Room & listening
Crest factor
The power gap between average loudness and brief musical peaks.
Simple
The gap between the average loudness of music and its brief loud spikes — a drum hit, a cymbal crash. Even at a modest average volume, those spikes need a lot more power for a split second.
The theory
Crest factor is the ratio of peak amplitude to RMS (average) amplitude, expressed in dB. The engine adds it to your target SPL to get the peak SPL the system must reach cleanly: peakSPL = targetSPL + crestFactor. The default of 15 dB reflects typical well-recorded music; the power for those peaks is what separates "loud enough" from "loud enough and clean."
For experts
15 dB is a deliberately honest default. Heavily compressed pop ("loudness war" masters) can sit at 6–9 dB crest, while uncompressed orchestral or live recordings hit 18–20 dB. The catch: peak power scales as 10^(crest/10), so a 20 dB-crest recording demands ~30× the peak watts of its 5 dB-crest counterpart at the same average level. This is why a 50 W amp can sound "underpowered" on classical but bulletproof on EDM — the music's dynamics, not just its loudness, set the bill. It's also why peak-vs-RMS amplifier specs matter and why momentary clipping on transients is the most common (and most overlooked) form of distortion.
Crest factor: the peak-to-average ratio of the signal, in dB.
A 15 dB crest factor means transients demand ~32× the average power.