Reverberation is one of the most important acoustic phenomena in our everyday experience of sound, yet we rarely consciously notice it because our brains are exquisitely adapted to interpreting reverberant spaces. When a sound is produced in any enclosed environment, the sound waves radiate outward from the source and interact with every surface they encounter—walls, floors, ceilings, furniture, and even people. Each surface reflects a portion of the sound energy while absorbing the rest, and these reflections continue bouncing between surfaces, creating an extraordinarily complex pattern of delayed copies of the original sound that gradually diminish in energy over time.

Acousticians divide this reverberant response into distinct temporal regions. The direct sound arrives first, traveling straight from source to listener. Early reflections arrive within the first 50-80 milliseconds, having bounced off one or two surfaces. These early reflections provide crucial spatial information about room size and listener position—our auditory system uses the timing and spectral content of these reflections to perceive the dimensions and character of a space. The late reverberant field develops as reflections multiply exponentially, becoming so dense and numerous that individual reflections can no longer be distinguished. This dense reverberant tail is perceived as a continuous wash of ambient sound whose decay rate characterizes the room's acoustic signature.

Digital reverb algorithms simulate these acoustic phenomena through several approaches. Algorithmic reverb uses networks of delay lines, allpass filters, and feedback loops to generate synthetic reflection patterns that approximate natural reverberation. Manfred Schroeder and James Moorer pioneered many of these techniques in the 1960s and 1970s, creating architectures that remain the foundation of modern reverb algorithms. The parameters of these networks—delay times, feedback coefficients, and filter characteristics—are tuned to emulate specific room types. Pre-delay sets the time between direct sound and first reflection, decay time controls how long the reverberant tail persists, and damping filters simulate the frequency-dependent absorption of real-world materials, where high frequencies are typically absorbed more quickly than low frequencies.

Convolution reverb takes a fundamentally different approach by capturing the acoustic signature of a real space as an impulse response—a recording of how the space responds to a brief, broadband impulse like a starter pistol or balloon pop. This impulse response is then mathematically convolved with the input audio, effectively imprinting the room's complete acoustic character onto the dry signal. Convolution reverb produces stunningly realistic results but is computationally expensive and offers less flexibility for creative parameter adjustment compared to algorithmic approaches.