Cardioid Bass

[gedlee]

I've seen more technically inadequate papers reviewed and published in respectable journals in my career than I would care to admit, some of then just outright wrong from the get go. Like everything else, it's largely a political environment.

As an AES reviewer I find this offensive. I have never let a poor paper through. This position isn't fair to assume.

Actually it would seem more likely that the discussion of multiple subs is the academic one since it is far more unlikely a scenario than a typical set up with a single or stereo subs. It may be the best solution, but highly impractical and far more difficult to set up optimally. For a give room, how many and where, not to mention all the additionally (cost) amplification, crossovers...... required, and didn't you mentioned de-correlators? Developing an algorithm for placement and number of sources to produce optimal response in any room is an interesting problem, academically, but it just doesn't offer a practical solution in the majority of cases. [/B]

Now I think that you are being unreasonable. My solution is eminately practical, inexpensive and used in numerous rooms that I know of. You don't need big subs because you use more of them, they are inexpensive, and they are small so they hide well. You can put them just about anywhere - that was the point that I made in my AES letter - multiple subs are not very sensitive to location when placed randomly. In my room you can't see any of the three subs.

You just don't seem to want to understand my point of view so I guess I'll leave it there.

 

[gedlee]

Please remember that the pro installation industry product volume and service or renewal rate is such, that dwarfs today's Hi-Fi industry size. There, and in true HT, whatever works is adopted. 2 distributed WB subs and a narrow band bigger one plus a couple of dsp cross channels are peanuts as an economic or technical requirement in clubs and dedicated HT rooms.

Agreed - and I don't even use the DSP. The subs are self powered so the X-overs are built in. The whole thing is quit practicle.

 

[MaVo]

Gedlee, is your method of multiple subs similar to the one described in the harman paper? Link is below. Simply putting 4 small subs instead of one big one into the room seems to be a great idea.

http://www.harman.com/wp/pdf/multsubs.pdf

 

[gedlee]

Gedlee, is your method of multiple subs similar to the one described in the harman paper? Link is below. Simply putting 4 small subs instead of one big one into the room seems to be a great idea.

http://www.harman.com/wp/pdf/multsubs.pdf

It is similar, but in a letter to the JAES I commented on several aspects of the Welte paper that I disagreed with He ONLY tested symetrical locations on a symetrical grid of data. This is a mistake.

I did a similar study where I used a random placement against a random (although centered in the space) data grid. Welte's and my results agreed "almost" completely, except that I found that random placement worked as good with three sources as his did with four placed in the symetrical locations, as long as a few rules were kept:
1) place the source as far away from each other as possible
2) one source should be in a corner
3) one source should be above the rooms midpoint

I have done this in several rooms and when the gains, cutoffs and phases are properly set an extremely smooth LF field will result. Each sub is only a 10" or 12" driver in a bandpass box - very small, about 24 x 24 x 18.

I sometimes use a very low freq sub to hit the lowest octave that the mains miss. This sub is larger. But the monopole subs do get down pretty low in any case.

 

[john k...]

As an AES reviewer I find this offensive. I have never let a poor paper through. This position isn't fair to assume.

You just don't seem to want to understand my point of view so I guess I'll leave it there.

I'm sorry you find what I said about the review process offensive. But it is the truth. I never said anything about you, just the review process in the technical field, not particularly the AES journal. I served as a reviewer on numerous engineering and scientific journals during my career, including the AMSE Journal of Fluids Engineering and the AIAA Journal. Ultimately what gets published and what doesn't get published is the editor's decision. And as much as you may disagree and object, it is an old boys club and the editor can pass through or reject any paper he likes, with or without review or in spite of the review (positive or negative). Being invited to be a reviewer is part of the political process, nothing more. Been there, done that.

I don't see why you should take any of this personally, it’s just the process. . As I said I would gladly consider looking at your paper. I have no problem there. I just don't feel that it necessarily follows that a paper must be reviewed to be technically competent. 95% of the technical reports I authored under government contract were never reviewed. Many of the classic references in the fields I worked in never made it past the meeting presentation stage. Are they therefore lacking in technical merit? Rather than be offended by my statements about the review process maybe you should look at it from the other side. You comments imply that the work of Backman and Salmi, since were apparently not reviewed, are therefore technically of lower quality than your work. That is certainly a slap to those researchers, IMO, don't you think? Under the correct circumstances the review process may imply technical competence, but the lack of a review does not imply a lack of competence.

You have repeately referred to your work here yet have failed to provide a link or suitable reference to it (unless I missed it). So please, supply me a reference to your work. Return the favor. I gave you the info on the reference you requested.

And this and all my comments have nothing to do with not wanting to understand you work. I understand it. I agree with it. I just have a difference of opinion as to the partiality of it. Neither your, nor my points of view are the only ones acceptable. I see no reason to become defensive on this. What you fail to accept is that maybe someone doesn’t want the complexity of 4 or 5, or 10, or what ever number of woofers distributed around the room, with associated wiring, amplifiers... whatever it takes to achieve you solution, not matter how (un)intrusive it is. And maybe, just maybe there are those who just enjoy the fact that their woofer system excites a 25Hz mode in their room and shakes the H out of it.

 

[gedlee]

You have repeately referred to your work here yet have failed to provide a link or suitable reference to it (unless I missed it). So please, supply me a reference to your work. Return the favor. I gave you the info on the reference you requested.

What you fail to accept is that maybe someone doesn’t want the complexity of 4 or 5, or 10, or what ever number of woofers distributed around the room, with associated wiring, amplifiers...

I will have to look up that reference. It was published in the JAES sometime about 1999.

You don't seem to even read what I write - I said 3 subs not "4 or 5, or 10" I don't see where your gross exageration of the issues benefits the discussion.

 

[john k...]

...I found that random placement worked as good with three sources as his did with four placed in the symetrical locations, as long as a few rules were kept...

Each sub is only a 10" or 12" driver in a bandpass box ... about 24 x 24 x 18.

You don't seem to even read what I write - I said 3 subs not "4 or 5, or 10" I don't see where your gross exageration of the issues benefits the discussion.

I apologize for missing that detail. However, I am conflicted by this and your previous statements.

a) First you say you can place the subs anywhere. b) You impose rules for placement.

c) Then, big subs aren't required, and since they are small they hide easily, and d) Each sub is only a 10" or 12" driver in a bandpass box ... about 24 x 24 x 18.


I guess we have differences in opinions of what big is and what random placement and anywhere mean. In my 19' x 16' room I would not consider three subs that size to be small nor easy to hide, particulary if one has to be above 1/2 the room heighy. But granted, that is just a difference in perspective.

 

[gedlee]

I apologize for missing that detail. However, I am conflicted by this and your previous statements.

a) First you say you can place the subs anywhere. b) You impose rules for placement.

Why does everything have to be so black and white? Whats wrong with saying "Random with some constraints"? Is there only totally random and totally specific? Discussions with you seem to always get down to this either/or scenario. Engineering is never like that and it really makes discussion difficult.

 

[MaVo]

Hi gedlee, i found a thing called double bass array, which is supposed to completely nullify the modal problems in a room. There seems to be only a german paper on it, but i found a link to an english forum with a description of one specific implementation of it.

http://www.avsforum.com/avs-vb/showthread.php?t=837744&page=1&pp=30

Here is a link to the paper, written by someone called Anselm Goertz. It is in german, but the pictures are quite self explanatory. It gets interesting starting with page 9.

http://www.sennheiser.com/klein-hummel/globals.nsf/resources/tmt2002.PDF/$File/tmt2002.PDF

May i ask, what you think about this? The single bass array is related to your approach, since multiple small subs are placed in the room, but the rules of placement are totally different.

 

[gedlee]

Hi gedlee, i found a thing called double bass array, which is supposed to completely nullify the modal problems in a room. There seems to be only a german paper on it, but i found a link to an english forum with a description of one specific implementation of it.

http://www.avsforum.com/avs-vb/showthread.php?t=837744&page=1&pp=30

Here is a link to the paper, written by someone called Anselm Goertz. It is in german, but the pictures are quite self explanatory. It gets interesting starting with page 9.

http://www.sennheiser.com/klein-hummel/globals.nsf/resources/tmt2002.PDF/$File/tmt2002.PDF

May i ask, what you think about this? The single bass array is related to your approach, since multiple small subs are placed in the room, but the rules of placement are totally different.

First comment - way to complex for practical implimentation. With my approach all you need is a classic self power sub, plug it in the nearest wall outlet and run the LFE channel signal out to it. Setup of the LF cutoff freq, gain and phase is a little more complicated, but nothing like in the paper shown.

Second - the idea should work in theory, but in a real room with non rigid walls and furniture all around the simple parallel wall assumption is no longer valid and there is no telling how well this approach may or may not work.

Hence - if you have lots of time to develop the idea and lots of money to pay for it, it just MIGHT pan out to be something. But if you are concerned about your time and how much money you are spending then I don't recommend going this way.

 

[MaVo]

True, to use eight subs instead of three and still have less output doesnt sound like an advantage. I am still curious if the performance of both approaches, assuming both are setup in an optimal way, would be different. I wonder if they actually do something completely different to arrive at a similar goal. But my knowledge isnt sufficient to figure it out. Well, thats were trial and error comes into the game

 

[audiokinesis]

David Griesinger advocates using two subs placed to the extreme left and right of the listening position, with 90 degrees of phase difference between them, apparently in an attempt to simulate an interaural phase difference in recordings where the low frequency information is in mono (which is the vast majority of recordings). The idea being that an interaural phase difference at low frequencies gives a sense of spaciousness, or "envelopment' as he calls it. I tried a Griesinger setup, and I'm not sure that I could hear an improvement from the 90 degrees of phase difference.

But one time I was doing a multisub setup (four small subs) and we got this wonderful sense of being immersed in the recording that seemed to draw us into the music (speakers were modded Maggie SMG's). The next day, I discovered that I'd accidentally wired the two subwoofers on the right-hand side of the room in reverse polarity relative to the ones on the left (180 degrees of phase difference, from left to right, instead of only 90 degrees of phase difference). When I "fixed" it, the sense of spaciousness seemed to diminish. There was little if any difference in the percieved quantity and quality of bass between the two, which surprised me quite a bit - I would have expected having half of the subs 180 degrees out of phase with the other half would have resulted in a lot of cancellation, but if so I couldn't hear it.

Later I came across a suggestion by Jenna Crock to use four spread-out subwoofers with an X-polarity configuration - that is, left front and right rear in one polarity, and right front and left rear in the other polarity. I presume the focus here is on smoothing the in-room bass, rather than synthesizing a sense of envelopment.

I think Magnetar once described a distributed multisub system he was using wherein most of the subs had the same polarity but one was in reverse polarity.

Earl, do you have any thoughts on "alternative polarity" multisub configurations?

Thanks,

Duke

 

[gedlee]

I tried a Griesinger setup, and I'm not sure that I could hear an improvement from the 90 degrees of phase difference.

Earl, do you have any thoughts on "alternative polarity" multisub configurations?

Thanks,

Duke

How do you get 90° of phase shift? This is not easy to do. Unless your amps have that "phase" control, which I was never sure did what it said it does. To get a constant phase shift at ALL frequencies is not a simple thing and likely not doable with analog. I suspect that the amps have an all-pass network and this changes the phase, but NOT with a constant phase shift - it will vary with frequency. So please be careful that you are sure that you are doing what you say you are doing.

In my experince there is no one way to set the phase. I do it by measurements and its always different and never absolutely either a benefit or a detriment. I think that one of the ideal things about having multiple subs with various controls is the vast increase in the number of degrees of freedom that one gets to play with. Well, perhaps "play" is not the right word since everyone will run out and "tweak" the LF response by ear (and claim that Earl said to do that!).

I really do mean to write up my procedue for seting up subs since Kenny, who has built hundreds of sound installations, asked me to do that because he was so impressed with the results that I could achieve. There is likely no way that someone could randomly find the optimum set of values for the phase, gain and LF cutoff settings of the sub amps. It takes a well thought out procedure, some equipment and some time. But its well worth the effort.

Duke, you have probably not heard my system using all my own subs and the procedure that I am talking about. Its really quite effective.

 

[audiokinesis]

Earl, I was using a pair of subwoofer amps that had a control labelled "0-180 degrees", and cranked the control of one of them about halfway around. I have no idea whether it worked as advertised or not - indeed, until reading your post I had no idea that its efficacy was problematic.

Do you see any benefit to interaural phase differences at low frequencies, synthesized or not?

I'd be very interested in reading about your subwoofer set-up procedure. If you decide to save it for a book, I can guarantee you at least one sale.

Duke

 

[gedlee]

Do you see any benefit to interaural phase differences at low frequencies, synthesized or not?

I'd be very interested in reading about your subwoofer set-up procedure. If you decide to save it for a book, I can guarantee you at least one sale.

Duke

I know that David Griesinger claims that this works, but I also know that this recommendation is based on experiemnts with a sample size of one - himself - and not done blind. While it may be plausible, I'm not about to jump on the bandwagon based on that data.

I know that I can smooth out the frequency response with multiple subs and my setup technique better than anything else that I've ever used, but I don't know if this improves (lowers) the interaural cross correlation. I really have trouble seeing how the cross correlation can be anything but 1.0 at these very low frequencies and so at best we are talking about a few % change - if at all. And statistical significance of small changes like this is highly questionable.

It is the book idea that I was thinking about.

I'm thinking of posting all of my Home Theater book chapters on the web. Things in this area change so fast that its already out-dated and won't sell much more. Do you think that this would be a good idea?

 

[john k...]

Why does everything have to be so black and white? Whats wrong with saying "Random with some constraints"? Is there only totally random and totally specific? Discussions with you seem to always get down to this either/or scenario. Engineering is never like that and it really makes discussion difficult.

In my experince there is no one way to set the phase. I do it by measurements and its always different and never absolutely either a benefit or a detriment. I think that one of the ideal things about having multiple subs with various controls is the vast increase in the number of degrees of freedom that one gets to play with. Well, perhaps "play" is not the right word since everyone will run out and "tweak" the LF response by ear (and claim that Earl said to do that!).

I really do mean to write up my procedue for seting up subs since Kenny, who has built hundreds of sound installations, asked me to do that because he was so impressed with the results that I could achieve. There is likely no way that someone could randomly find the optimum set of values for the phase, gain and LF cutoff settings of the sub amps. It takes a well thought out procedure, some equipment and some time.

Earl, I don't want to argue with you. Things need not be black and white, but the shades of gray need to be defined. You introduce new information in new posts. First you say multiple small sources positioned randomly. Then you say well, not so random and not so small. At last count we have multiple sources (at least 3), with somewhat specific positioning, and further constrained by adjustments of amplitude, phase and LF cut off, resulting is a vast increase in the number of degrees of freedom of the system for which, in your own words, “There is likely no way that someone could randomly find the optimum set of values for the phase, gain and LF cutoff settings of the sub amps. It takes a well thought out procedure, some equipment and some time." This is a long way form monopole woofers are superior to dipoles or cardioids. I certainly have an interest in what you are doing, but buy your own words it isn't something that average DIY'er ot stereo/HT owner is going to be able to acomplish by themselves.

 

[Salas]

gedlee

''I'm thinking of posting all of my Home Theater book chapters on the web. Things in this area change so fast that its already out-dated and won't sell much more. Do you think that this would be a good idea?''


Yes. Do it.

 

[audiokinesis]

Thanks for the additional information, Earl.

On the Home Theater book... well, if you have a fair number of copies left, you might consider having a sale. Some of the specific video-system-related information might be outdated, but the acoustic and psychoacoustic information and room isolation information is still currrent, isn't it?

Maybe post a couple of chapters, and see if that boosts sales?

I have nothing against you posting it for free of course, and a lot of people would probably appreciate that.

Duke

 

[Brett]

gedlee

''I'm thinking of posting all of my Home Theater book chapters on the web. Things in this area change so fast that its already out-dated and won't sell much more. Do you think that this would be a good idea?''


Yes. Do it.

I'd like that too please.

 

[gedlee]

You introduce new information in new posts.

Hopefully all new posts have new information.

“There is likely no way that someone could randomly find the optimum set of values for the phase, gain and LF cutoff settings of the sub amps. It takes a well thought out procedure, some equipment and some time." This is a long way form monopole woofers are superior to dipoles or cardioids. I certainly have an interest in what you are doing, but buy your own words it isn't something that average DIY'er ot stereo/HT owner is going to be able to acomplish by themselves. [/B]

New Information!

Its not all that hard to do and any DIY could do it with nothing more than ARTA for example (any RTA) - hardly out of reach for anybody. By themselves - no, I'll explain the procedure, its not difficult once its explained to you.

And you don't set gains and phase of the amps for cardiods and dipoles to optimize the room? Thats a shame, its very useful.

You are trying to make it sound like "Just buy ONE of MY subs, throw it in any room anywhere, don't worry about setting it up at all and you will get great bass." Is that about right?