Originally Posted by stevekale
One comment on this. Of course there are different flavors for different tastes, but just because the room dominates in one portion of the frequency range and the speaker tends to in another doesn't necessarily mean that one shouldn't use Dirac (or any other frequency response compensation) to correct the region dominated by the speaker's response. That can still be improved. How far up the curve you go is up to you and what you think you hear.
IMHO, it's actually not that there is a frequency range that room "dominates" more and another that speaker "dominates" more.
To me, the room "dominates" throughout frequency range, period. (Better word probably is "affects.") It's the TYPE of "domination"
that leads to above observation. Below Shroeder frequency, room affects sound because of room modes
, above, room affects sound because of room reflection
(power response versus anechoic response).
We should correct for room modes, not for room reflection, is the argument. But as you wrote and I partially agree, "how far is up to you." Having said that, either you do or you don't
, either you correct for midrange and treble reflection, or you don't. The Shroeder frequency is the only point where you make a decision whether to correct above it. It wouldn't make sense to randomly, for example, picks 2000 Hz as the cut-off. IMHO.