Concert hall sound varies 20-30 dB across frequencies yet listeners barely notice.

Evidence

When the equipment first became available to measure it, acoustical engineers were horrified to find that the point-to-point propagation of sound in rooms and especially in concert halls is strongly colored by peaks and dips as a function of frequency (see figure 27.15). Maxima and minima in the amplitude of the sound arriving at a given point can differ by 20 to 30 dB from one frequency to the next; the figure shows that in the case of the concert hall measured, the maxima average about 10 Hz apart, and have a width of a few Hz. The inventor of the condenser microphone, Edward Wente (1889–1972), faced with such data, questioned whether concert halls should even be a good place to listen to music! It seems empirically that such variations actually don’t matter very much. — Eric J. Heller, Why You Hear What You Hear- An…, 2547

Categorization

• SEED: Angle: why vast spectral irregularity in rooms doesn’t harm musical perception — what this says about how the ear’s integration works.
• QUESTION: Why don’t 20-30 dB spectral peaks and dips matter perceptually? What perceptual mechanism flattens them out?
• XREF: Relates to the reader’s underlying question of how sound becomes hearing — psychoacoustic integration over frequencies.
• SHARE: Good spark for an audio/engineering-minded friend — the fact that Wente, inventor of the condenser mic, questioned whether halls should host music at all.