Velocity layers map different recordings to one pitch so harder playing triggers a different sample, not just a louder one

A single MIDI note can trigger a different audio file depending on how hard the key is struck. This matters because an instrument played gently has a different timbre, attack, and overtone content than one played hard — not merely a quieter version. Velocity layering assigns distinct recordings to velocity ranges for the same pitch, so soft presses fire the gentle recording and hard presses fire the full-strength one. Splits are even by default: two layers cover 0–63 / 64–127; three cover roughly 0–42 / 43–85 / 86–127, with higher layer numbers mapped to higher velocities. This is fundamentally different from scaling amplitude by velocity (available in any sampler): amplitude scaling cannot reproduce the timbral/overtone changes captured by separate recordings, which is why multisampled virtual instruments sound far more convincing than a single sample turned down.

Examples

Two piano samples, soft and forte, assigned as velocity layers 0 and 1: velocity 40 plays the soft sample, velocity 100 plays the forte — different recordings, not a gain change. A whisper-quiet snare made by turning down a hard-hit sample has the wrong timbre; a properly layered kit gives each intensity its own recorded character (as in EZdrummer, Superior Drummer, Battery).

Assessment

For a pitch with three velocity layers, draw the velocity range each layer covers. Explain why the same sample at velocity 30 vs 100 sounds fake and why velocity layering beats amplitude scaling for realistic simulation, referencing overtone/timbre change.