Originally Posted by cap'n
If I may be so bold as to summarize what you are saying, it is this: there is some quality of large rooms that we would like to duplicate in small rooms. I think you agree now that room modes cannot be created, but perhaps some resonances could be added.
That is incorrect. Further, your "now" directly implies ignorance on my part. What is more, you're again mistating the facts.
I know of nobody who suggests that you can get RID of room modes in a small room. You can, however, with multiple radiators, control them. You can alter them actively, etc, but you can't change the basic physics, except to do equalization via multiple woofers. The modes exist, but you are dealing with them in a rational fashion.
It seems, since you appear to understand that, quite surprising to me that you do not realize that you can, as well as mitigate modes, add
other modes, of course, due to the loudspeaker, not due to the basic acoustics of the listening room. The two are parts of the same process. Perhaps your presumption that you would never want to do that is the problem?
You can, (to use the artificial method, which is not the one I endorse, but which is easier to describe) add MORE modes, with more woofers, of course, by (no, I don't suggest that this is the best way) putting in delay time to the woofer that substantially mimics the image in the ORIGINAL HALL that would create the mode in the ORIGINAL HALL. Yes, it's artificial. So what? It's another mode. In fact, with multiple woofers, you would actually add such information to all woofers, of course differently for each, corresponding to the set of new images and modes you wish to have.
So, WITH MULTIPLE WOOFERS, you can simulate characteristics of a larger room. This is not the near-complete cancellation, etc, that opposed array can accomplish (but I think 'easily' is quite a misnomer, and I don't see it happening under consumer control
any time soon), but rather adding some of the modes in the large room to the small room. Yes, they will actually show up there, unless you're about to argue that the method of images is in fact flawed. (it's not that simple, of course, you have to account for the wall diffusion at each bounce, etc, etc, there are myriad of fiddly details)
So, your statement is suborned. I said you can not REMOVE modes due to a small venue. I absolutely do not agree with your statement that I can not add them. Yes, electronically and with more radiators, not by using the basic physics of the room, of course. And before you attempt the fallacy of the excluded middle again, yes, you can control modes due to the basic physics of a small venue.
But it seems to me that those types of resonances are incorporated into the recordings themselves, i.e. the concert hall reverberation.
Again, you fail to distinguish between frequency response shaping (by which I will presume you mean power spectrum) and interaural correlation.
Using coincident miking, you have no spatial separation. You can, to some extent, by using XYZW miking, capture the whole soundfield at one point, but you're not capturing the decorrelation as a function of distance. You are missing information
present at your ears in the original soundfield.
Using part-velocity microphones, at an appropriate distance, you capture at least some of the perceptually important cues. The reason for using velocity microphones is, of course, that 1) they are directional, and 2) they capture more of the soundfield information, and from a direction where it will ultimately be rendered.
So, you are again ignoring binaural cues when you refer to "resonances".
Your ears are not coincident. The coherence length in a good, large room is surprisingly shorter than a wavelength in some cases.
But you are looking for some further enhancement of the “spatial” properties, which leads us to the decorrelation issue we have already flogged to death, and the lack of convincing evidence.
No, I am looking to capture the information. "Resonances", which is to say magnitude spectrum information alone, does not capture the information in the first place. A single-point pressure measurement is discarding 3/4 of the information in any given soundfield at any given point.
As to evidence, you haven't cited a single test that addressed the subject I'm discussing.
Edited to add:
By the way, when you "quote" me, be careful to quote me in context, and with included qualifications, and avoid fallacies of the excluded middle, thank you.