AudioRoundTable.com: Speaker

Home » Audio » Speaker » Sub placement

Show: Today's Messages :: Navigator :: Polls
E-mail to friend

Re: We've discussed this before [message #18580 is a reply to message #18579]

Thu, 23 February 2006 13:02

Earl Geddes
Messages: 220
Registered: May 2009

Master

In a space with modes all are connected. In other words, unlike above the modal region, all points in the room are correlated and depend on what is happening at every other point. This is why random works as well as it does. Go random, it works better even for a small number of subs.

Report message to a moderator

Re: We've discussed this before [message #18581 is a reply to message #18580]

Thu, 23 February 2006 13:46

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

I understand what you are saying, and agree with you in principle. But I ask again, how can you be sure that a random placement will produce the desired result in all cases? I think everyone agrees that bass smoothing in a small room is accomplished with increased numbers of bass sound sources. The question is where is the best place to put them. Welti suggests a handful of specific placements, you seem to suggest rolling the dice.

AES Preprints: AES 112th Convention

Report message to a moderator

Re: We've discussed this before [message #18582 is a reply to message #18581]

Thu, 23 February 2006 14:00

Earl Geddes
Messages: 220
Registered: May 2009

Master

Its been a while since I read it, but as I recall a finite number of studies were done because it was mostly an experimental one, with a little simulation. But what I remember was that he had missed some important configurations and the ones that I recommend were not among them.

Maybe I should reread it, but as I've posted, I did my own studies and my results were not exctly the same as Todds. I remeber rereading his to see how it was that we disagreed and found that he hadn't tested what I found to work best. In other words he was drawing broad conclusions based on a limited investigation and my work did not cooberate his.

If I get a chance I will run the four subs in corners against four radndom subs and I think that you will find that they are not much different. And then I'll run three subs at random and show that this is only marginally worse than the four corners, with one fewer sub.

Report message to a moderator

Re: We've discussed this before [message #18583 is a reply to message #18582]

Thu, 23 February 2006 14:23

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

The study indicates that random placement was good. It also indicates that four corner placement is equally good. Like you said, if there are a lot of subs involved, their placement matters less. But I am still concerned though about the "roll of the dice" issue. The decorrelation idea by random placement sounds good, but I can't help but wonder how many random placements happen to line up nodes in an undesirable way.

Report message to a moderator

Re: We've discussed this before [message #18584 is a reply to message #18583]

Thu, 23 February 2006 14:36

Earl Geddes
Messages: 220
Registered: May 2009

Master

As I said I cannot cooberate Welti's study, so that leaves me with some doubt. Its natural I guess to believe in your own study and doubt the other guys, but the fact remains that I did not get the same results he did. Todd and I did tend to agree on the random placement, but we did not tend to agree on the symmetric placement. So what has been cooberated is the random placement and, to me, what is in doubt are the symmetrical placements.

Report message to a moderator

Re: We've discussed this before [message #18585 is a reply to message #18584]

Thu, 23 February 2006 15:02

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

I hate to sound like a broken record, but how can you be comfortable that random placement always provides desirable results? How many tests have you setup and run?

Report message to a moderator

You don't seem to listen [message #18586 is a reply to message #18585]

Thu, 23 February 2006 15:15

Earl Geddes
Messages: 220
Registered: May 2009

Master

My PhD thesis was on LF room modes. The Welti study was incomplete because it did not include the configurations that worked best. Any conclusions he drew were shortsighted at best.

Report message to a moderator

Re: You don't seem to listen [message #18587 is a reply to message #18586]

Thu, 23 February 2006 16:30

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

I read your comments. And I respect and consider them. But the only data I see is from the Welti study. You have presented no other data to the contrary, just unsupported assertions and qualifications from your resume.

Please don't take that as being said in a condesending tone. It isn't mean that way. But I do not see your data rising to the level of Welti's, in fact, I do not see any data from you at all. What I would prefer to see, is a study of a large number of random setups that showed the energy distributions in each. That way it could be determined how reliable this approach is over a number of different iterations, different rooms, different random setups, etc.

Report message to a moderator

Thanks for giving the Welti source [message #18588 is a reply to message #18581]

Fri, 24 February 2006 14:27

Duke
Messages: 297
Registered: May 2009

Grand Master

I'm off to read it now.

By the way, I have enjoyed this exchange immensely. If it weren't for the passionate and the stubborn, us in the peanut gallery wouldn't learn nearly as much.

Duke

Report message to a moderator

Re: Sub placement [message #18589 is a reply to message #18565]

Fri, 24 February 2006 15:29

Manualblock
Messages: 4973
Registered: May 2009

Illuminati (13th Degree)

Can I ask why they in the Welti Study feel that summing all the sub-woofers is the most optimal implementation? Why do they considor stereo bass of no value?
Just asking.
I assume the original signal is recorded in stereo; is that right?

Report message to a moderator

Pages (4): [ « ‹ 1 2 3 4 › »]

Previous Topic:	Raw Acoustics Omega's
Next Topic:	4.5 " driver 8 per side for a mini array

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sun Nov 24 14:46:54 CST 2024