The Delay Line: One Primitive, Many Effects
Comb filter, flanger, chorus, echo, and Karplus-Strong synthesis are all the same thing: a delay line with feedback at different time scales.
The Simplest Effect
A delay line does one thing: it holds audio for a set amount of time, then outputs it. Mix this delayed copy with the original, and you get an effect. Feed the output back into the input, and echoes stack up. That is literally the entire mechanism behind five of the most important effects in synthesis and music production.
What changes is the delay time. At sub-millisecond delays, the copies are so close together that they interfere constructively and destructively at specific frequencies, producing a comb filter. Push the delay to 1–5 ms with a slow LFO sweep and you get a flanger. Extend to 5–30 ms with gentle modulation for a chorus. Beyond 100 ms, you hear distinct repetitions: echo. And at very short delays with extreme feedback, the delay line resonates at a pitch. That is Karplus-Strong synthesis, where a delay line becomes an oscillator.
One Knob, Five Effects
The demo below has a single delay line. As you sweep the delay time from sub-millisecond to over a second, a label tells you which effect region you are in. There is no discontinuity; the boundaries are arbitrary conventions. The physics is one continuous spectrum of time-domain repetition.
Add LFO modulation to the delay time and you get the characteristic sweep of flanging and chorus. Add feedback and echoes pile up. Combine very short delay with very high feedback and the delay line rings like a tuned string.
The Presets: A Guided Tour
Comb Filter: at 0.5 ms with high feedback, the delay is so short that it creates interference patterns: certain frequencies reinforce while others cancel. The result is a metallic, hollow timbre. This is literally what a comb filter is. The frequency response looks like the teeth of a comb.
Flanger: push the delay to 2 ms and add a slow LFO sweep. Now those comb filter peaks slide up and down the spectrum. The result is the classic jet-plane whoosh that defined 1970s guitar and synth sounds. The feedback makes the peaks sharper and more pronounced.
Chorus: at 15 ms with no feedback and gentle modulation, the delayed copy is far enough behind the original that you hear it as a separate voice slightly out of tune. The LFO creates a subtle pitch wobble, and your brain interprets this as multiple instruments playing in unison, hence "chorus."
Slapback: at 80 ms, you hear a distinct repeat, like singing in a tiled bathroom. Rockabilly and early rock & roll used this constantly. It adds presence and energy without the complexity of longer delays.
Tape Echo: at 350 ms with moderate feedback, each repetition arrives well after the last, creating a rhythmic trail of echoes. Real tape machines added warmth as each repeat degraded slightly. Our digital version stays clean, but the rhythmic effect is the same.
Karplus-Strong: the surprise ending. At 0.8 ms with feedback cranked to 0.98, the delay line resonates at a specific pitch (1/0.0008 ≈ 1250 Hz). This is Karplus-Strong synthesis: a delay line becomes an oscillator. Plucked string sounds in many synthesizers use exactly this technique.
Try it: play a note on the keyboard and slowly sweep the delay time from left to right. Listen to how the same mechanism transforms from metallic resonance to jet-plane flanging to warm chorus to distinct echoes.