[youngho]

Hi Brian, Welti was looking at multiple seat response. However, this was WITHOUT equalization. WITH equalization, the corner placement was the third or fourth best configuration IF the listening positions are centered in the room. However, when the listeners are shifted somewhat towards the rear of the room in investigation #7, corner placement of the four equalized subwoofers definitely falls short of many of the other configurations.

Duke, thinking about it a bit more, I realized that the 90 degree phase difference will maximize the velocity vector. In terms of decorrelation of bass energy, it seems to me that this would be only really relevant to the width mode (assuming that the two subwoofers are placed at the midpoints of the side walls in a rectangular room) and height modes. Also, I forgot that corner placement of a single subwoofer may be ideal in the scenario where the subwoofer's fourth-order low-pass crossover frequency is set below the lowest room mode.

Young-Ho

[Duke]

Young-Ho,

I must have missed where the metrics for the corner location are given, presumably along with other metrics, so that they can be compared.  Can you tell me where you see that in Welti's Harmon paper?  Or are you looking at his AES paper? 

Note that the smoothest result of all is shown in figures 40 and 41, with four subs each 1/4 the way out from a corner - but that was evidently judged an "impractical" configuration.  It looks to me like Welti's study points to four subwoofers at the wall midpoints as being the smoothest "practical" configuration of the ones he examined.  Of course without a rectangular room, I don't know whether any of Welti's multisub configurations could be precisely replicated.   

He goes on to say that equalization has "some effect, but does not change [the] overall conclusions."   So I don't think the fact that Welti uses equalization and Geddes does not disqualifies the comparison Geddes made.

While the corner may be better in some ways than some symmetrical multisub configurations, it is the successful configurations we're interested in - which according to Geddes would include asymmetrical placement. 

Duke

[bpape]

When we set up home theaters and other multi-seat setups, if space allows, we do 1 sub centered each front and rear. If there is still room (not very often), we add 1 centered each left and right.   

Bryan

[jimdgoulding]

Just a little note to Sat and to describe sound in my lively 12x15x8 room.   My room is like a horn speaker itself about the size of the front end of my room and my speakers are like a small diaphragm deep inside the horn.  I see em but I don't hear em.  I'm just sittin there in front of a bigger than it oughta be stage inhabited by musicians.  And there is no mystery about how they were recorded.  It doesn't seem to mess up recordings that have lots of ambient information in them.  It doesn't seem to me to mess up anything.  The sound of instruments remains characteristic to my speakers (got to live with that).  I figure there is a marriage of things going on here . . the size of my speakers, the size of the mid/bass drivers they use (4), where they are placed, and how they load to the room.  If someone had given me an analogy like I opened with a year ago, I would have smiled politely while I cringed inside.  That was before.  The reason I put so much emphasis on time, waveform behavior, and the listening room is because of what I'm getting.  The knowledge of which didn't originate with me, for sure, I just started fooling around and fell into right.  The definition of synergy, in my case, has expanded to include the behavior of energy to air and my perception of it.  Thanks for your indulgence.

[youngho]

Duke, please look at slides 69-71 for modelled results using unequalized subwoofers. Configuration #10 is the one with four subwoofers, one in each corner. Configuration #1 is the one with a single subwoofer in a corner. The metrics are basically the points on the graph: Std, Max-min, Max-ave, and LF factor. As you can see from the graph, for the four corner subwoofers, Std and Max-min are SLIGHTLY better, Max-ave and LF are worse, but I think you'll agree that the metrics overall are remarkably similar. Compare to configuration #6 with the two subwoofers on opposing wall midpoints and configuration #11 with four subwoofers, each on the midpoint of a wall. Definitely better.

I'm afraid that I completely miss your point about slides 40 and 41. There is no data about Std, Max-min, Max-ave, or LF, so the result from this one particular configuration cannot be compared to the majority of the others. This configuration of subwoofers is entirely non-random. It's specifically designed to take advantage of mode cancellation of the odd-order modes, as well as placement in the nodes of the even-order modes. However, in the absence of metrics, it's not possible to relate these results to the other investigations.

I agree with you entirely about the difficulty in relating the results of the Welti paper to nonrectangular rooms, which is why I wrote "Welti paper (which is really only applicable for rectangular, symmetric rooms with certain dimension ratios..." five posts ago.

Welti may written that sentence about the conclusions, and he's right about the first two conclusions, but it's important to look at the actual data in these sorts of papers rather than relying on the authors' interpretations alone, and I believe that anyone who looks closely at the actual data will have to agree that corner placement of four unequalized subwoofers performs relatively poorly, as I discussed above. This is why I wrote "the Geddes one, which did not compare random subwoofer placement with Welti's suggested placement but rather placing the four subwoofers in the corners of the room" and "Geddes' study did not use equalization, so I believe that you cannot relate the results of one study with the other." I never said that Geddes' study should be disqualified. In his model, four randomly located unequalized subwoofers performed better than four corner unequalized subwoofers, but this was a miniscule part of Welti's study. It doesn't say anything about equalization, typical listener positions (for example, Geddes' model positioned the listeners near the center of the room, which for a 3.5 meter tall room means that the modelled data would be relevant to a standing but not a seated listener. I like to sit down when I listen to music, so Geddes' results are not relevant to me), or actual real-world performance. Two subwoofers, each located at the midpoint of opposing walls, or four subwoofers, each located at a wall midpoint, may perform as well as or better than four randomly located subwoofers when they are all unequalized, but I don't know because the results of Welti's study cannot be adequately related to the results of Geddes' study, in my opinion.

As an aside, look at slide 51. "Not surprisingly, symmetrical configurations seem to work better than nonsymmetrical ones." However, no conclusion can be drawn from this sentence with respect to Geddes' paper, since the two cannot be adequately related.

You wrote "If we have two subwoofers spaced somewhat apart, they produce different peak-and-dip patterns at the listening position - not only smoothing the net result, but increasing the density of peaks and dips so that the ear has a better chance to average them out.   Most people who have gone from one sub to two report smoother, more natural-sounding bass, and I believe the beneficial interaction of the two subs' differing peak-and-dip patterns is largely responsible.  In fact, the improvement doesn't stop there; the more independent low frequency sources we have in a room, the smoother the in-room bass will be." Welti's paper shows that two subwoofers spaced somewhat apart may not produce different peak-and-dip patterns at the listening position, etc. It shows that having more independent low frequency sources may not result in smoother in-room bass. That's all I'm pointing out.

Asymmetrical placement might be the greatest thing since sliced bread, but there isn't good data to show that yet. I'm afraid that my posts have a decreasing signal-to-noise ratio at this point. I read a little, and although I have a good memory and attention to detail, I've approached the limits of my knowledge about these things.

Young-Ho

[satfrat]

Just a little note to Sat and to describe sound in my lively 12x15x8 room.   My room is like a horn speaker itself about the size of the front end of my room and my speakers are like a small diaphragm deep inside the horn.  I see em but I don't hear em.  I'm just sittin there in front of a bigger than it oughta be stage inhabited by musicians.  And there is no mystery about how they were recorded.  It doesn't seem to mess up recordings that have lots of ambient information in them.  It doesn't seem to me to mess up anything.  The sound of instruments remains characteristic to my speakers (got to live with that).  I figure there is a marriage of things going on here . . the size of my speakers, the size of the mid/bass drivers they use (4), where they are placed, and how they load to the room.  If someone had given me an analogy like I opened with a year ago, I would have smiled politely while I cringed inside.  That was before.  The reason I put so much emphasis on time, waveform behavior, and the listening room is because of what I'm getting.  The knowledge of which didn't originate with me, for sure, I just started fooling around and fell into right.  The definition of synergy, in my case, has expanded to include the behavior of energy to air and my perception of it.  Thanks for your indulgence.

Everyone's musical preferences are different to some degree or another, I'm sure Brian will attest to that. What moves one person might just turn another off. Most audiophiles I run into want that accuracy where the performer is right in their face and they are sitting in the front row...... Not me. I want to be sitting in the middle of the auditorium and I want my music dangling in the air and from behind the speakers location. I understand what's working for you Jim and that's a good thing.(for you) 


Cheers,
Robin

[Duke]

Young-Ho, thanks for discussing this with me - I have enjoyed it.  Perhaps part of our occasional difference of opinion lies in my giving greater weight to the ideas of Earl Geddes, admittedly in large part from having worked with him and spoken with him at length on the subject of low-frequency reproduction in typical home rooms.   

Until we meet again, very best wishes.

Duke

[jimdgoulding]

My comments are just in regards to in room realization of recordings in domiciles on the small side* and what is possible in my experience.  Except for the kind of bass reproduction under discussion here.  That is beyond my reach.  As to perspective, that’s a function of the recording in a good system.  If your speakers are out from your back wall three or four times their depth and two or three times their width from your side walls, you’ll get good staging and a neutral perspective.  Mine are and my approach is to incorporate my room to allow the acoustics to expand in all directions so my recordings, any recordings, live or otherwise, are free to happen as they might.  Focus and separation are excellent and I don’t perceive any nodes or roughness in the frequency response.  But then my speakers roll off quickly below 38Hz and my off axis frequency response is measurably smoother than ever (a bonus of my and yorn woolies!).  To put things in perspective (pun intended).  LOL

*for anyone this might include

[rockadanny]

Fascinating stuff, thank you guys, but back to my original concern, live music in a smallish (with a high ceiling) church ...
I though I might try slapping up some diffusion panels here and there to see if I can generate a decent reverberation field for most seats. Do you you think this is a waste of time because the whole building (architecture, etc.) needs to be addressed by a professional acoustician, or might I have some luck with these panels?

[bpape]

Diffusion in a church will require relatively large panels and normally, they are barrel type diffusors.  IMO, you'd be better off scattering some 2" absorbtion around the upper portion of the room.  What you're fighting is general decay times more than anything else.  People make great absorbers but they're down on the floor.  Address the upper portion minimally and things will improve.

Bryan

[rockadanny]

Thank you Bryan.

[youngho]

Have you measured reverberation times? Everest does discuss large spaces like churches a little in his book. You might try just browsing here, to start: http://books.google.com/books/mhprofessional?q=reverberation&hl=en&vid=ISBN0071360972

[rockadanny]

youngho - no, haven't done any measurements. Thanks for the link, I will check it out!