The energy after an onset
After a pluck or strike, a sound often falls from an initial peak as stored energy leaves its vibrating parts. In an ADSR synthesizer envelope, decay has a narrower definition: the segment from the attack peak down to the sustain level. In everyday descriptions it can mean the whole fading tail. Read the lab’s labels to know which meaning applies.
Evidence: 1
A falling curve has a shape
A linear fade loses an equal amount of amplitude in each equal time interval. An exponential fade loses an equal fraction. They sound and look different, especially near the end. For an exponential amplitude model with time constant 0.4 seconds, the value after 0.4 seconds is about 37% of its initial value; after 0.8 seconds it is about 14%.
Evidence: 2
One sound can have several decays
A string’s high modes may lose energy at different rates from its low modes, and the body can continue responding after the initial contact ends. Thus a note may become darker as it fades. A single overall envelope scales all components together, so it illustrates duration while deliberately omitting this spectral evolution. Recognizing that limit is more useful than treating every short envelope as a piano.
Evidence: 3
Hear the musical space between notes
At a fixed pulse, a longer decay can overlap the following note while a shorter decay leaves a gap. Neither is inherently better: one may support continuity, another articulation. In the envelope lab, repeat the same note with two decay settings and listen to the tail as carefully as the onset. Then imagine a sequence where pitch changes before the previous tail has ended.
Evidence: 4
Linear versus exponential
A linear fade can use A(t) = max(0, 1−t/D). An exponential fade uses A(t) = e^(−t/τ). Neither curve is a full physical model of every instrument.
Decay describes falling energy or an envelope segment; its shape and frequency dependence affect articulation.
Make the connection
The beginning changes everything
Keep attack fixed and compare short and long decay times.
Open the experiment ↗Useful words
Decay
A decrease in a sound’s level over time. Different partials can decay at different rates, changing timbre during the note.
See in glossary ↗Sources & further reading
These sources support the explanations and model assumptions. Recording credits are listed in the listening panels and the source register.
- Envelopes ↗Chris Donahue / Carnegie Mellon University · 2026
- Oscillations and simple harmonic motion ↗Joe Wolfe / UNSW Physclips · Undated
- Musical sounds and musical signals ↗Joe Wolfe / Sound Musicianship · 2012
- Scores versus timbre ↗Chris Donahue / Carnegie Mellon University · 2026