Too bad this poll isn't multiple selection! I really want to answer with three of the choices:In general, it does more harm than good
Many RC systems try to do too much,
IMO. I echo Sanjay's thoughts above about EQ'ing HF from the listening position. (A nearfield
correction may arguably have more merit.)
Basically, the way I see it is that RC should do three things:
1) Equalize the response in the modal region and below. This action can be very effective, and in a multisub system it can even be effective over a wide listening area;
2) Set up a calibrated system for modern DSP-based loudness correction (Dolby Volume modeller, Audyssey DynamicEQ, etc.) to base its correction curves on. One still needs to play with reference level offset, etc. to taste, but overall very useful; and optionally
3) Set a target curve in the highs. Ignoring the measured peaks and dips, there's something to be said for having some control over the big picture treble response.
When it tries to do more, it usually screws up everything. The only exception of which I can think is Trinnov's spatial processing, which is kind of neat if one can't place speakers optimally. But good placement is still better, and useful over a wider listening area.In general, it does more good only in the bass frequencies
Self-evident. In the modal region (say, ~200Hz down to ~50Hz in a typically living room) and the 1st mode region (~50Hz down) the room's volume matters more than the loudspeakers' native response. So correction of several types (multiple subwoofers + DSP, ideally) is needed to bring it under control. Also, IMO, any correction system that does not
have a "no correction above [user-selectable] frequency" option is inadequately designed.It depends on the room-correction system; some work better than others
This one has mostly to do with target curves. It's fairly well-established that a target curve that allows for "room swell" is perceptually more natural than one that levels the upper bass to flat. Likewise, the target curve should not presume that the user is incapable of picking competently-designed loudspeakers, by introducing notches into the midrange in attempts to band-aid directivity shifts that are not present in competently-designed loudspeakers.
Originally Posted by HD-Master
Audyssey works rather well actually. I'd like to hear from the folks who voted that it does more harm than good, but didn't bother to say why.
Audyssey's biggest problem is that to undo the flawed assumptions it starts with (user has crappy speakers, measured-flat-in-room upper bass) one needs to spend a fairly large sum to get the Pro license and measurement kit. Then one can draw a target curve that preserves natural "room swell" and delete-options Audyssey's crappy speakers compensation midrange notch. Pro should really come standard on any four-figure AVR with Audyssey technology embedded.
Originally Posted by adidino
Here's one I currently use for movies. I have 3 versions of this curve which differ slightly in target bass response.
What is the scale of that graph?
Originally Posted by HowardV
We were in the same room correction seminar at the Newport Beach show. If you recall, a "general" rule was cover about 1/3 of the wall with acoustical padding.
Room mutilation is a non-starter for many. Fortunately, if one is sensible about picking loudspeakers with smooth off-axis response, and chooses loudspeakers with directivity suited to the room and one's preference in the specificity/spaciousness continuum, room mutilation such as "cover[ing] 1/3 of the wall ***" is entirely unnecessary. Problem is, most speakers on the market are rather badly designed, so to get them to sound right one has to kill the excess off-axis midrange energy through room mutilation.