Low-Q resonances become audible at peaks below 0.5 dB in the midrange

Evidence

The middle curve shows just detectable medium-Q (Q = 10) resonances. For pink noise the 200 Hz peak was about 2 dB, the 1 kHz peak was about 1 dB and the 5 kHz peak was about 0.5 dB. With pop/jazz the 200 Hz peak went up to about 6 dB. The bottom curve shows detection thresholds for low-Q (Q = 1) resonances. For pink noise the 200 Hz peak was about 1 dB, and the 1 and 5 kHz peaks were around 0.25 dB. With pop/jazz the 200 Hz resonance peak went up to about 2.5 dB. — Toole, Floyd E, Sound reproduction- the acoustics and…, 505

Here the dashed line in Figure 2.13(b) tells us that it is something else. Broad low-Q undulations, with superimposed bumps and wiggles together use up the full 4 dB tolerance. It is commendable that the product meets its specification, but the specification isn’t informative. What do the curve fluctuations indicate in the way of anything audible? We know that low-Q resonances are audible when they are revealed by humps of about 1 dB amplitude or more, so there are likely be some audible broadband spectral colorations. The smaller bumps might be evidence of acoustical interference, which is innocent, or resonances, which are not. Without more information we don’t know which they are, but the suggestion is that this is not a neutral-sounding loudspeaker. — Toole, Floyd E, Sound reproduction- the acoustics and…, 239

Categorization

• DEFINE: Defines what Q-values (50, 10, 1) mean for resonance bandwidth and how they shift detection thresholds.
• corroborated: 2 independent passages support this claim — Toole, Floyd E, Sound reproduction- the acoustics and…, 505; Toole, Floyd E, Sound reproduction- the acoustics and…, 239