Listener preference is driven mostly by timbral neutrality across the listening window

Evidence

Again, to explain in plain language: PIR is an empirical prediction of a steady-state room curve based on 12% listening window, 44% early reflections and 44% sound power (Devantier, 2002). In this case, my opinion is that in-room response, by itself, has not taken on a special importance. Instead, the model has recognized NBD and SM, both indicators of smoothness in the data from which PIR is calculated. When such smoothness is seen in all three of these curves it is a strong indicator of an absence of resonances. Smoothness is also seen in the on-axis (direct sound) curve. Combined, indicators of timbral neutrality accounted for 69.5% of the total score. Bass extension is highly valued, at 31.5% of the total score. Bass smoothness, LFQ, was not consequential in this model, probably because many of these loudspeakers had the bass energy to excite invasive room resonances, that would have masked any intrinsic smoothness the loudspeakers may have had. — Toole, Floyd E, Sound reproduction- the acoustics and…, 1216

Categorization

• DEFINE: Plain-language definition of PIR as a weighted blend of listening window, early reflections, and sound power.
• XREF: Connects to Olive/Harman spinorama and NBD/SM smoothness metrics used in loudspeaker measurement literature.