Delay and reverb: the space around the sound
Both effects are made from copies of the sound, played back late. A delay's copies are few enough to count; a reverb's are thousands, fused into one tail.
Everything up to this point shaped the sound itself. This lesson shapes the space around it. Delay and reverb are built from the same raw material: copies of the input signal, played back after a gap. The only real difference between them is density. A delay's copies arrive far enough apart that your ear counts them one by one. A reverb's copies arrive so close together that they fuse into a single wash. Audio engineers file both under time-based effects, because they work on when a sound repeats rather than on its tone or its level.
One control matters more than any other on a time-based effect: wet. Wet sets the volume of the effect's own output, mixed in alongside the untouched dry signal. At 0% wet the effect is running but silent, so every other knob on it will seem to do nothing — you are adjusting an echo or a tail you cannot hear. The fix is always the same, on any effect: raise wet until you can hear it, then shape it. Because wet and dry are simply added together, raising wet never removes anything from the original signal the way a filter or a drive stage would; it only adds a second, later copy underneath it. In a finished mix, delay and reverb sends usually sit lower than feels natural while learning, often 10–30% wet, because the dry signal is meant to stay in front. This lesson turns wet up louder than that on purpose, so the mechanics stay easy to hear.
Delay: copies you can count
A delay is an echo machine, and its two knobs mean exactly what they say. Time sets how long after the source each echo arrives. Short times, 80–120 ms, sit close enough to the source to thicken it rather than repeat it — the classic slapback heard on rockabilly guitar and vocals. Times from 200–500 ms read as a clear, separate echo, a reflection off a wall across the room. Push time toward the toy's 900 ms ceiling and the echo answers the note almost a full second late instead of following it. Musicians usually set time to the song's tempo rather than to a fixed millisecond value: at 120 BPM a quarter note is 500 ms, an eighth is 250 ms, and a dotted eighth is 375 ms — the interval behind most dub and reggae echo.
Feedback sets how many of those echoes you hear. Each repeat is quieter than the last by the same fraction, so feedback behaves like a volume knob applied over and over. At 50% feedback, each echo is exactly half the level of the one before it — a 6 dB drop every repeat — and by the sixth repeat the tail is 36 dB down, close enough to silent to stop counting. Near the toy's 85% ceiling, each repeat only loses about 1.5 dB, so dozens of audible repeats stack up before the tail fades: the classic dub effect of a delay that seems to keep going on its own. Push time down near the toy's 60 ms floor with feedback still high and the repeats start arriving fast enough to brush against the boundary lesson 1 drew between rhythm and pitch. Real hardware delays pushed shorter still cross that line outright, and the echo stops sounding like a series of repeats and starts sounding like a hollow, ringing tone.
Six repeats at two feedback settings. At 50% each echo drops 6 dB and is essentially gone by the sixth repeat; at 85%, the toy's ceiling, each drop is only about 1.5 dB and the repeats are still going strong.
Reverb: copies too dense to count
Clap once in a stairwell and thousands of reflections arrive so fast and so close together that your ear cannot separate any single one. That fusion is reverb: statistically many delays at once, each individually inaudible. Its master control is decay, the time the tail takes to die out, and your ear reads that time directly as the size and hardness of the room. Real reverb units add a second control this toy leaves out: pre-delay, a short gap of silence before the tail begins. A few milliseconds of pre-delay reads as a small, close space; 20–50 ms opens a gap the ear interprets as distance to the nearest wall before the room's reflections arrive.
- Booth — decay under 0.8 s. Tight and dry, barely a room at all; the low end of the toy's own range.
- Room — decay 1–1.5 s. The ordinary size for a mixed vocal or drum bus, close to the toy's 1.6 s default.
- Hall — decay 1.8–3 s. Big enough that the tail becomes part of the sound rather than a finishing touch.
- Cathedral — decay past 3.5 s, near the toy's 4 s ceiling. The tail can outlast the note that triggered it.
The same idea at two densities on one time axis. On top, a handful of separate echoes, each one countable. Below, hundreds of unseparated reflections under the same kind of decaying envelope — the individual spikes are inaudible, and only the smooth curve over them is what you actually hear.
A pluck sounds every ~1.7 s, with silence after it — that gap is where the effects live. Work one effect at a time: zero the reverb wet, raise the delay wet, and drag time left/right — hear the echo hug the note, then answer late; raise feedback and count the extra repeats. Then zero the delay wet, raise the reverb wet, and drag decay — the room grows from booth to cathedral. Only then mix both.
On the Digi-series boxes these are send effects: DELAY and REVERB each get their own pattern-level page, and every track chooses how much to send from its AMP page (DEL/REV). Tonverk turns both into swappable FX machines on its send tracks, including three separate reverb algorithms, each with the PRE (pre-delay) control this lesson described: