Vocal delay settings: slapback, dotted eighths, and throws

Why mixers reach for delay before reverb

Reverb fills all the space around a vocal continuously; delay places discrete copies in rhythm and leaves everything between them silent. In a dense arrangement that difference is decisive — a delayed vocal keeps its pocket and clarity where a reverberant one washes out. Plenty of modern rap and pop vocals carry no audible reverb at all: their sense of space is delay doing the work. The two aren't rivals, though; a short reverb for close space plus a filtered delay for depth is the standard pairing.

Setup 1 — Slapback: thickness, not echo

A slapback is one fast repeat, quiet enough that listeners hear a thicker vocal rather than a delay. In Ableton's Delay: sync off, time 80–140 ms, feedback at or near 0%, and the repeat tucked well under the dry vocal — Dry/Wet around 10–15% as an insert, or a send with the return faded until the repeat is felt more than heard. Below ~80 ms it starts to comb-filter into the dry signal; above ~140 ms it reads as a distinct echo.

Slapback earns its keep on sparse arrangements and verses where a full reverb would feel too produced — it adds size while sounding like nothing at all. On a return-track setup, pan the repeat slightly off-center for a subtle widening bonus.

Setup 2 — The dotted-eighth workhorse

The default modern vocal delay: Ableton Delay with sync on, left channel on a dotted eighth, right on a quarter note (or both dotted-eighth for a tighter feel), feedback 20–35%, Dry/Wet 10–15%. The offset between channels makes the repeats dance around the vocal instead of stacking behind it.

The setting that separates clean from cluttered is the filter: enable Delay's filter section — it's a band-pass with Freq and Width controls — and set it so the repeats pass roughly 300 Hz to 5 kHz. Filtered repeats read as depth; full-bandwidth repeats read as a second singer talking over the first. This mirrors the return-EQ logic from the reverb world — tails and repeats always want to be darker than the source.

Setup 3 — Throws: the automated moment

A throw is a delay that exists for one word: the last word of a hook 'thrown' into a long repeating echo that fills the transition into the next section. Set up a return track with a quarter- or half-note delay, feedback 40–60%, filtered like the workhorse delay. Keep the send at −∞ by default, then automate it up for just the target word and back down immediately.

1. Create a return track with Delay: sync on, 1/4 note, feedback 50%, filter on and narrowed so the repeats stay dark, 100% wet.
2. Add a send from the vocal track, defaulted to −∞ (fully down).
3. Find the phrase-ending word, automate the send up to taste (start around −10 dB) for that word only.
4. Let the repeats ring through the gap, and make sure the tail clears before the next vocal entrance.

Throws are automation, not a static chain setting — which is why generated racks and presets don't include them. They're the human layer you add on top; see where automation fits in our guide to making vocals sit in the mix.

Delay vs. Echo in Ableton

Stock Delay (in every edition — Live 10.1 merged Simple Delay and Ping Pong Delay into it) covers every setup above: per-channel sync, ping-pong mode, filtering, and clean repeats. Echo, included with Suite, is the character option — analog-style saturation, modulation for tape-like wobble, a gate, built-in ducking, and a reverb inside the delay. If the reference sound has dirty, wobbling repeats, Echo gets you there faster; if it just needs clean rhythmic depth, Delay is lighter and available to every Standard user.

PhantomRack generates its delay stages on stock Delay for exactly that portability reason, setting sync note values, feedback, and filtering from the repeat pattern and tail brightness it hears in your reference — see the full vocal chain guide for where the delay stage sits.