compression that hurts.

Compression is automatic gain reduction triggered by a level threshold. When a signal goes above the threshold, the compressor turns down the gain. When the signal drops below the threshold, the compressor lets the gain return to normal. The result: loud parts get quieter, the dynamic range narrows, and the average level can be raised without clipping the peaks. That is the entire mechanism. Everything else is a parameter that controls how the gain reduction behaves over time.

Threshold sets where the compressor starts working. Anything above threshold gets compressed; anything below it passes through unchanged. Set the threshold by listening to the gain reduction meter — you want to see the meter dip down on the loudest parts of the performance and return to zero between them. If the meter is constantly active, your threshold is too low. If the meter never moves, your threshold is too high.

Ratio sets how aggressively the compressor reduces gain when the signal exceeds threshold. A ratio of 2:1 means that for every 2 dB the signal goes above threshold, only 1 dB comes out — a gentle reduction. A ratio of 10:1 means that for every 10 dB above threshold, only 1 dB comes out — aggressive limiting. A ratio of ∞:1 (infinity to one) means nothing exceeds the threshold ever — that is a brick-wall limiter.

Attack sets how fast the compressor reacts after the signal crosses threshold. Fast attack (0.1 to 5 ms) catches transients aggressively, killing snap and punch. Slow attack (10 to 50 ms) lets transients through and only compresses the body of the sound. The most common amateur mistake is using fast attack on percussive instruments (kick, snare, plucked bass), which kills the transient and leaves a flat, dead sound.

Release sets how fast the compressor lets go after the signal drops below threshold. Fast release (10 to 100 ms) means the gain returns to normal quickly, which can pump audibly with the rhythm. Slow release (300 ms to 1+ second) means the gain returns slowly, providing smooth, transparent leveling. Auto-release modes are program-dependent and usually a safe starting point.

Vocal compression: ratio 3:1 to 4:1, attack 10 to 30 ms, release set so the meter pumps with the syllables, 4 to 6 dB of gain reduction on the loudest words. The 1176 (FET) and LA-2A (opto) emulations are the workhorses. Use the LA-2A for slow musical leveling, the 1176 for fast peak control. Stack both for serial compression — the LA-2A sets the average level, the 1176 catches the peaks.

Kick compression: ratio 4:1, attack 5 to 10 ms (let the click through), release 100 to 200 ms, 4 to 6 dB of gain reduction. The slow-ish attack preserves the beater click; the fast-ish release lets the kick recover before the next hit.

Snare compression: ratio 4:1 to 6:1, attack 5 to 15 ms, release 100 to 200 ms, 4 to 8 dB of gain reduction. Same principle as kick — attack slow enough to let the crack through, release fast enough to recover.

Bass compression: ratio 3:1, attack 10 to 30 ms, release set by ear so it pumps with the song, 3 to 6 dB of gain reduction. The slow attack preserves the pluck transient; the goal is even sustain, not punch reduction.

Acoustic guitar compression: ratio 2:1, attack 30 to 50 ms (preserve the pick), release auto, 2 to 4 dB of gain reduction. Light, transparent, just to even out the dynamics between picking and strumming.

Bus compression: ratio 2:1, attack 30 to 100 ms (very slow), release auto or 0.3 to 1 second, 1 to 3 dB of gain reduction. The goal of bus compression is glue — the entire bus moves slightly together, creating cohesion. Heavy bus compression sounds amateur; light bus compression sounds expensive.

Parallel compression: send a signal to an aux, slam it with extreme compression (ratio 10:1, fast attack, fast release, 10+ dB gain reduction), and blend the crushed copy in under the dry signal. Result: punchy and present without losing the transients. This is how every modern drum mix gets its impact.

Common mistakes: too much gain reduction (kills the life), fast attack on percussive instruments (kills the punch), too low threshold (compressor working constantly), wrong release (audible pumping when you don't want it), trying to fix a bad recording with heavy compression (compression amplifies everything bad in the source).

Pull up a vocal track. Apply a compressor with 3:1 ratio, 20 ms attack, auto release. Pull the threshold down until the meter shows 4 dB of gain reduction on the loudest words. Listen to the difference. That is the foundational vocal compression move — and from there you can adapt everything else.

Start Free