New plugin

Airchain

FM radio airchain emulation — AGC, four-band squash, emphasis, clipper and transmission artifacts behind one big SQUASH knob.

VST3 / AUmacOS14 presets
Discover Airchain
support@squaresmile.de Köln, Germany
All guides

Help Center · SoundSense

Match games: compressor, delay, reverb, saturation, gate

Five games where you turn stepped knobs until your version sounds like a hidden target — how the levels grow, how matching by ear is scored, and what the processors are.

Open SoundSense

On this page
  1. How a round works
  2. Compressor match
  3. Delay time match
  4. Reverb decay match
  5. Saturation match
  6. Gate match
  7. On a phone speaker

In a match game you don't name a setting — you rebuild it. A stem plays through a hidden setting (the Target). You turn knobs until Yours sounds the same, then press Check my match. The five games are Compressor match, Delay time match, Reverb decay match, Saturation match and Gate match. All five sit in the Mix Engineering row and run on the eight-level ladder (see The level ladder and pre-listen).

How a round works

  1. The switch Original / Target / Yours picks what you hear. The stem loops; switching keeps the place in the music, without clicks.
  2. Turn the knobs. The knobs move in steps; greyed knobs are fixed at this level. When you let go of a knob, Yours is rendered again with the new setting.
  3. Press Check my match. The result shows how far apart the two sound, the target setting and, where there is one, the picture of both. The switch stays, so you can compare Target and Yours once more.

Every version plays as loud as the original. A stronger setting is never heard as "louder", so you listen to the movement, the time or the grit, not the level.

Scored by ear, not by knob position. Each game compares what the two versions do to the sound. Two settings that sound alike score alike, even when the knobs differ.

The picture fades. On levels 1–2 a picture shows Target and Yours, on levels 3–4 only Yours, from level 5 nothing. Points with a picture count ×0.5 (both) or ×0.85 (Yours only), and only answers without one count as exact — see The level ladder and pre-listen.

These games run on our own models of a compressor, a delay, a reverb, a saturator and a gate, written for SoundSense. The same model makes the sound and the score. They are not emulations of any product.

Compressor match

The knobs grow with the level:

Level You set The rest is fixed at
1–2 Threshold (−2 to −18 dB, in 2 dB steps) ratio 4:1, attack 10 ms, release 120 ms
3–4 Attack (0.3–100 ms) and Release (30–1000 ms) threshold set for about 7 dB of gain reduction on this stem
5–6 Threshold and Ratio (1.5:1 to 10:1) attack 10 ms, release 120 ms
7–8 all four —

Odd levels draw the target from the clearer settings, even levels from the whole range. Threshold is relative to the clip's average level, so the same setting squeezes every stem alike. Makeup gain is automatic.

  • The model: a feed-forward compressor with an RMS detector, a 6 dB soft knee and attack and release in dB.
  • The picture: the gain-reduction line of each version over the first seconds of the loop.
  • Points: how alike the two gain movements sound, weighted by the music (a difference during silence doesn't count) — within 0.5 dB 100, 1 dB 80, 2 dB 50.

Delay time match

Set the delay time until your echoes land where the target's do. Feedback (35 %) and mix are fixed; only the time changes. Range 80–700 ms.

  • Levels 1–2: note values at a tempo the prompt shows (90–130 BPM), such as 1/8, dotted 1/8 or 1/4. The picture shows where the echoes of one hit land.
  • Levels 3–4: milliseconds in steps of about 10 %.
  • Levels 5–6: steps of about 5 %. Levels 7–8: about 2.5 %.
  • Points: by how far the times differ — within 3 % 100, 7 % 80, 15 % 50. 10 % too long and 10 % too short score alike.

Each repeat is a little darker than the last, through a low-pass in the feedback.

Reverb decay match

Set the decay time (RT60, the time the tail takes to fall by 60 dB) until your tail dies away like the target's. Pre-delay and mix are fixed, and the reverb is just as loud at every decay. Range 0.4–4 s.

  • Steps: about 30 % on levels 1–2, 22 % on 3–4, 16 % on 5–6 and 12 % on 7–8.
  • The model: an eight-line feedback delay network that decays at the set RT60 at every frequency.
  • The picture: the decay to −60 dB of each version.
  • Points: within 8 % 100, 16 % 80, 32 % 50.

Listen to the gaps between notes: how long the tail is still there.

Saturation match

Set the drive until Yours grits like the target. The output is matched in loudness automatically, so listen to the grit, not the level. Drive is relative to the clip's average level, 0–30 dB.

  • Steps: 6 dB on levels 1–2, 4 dB on 3–4, 3 dB on 5–6 and 2 dB on 7–8.
  • The model: a soft clip that bends one half of the wave a little earlier than the other, so it adds even harmonics as well as odd ones.
  • The picture: the transfer curve — input across, output up.
  • Points: by how much distortion each version adds — within 1.5 dB 100, 3.5 dB 80, 7 dB 50. High drives distort more alike than low ones, and score so.

Gate match

Set the gate until it chops the quiet parts like the target.

Level You set
1–2 Threshold
3–4 Release (10–800 ms); the threshold is held where the gate is closed about a third of the time
5–6 Threshold and Release
7–8 Threshold, Release and Range

Threshold is relative to the clip's loudest peak (−6 to −42 dB). The gate opens at the threshold and closes 4 dB below it, after a 20 ms hold.

  • The picture: the gate's gain over time, for each version.
  • Points: how alike the two level movements sound while a gate works — within 1.5 dB 100, 4 dB 80, 8 dB 50.

Listen to the ends of the notes: what is cut off, and how fast.

On a phone speaker

All five games work in phone-speaker mode, in mono and without bass stems. Compressor match also leaves out stems whose energy sits mostly below 400 Hz, where a phone speaker would hide the pumping. See Listening devices and phone-speaker mode.

Was this page helpful?