Hello
@Thomas Lund I have read your
study. It appears that if we want "Acoustic Envelopment", we need uncorrelated bass signal, ideally with subwoofers placed on either side of the listener. I have a few questions.
View attachment 420710
The first is about the test signal. Pink noise has equal energy per octave, so an uncorrelated test signal would have the same equal energy per octave as the correlated signal. But music is not like this - it does not necessarily have the same equal energy per octave across both channels. So I wonder whether your uncorrelated pink noise was played at the same SPL in both channels, and how much the effect is dependent upon SPL per channel.
Second, the effect of AE was not studied in an enclosed space. To very quickly summarise your findings: with the headphone listening test (Test B), the listeners reported that the correlated signal sounded "small", "constricted", etc. while the uncorrelated signal sounded more "open" and "free". With the subwoofers in the open field (Test C), the same result was reported, but the sensation was more pronounced when the subwoofers were on either side of the listener's ears (Test C1 and C4). And finally, Test D showed that sensitivity to bass direction dropped with longer wavelength with fewer listeners able to identify the difference between correlated and uncorrelated signal.
I wonder whether the effect would still be as pronounced in a listening room, and whether the sensitivity to the effect would start dropping off at shorter wavelengths than those thresholds you found in test D. We already know from other studies (like those of your colleague Makivirta) that sensitivity to group delay is poor for longer wavelengths, and those studies were conducted with headphones and not in a listening room. I imagine it would be even worse in a listening room.
So if it is not uncorrelated SPL that produces the sensation of AE (or more precisely, we don't know since uncorrelated SPL was not tested), it should be uncorrelated timing. But enclosed spaces mess up the timing.
Please don't get me wrong, I really appreciate you doing this research and advancing our knowledge. I am concerned about the generalisability of your findings for typical applications - i.e. people who listen to music in listening rooms.