Some AR2ax measurements

[Zilch]

Only that you seem to be missing an opportunity to try and define what is so uniquely GOOD with the AR 2AX or 3a. I don't have to know more about the science and theory to see that your process misses this critical point of all the folks who keep telling you they prefer AR's for their unique "musicality". And it appears to be because you ignore the qualities that make these original AR's so good and choose to focus instead on their flaws.

I have said what I believe it might be several times on two forums now, Shacky.

If you weren't so focused upon bashing the messenger in both of them, and actually attempted to understand my posts, you wouldn't be beating this dead horse ad infinitum. We're well beyond it, now, actually.

You also know what my mains are, so stop fraudulently asserting that I'm "withholding" anything from you in this regard. And, "No," you do not have my permission to reveal that information, disclosed to you in confidence, here....

In summary, it is clear that the establishment of a subjective preference for the sound from a loudspeaker incorporates aspects of both sound quality and spatial quality, and there are situations when one may debate which is more important. The results discussed here all point in the same direction: that wide-dispersion loudspeakers, used in rooms that allow for early lateral reflections, are preferred by listeners especially, but not exclusively, for recreational listening. There appear to be no notable sacrifices in the "imaging" qualities of stereo reproduction. Indeed, there are several comments about excellent image stability and sensations of depth in the soundstage.

It would seem that what listeners like about AR3a in reverberant spaces is the apparent source width enhancement contributed by wide lateral dispersion, to the extent AR3a actually provides it. Given the facility to vary that, listeners will adjust it to extremes well beyond any realistic scale.

[genek]

All I want to know is do you (can you) appreciate the unique sound of the AR"s so you'll be able to chase it with your science. If all you can do is compare it to your econo-wave gold standard then I'm not very interested in your work.

It really shouldn't be necessary for someone collecting a properly-staged dataset of audio gear to "appreciate" the unique sound of anything, just to recognize it. I doubt that anyone conducting audio tests of exploding movie death stars really "appreciates" the sound when playing it back repeatedly for several hours at a time.

It would seem that what listeners like about AR3a in reverberant spaces is the apparent source width enhancement contributed by wide lateral dispersion, to the extent AR3a actually provides it. Given the facility to vary that, listeners will adjust it to extremes well beyond any realistic scale.

Oops. There's another one of those "hypothesis/assumption" conclusions you didn't remember making, Zilch. I forgot about that one too.

[Zilch]

Oops. There's another one of those "hypothesis/assumption" conclusions you didn't remember making, Zilch. I forgot about that one too.

The "It would seem" part, or the "Listeners will adjust" part, neither of which is derived from my data, rather, merely express how I believe Toole's analysis might be applicable to AR3a?

Parse my dog eating marshmallows post, please. I'm SURE there's a hidden agenda to be discovered in there somewhere....

[genek]

The "It would seem" part, or the "Listeners will adjust" part, neither of which is derived from my data, rather, merely express how I believe Toole's analysis might be applicable to AR3a?

It's the belief that is being expressed that is the hypothesis. The "it would seem" firmly establishes it as a hypothesis, so it's not a hidden agenda, Ideally, it would be followed by some sort of experiment to see if doing something to suppress the phenomenon has a negative effect on some AR lovers' opinion of the sound, and without such a test of the hypothesis, it just sits there as another rhetorical landmine.

[Zilch]

It's the belief that is being expressed that is the hypothesis. The "it would seem" firmly establishes it as a hypothesis, so it's not a hidden agenda, Ideally, it would be followed by some sort of experiment to see if doing something to suppress the phenomenon has a negative effect on some AR lovers' opinion of the sound, and without such a test of the hypothesis, it just sits there as another rhetorical landmine.

I how suggested how YOU might test it, actually, but you find the rhetoric far more engaging than actually DOing anything, apparently....

[genek]

I how suggested how YOU might test it, actually, but you find the rhetoric far more engaging than actually DOing anything, apparently....

If I should ever offer up my own hypothesis on how anything AR-related works I will consider it my responsibility to back it up with testing. If you are looking for other people to test your hypotheses for you, I can probably put you in touch with someone who is much better qualified than I am so you can get a quote for the job.

[kkantor]

I overstate it, primarily to highlight the fundamental conflict between Allison's perspective as disclosed in the 1970, 72 AES papers (to which some wagons here are apparently hitched,) and Toole's more contemporary view. As should be apparent, I believe there is some mileage yet to accrue from pursuit of the implications.

Understood... hope springs eternal! That's why I'm following this and commenting.

In some ways, Toole, et al, probably wouldn't exist were it not for the work of Allison, et al. But, of course, research that brings nothing new to the table is not very interesting. Clearly, you have become interested in investigating the differences in their work. (I wouldn't call them "conflicts," and I doubt Floyd would either. But, I have yet to read his book....) It's a valid approach to history.

(OTOH, Personally, I've learned over time never to show prototypes to management, and never to commit ongoing research to general press but, hey, it's your dime!)

Clearly, you are privy to a universe of more detailed information and insight than is possible to glean from what is documented and readily available in the CSP library, but even here, there are indications of more afoot, as example, in the allowance for the mid and tweeter responses to overlap in support of each other more than prudent design practice would recommend.

To a certain extent, some of it is simply my exposure to AR's tribal knowledge, rather than anything privy. I don't think anyone yet has figured out how to really document this stuff completely, in a whole variety of engineering fields. For example, when I undertook the development of the AR303, I had to face down the issues of what to keep about the old school, and what to try and improve upon, while preserving the lineage. I found it a very complex and challenging undertaking. I think there are notes in the library here. And, of course, "prudent design practice" is goal dependent.

And do you find my measurements and findings consistent with the more comprehensive work you and your team performed?

Your measurements are certainly good enough that they are consistent with your thesis, and so don't warrant any red flags. As anyone who has tried to communicate such things even between labs within one building can attest, perfectly definitive driver response curves, particularly at higher frequencies, are a very elusive thing.

I am not a student of AR3a or AR in general;

Good thing, or you would have to change your avatar....

my purpose was to verify RoyC's L-pad substitution via in situ measurements. However, even the most rudimentary assessment reveals issues worthy of discussion and pursuit, in my view....

A noble purpose!

-k

[Zilch]

If I should ever offer up my own hypothesis on how anything AR-related works I will consider it my responsibility to back it up with testing. If you are looking for other people to test your hypotheses for you, I can probably put you in touch with someone who is much better qualified than I am so you can get a quote for the job.

It's Toole's thesis, to which I subscribe. I merely invite you to verify for your own edification whether it is valid or not; the task is easily accomplished. I recall your wanting to know, is all.

[No marshmallows involved in this one.... ]

[Zilch]

In some ways, Toole, et al, probably wouldn't exist were it not for the work of Allison, et al. But, of course, research that brings nothing new to the table is not very interesting. Clearly, you have become interested in investigating the differences in their work. (I wouldn't call them "conflicts," and I doubt Floyd would either. But, I have yet to read his book....) It's a valid approach to history.

Less conflict than conflicting views of the same data, and, as you suggest, Allison's perspective dead-ended in the late seventies. Toole provides a credible explanation as to why we don't hear combing with no diffusively integrative reverberant field required, and I believe I can demonstrate from Allison's 1970 data alone (once it is posted here) that AR3a's exhibit rising and substantially higher directivity than "max dispersion" theory supposes.

Toole describes what wide dispersion does that we like, again, with no reverberant field required, much as Holt observed regarding Bose 901, and Geddes teaches how to accomplish it more effectively using contemporary constant-directivity waveguides without the compromising response anomalies so evident in these early AR designs. I will discuss the details further with you in person on your recommendation, having already disclosed more than sufficient here; I'm going with Geddes....

[genek]

It's Toole's thesis, to which I subscribe. I merely invite you to verify for your own edification whether it is valid or not; the task is easily accomplished. I recall your wanting to know, is all.

Villchur and Allison had their theories on how speakers worked and used them to make speakers that worked fairly well, at least IMO. If others come along to discount those theories, I consider it incumbent upon them to demonstrate that their theories work in a similar manner. Measurements are moderately interesting, but the only thing I would consider truly edifying is to hear those with the conflicting theories apply those theories to produce speakers that work better for me than the ones produced by the people whose theories they are discounting. That would be "validation."

[Zilch]

Villchur and Allison had their theories on how speakers worked and used them to make speakers that worked fairly well, at least IMO. If others come along to discount those theories, I consider it incumbent upon them to demonstrate that their theories work in a similar manner. Measurements are moderately interesting, but the only thing I would consider truly edifying is to hear those with the conflicting theories apply those theories to produce speakers that work better for me than the ones produced by the people whose theories they are discounting. That would be "validation."

You forget - nobody give a whit what works for you - it's your job to discover that....

[genek]

You forget - nobody give a whit what works for you - it's your job to discover that....

I am a consumer with sufficient disposable income to buy anything that pleases me, something that appears to be in rather short supply these days. Anyone who doesn't give a whit about what works for me can go get their revenue from somebody else.

In the process of developing new product, my "job" is to consider things that are offered to me, spend money on the ones I decide work for me and reject everything else. It's a good job, and I enjoy it immensely.

[Zilch]

In the process of developing new product, my "job" is to consider things that are offered to me, spend money on the ones I decide work for me and reject everything else. It's a good job, and I enjoy it immensely.

O.K., I'm going to Zen it:

I offer you nothing but what you might know; in this you will find pleasure.

Got a link to Beranek's model?

[genek]

O.K., I'm going to Zen it:

I offer you nothing but what you might know; in this you will find pleasure.

The Zen of GeneK:

Knowledge is more pleasing than money when both are in short supply; otherwise, he who dies with the best toys wins.

[soundminded]

I how suggested how YOU might test it, actually, but you find the rhetoric far more engaging than actually DOing anything, apparently....

The problem with your point of view as I see it Zilch is that there is no correlation between what yours or anyone else's measurements show and subjective accuracy, that is the ability of audio equipent to produce convincing re-creation of music (I'm talking serious music, not what passes for music in the cheapest, most banal, and commonest use of this term today) for people who do have good acoustic memories, who do know what the real thing sounds like, and who can listen critically and make critical judgements about what they hear, double blind if necessary. In fact there isn't even a working technical definition of what musical accuracy is and as I've pointed out in numerous threads, depending on the context, there is more than one answer to that question because the subject covers more than one problem.

AR's unique status whether anyone likes it or not is that it proved that its products at least on some occasions even if rarely and under highly contrived conditions produce sound that did seem to many listeners very similar to real music on direct comparison. No other manufacturer before or since has demonstrated that such a thing was possible with their products. If that is the goal of high fidelity, then there has not been progress in the last forty years. If making lots of money by convincing people to buy a new more expensive piece of equipment to replace one they already had that still functioned because they liked the new one better, then there has been lots of progress.

Reading the "theories" of the people who are in this business, it seems obvious from my perspective that in the larger context of this problem, they are by and large clueless. Since they can't even get a working definition of the problem, it is hardly surprising that they haven't solved it yet, they haven't come close. At the rate they are going, I have to wonder if they ever will. Their perpetual "experiments" which is all I can call their products are an endless parade of failed attempts at ever escalating prices to their customers, each one lauded as a breakthrough. In fact, from what I can tell, their advertising no longer even makes a pretense at soliving the basic problem, just claims for technical superiority whose signifigance or usefullness to their prospective customers they can't demonstrate. Needless to say, a cottage industry of magazines and blog sites as well as a plethora of other useless junk like exotic wires has sprung up around them, all with the sole purpose of making money. This site has been until now one of the rare refuges from that perversion.

The flawed theories of these merchants include people who made genuinely valuable contributions such as Villchur. His knowledge seems to me to have been intuitive rather than rigorously analyticial. This was hardly surprising since he was not trained as a physicist or an engineer. (Those who were trained as scientists and engineers have fared no better, probably because the most talented of them don't waste their time with such mundane and inconsequential problems.) For An example of this is Villchur's explanation of how what he should have correctly called a "pneumatic suspension" loudspeaker rather than an acoustic suspension loudspeaker using theories of thermodynamics is laughable. Not that it is wrong, it just doesn't explain it, it's the wrong tool. I've also pointed out on numerous occasions that Newton's second law of motion as applied to forced oscillation is the correct tool. Those who understand it will see it instantly, those who don't will not. Yes he got the right answer for the wrong reason which neither diminishes its validity nor its value. Yet it was not a scientifc breakthrough. It was an ingenious way to achieve an already possible goal with an economy of size and price at a sacrifice of efficiency since technology already knew how to build horn speakers that would equal and outperform AR1's LF response if size and cost were no object.

[Carlspeak]

Here's a link to a 5 yr old article in "Test and Measurement World" magazine. The author visited JBL's research labs to discover how they test speakers and correlate their results with listener preferences. It's a bit old but contains some interesting details and puts into plain english much of what has been discussed here. FYI....

http://www.tmworld.com/article/CA475937.html

[genek]

The problem with your point of view as I see it Zilch is that there is no correlation between what yours or anyone else's measurements show and subjective accuracy

For the past 30 years or so the audio industry from the low end to the high end appears to have been marching in lockstep toward the same holy grail of loudspeaker specs, and in that time most every new speaker I've heard has sounded pretty much alike to me (I don't know where I score on the "good acoustic memory" scale, but I can remember the relative level of pleasure or irritation I experience from one listen to another). There must be some measure of correlation in that somewhere.

[kkantor]

Less conflict than conflicting views of the same data, and, as you suggest, Allison's perspective dead-ended in the late seventies. Toole provides a credible explanation as to why we don't hear combing with no diffusively integrative reverberant field required, and I believe I can demonstrate from Allison's 1970 data alone (once it is posted here) that AR3a's exhibit rising and substantially higher directivity than "max dispersion" theory supposes.

Toole describes what wide dispersion does that we like, again, with no reverberant field required, much as Holt observed regarding Bose 901, and Geddes teaches how to accomplish it more effectively using contemporary constant-directivity waveguides without the compromising response anomalies so evident in these early AR designs. I will discuss the details further with you in person on your recommendation, having already disclosed more than sufficient here; I'm going with Geddes....

Just for giggles, have a look at this article I wrote for High Fidelity in 1986. It's not my best.... just a contract consumer, piece, but I did plug Floyd's pre-Harman work in it.

http://www.kenkantor.com/publications/spea...ker_testing.pdf

I think it is great to keep the discussion of >issues< alive here. I think conclusions, data, even strong perspectives of the researcher's goals and opinions, are best left until the work is done. So, of course it would be good to illuminate the work of Geddes, et al. (Might be best in a Mods and Tweaks area though, since it is academic and broadly applicable subject, not particularly related to AR.

Just so you know, I have talked about this with Earl a few times over the last couple of years. Mostly in private emails, but also on a couple of BBS's, and the AES Loudspeaker Technical Committee "reflector." We also investigated waveguides for a while at Tymphany/Vifa, after we hired one of the lead designers of the Mackie monitor. This, in turn, brought me some time meeting with Cal Perkins. (A name I think you probably know.) So, while I have never chosen a waveguide for one of my own designs, I am generally familiar with them, and understand their advantages.

-k

OK, really, I have to get back to Faceb... I mean Work, now!

[Zilch]

AR's unique status whether anyone likes it or not is that it proved that its products at least on some occasions even if rarely and under highly contrived conditions produce sound that did seem to many listeners very similar to real music on direct comparison. No other manufacturer before or since has demonstrated that such a thing was possible with their products. If that is the goal of high fidelity, then there has not been progress in the last forty years.

Many of today's modern "lock-step" designs can do as well as or better than vintage ARs in contrived live vs. recorded demos. Contrary to common perception, loudspeaker design has significantly advanced over the last 40 years, to the point that even mini-systems from Best Buy can do it. You ascribe far more significance to AR's LVR "performances" than they deserve.

AR's design intent was to replicate the performance of acoustic music in concert halls, and as consequence, they adulterate the music in well-understood ways to accomplish that. If that's your standard of "realism," your bias is clear, but recoginize that:

1) It does not translate well to contemporary listening practices and tastes, and,

2) It can be accomplished as well or better using contemporary technology.

[P.S.: I provided the Beranek model at #233, above.... ]

********

I think it is great to keep the discussion of >issues< alive here. I think conclusions, data, even strong perspectives of the researcher's goals and opinions, are best left until the work is done. So, of course it would be good to illuminate the work of Geddes, et al. (Might be best in a Mods and Tweaks area though, since it is academic and broadly applicable subject, not particularly related to AR.

Geddes's approach to system design provides insights relating to loudspeaker/room interaction, which is highly relevant to this discussion, in my view. Here again is a link:

http://www.gedlee.com/downloads/Cum%20laude.pdf

[genek]

2) It can be accomplished as well or better using contemporary technology....

When is the listening demo for #2? I'll bring chips.

[soundminded]

Many of today's modern "lock-step" designs can do as well as or better than vintage ARs in contrived live vs. recorded demos. Contrary to common perception, loudspeaker design has significantly advanced over the last 40 years, to the point that even mini-systems from Best Buy can do it. You ascribe far more significance to AR's LVR "performances" than they deserve.

AR's design intent was to replicate the performance of acoustic music in concert halls, and as consequence, they adulterate the music in well-understood ways to accomplish that. If that's your standard of "realism," your bias is clear, but recoginize that:

1) It does not translate well to contemporary listening practices and tastes, and,

2) It can be accomplished as well or better using contemporary technology.

[P.S.: I provided the Beranek model at #233, above.... ]

********

Geddes's approach to system design provides insights relating to loudspeaker/room interaction, which is highly relevant to this discussion, in my view. Here again is a link:

http://www.gedlee.com/downloads/Cum%20laude.pdf

"Many of today's modern "lock-step" designs can do as well as or better than vintage ARs in contrived live vs. recorded demos. Contrary to common perception, loudspeaker design has significantly advanced over the last 40 years, to the point that even mini-systems from Best Buy can do it. "

Then why don't they prove it? I say they can't.

"AR's design intent was to replicate the performance of acoustic music in concert halls, and as consequence, they adulterate the music in well-understood ways to accomplish that."

Rolling off high frequencies does not replicate the performace of music in concert halls. That was just plain naive. BBC still does it but then what would you expect from them? The analysis of that problem is well beyond the current state of the art, so is figuring out how to solve it. That shortcoming in AR's speaker was compensated for in the demo with a preamplifier treble control. It is the results which count and I know what I heard, in fact twice. And I was surprized. Perhaps if I heard it again, I'd be more critical as I have become a far more discerning listener than I was many years ago. Listening is different from hearing and I think within the bounds of hearing ability it can be learned, some have greater potential than others. AR demonstrated on those two occasions that within the limits of my hearing and listening ability that it could produce from a recording it made itself the sound of a guitar and a Nickelodeon from a pair of loudspeakers in a relatively small room. AFAIAC, that is still the current state of the art or should I say was as it seems to have regressed, not progressed. Neither AR nor anyone else has demonstrated the ability to make sound in a small room sound like it was in a very large room, say close to a million cubic feet. Not even in their dreams. Not to anyone with normal hearing.

[genek]

I have never had a chance to evaluate the 303, but I know of at least one person who has and he believes it is an exemplary design. I believe that Julian Hirsch felt pretty much the same way, and I think you can be proud of that speaker. No doubt, you still consider it to be a really fine performer. It can probably hold its own with any number of contemporary designs.

I listened to some in showroom and at home about 10 years ago. At the time I had only my 2ax's and 6's to compare them to, no 3a's, but I thought the sound was every bit as good as I remembered from hearing 3a's back in the 70's. We have since learned that the tweeter dispersion is not quite the equal of the previous model in a direct comparison, but perhaps the redesigned grill compensated by reducing the impact the 3a's picture frame front had on its tweeters. What ultimately deterred me from buying that pair and put me on the path toward acquiring a pair of 3a's was the fact that the only 303 finish available to me at that time was the black laminate; if they had been the rosewood veneered version I never knew existed until after the model was discontinued I probably would have gone for them.

[Zilch]

See Toole p. 325 where he clearly states that "Uniformly dispersion over a huge horizontal angle -- almost +/- 90° -- is required to deliver similarly good sound in the direct and reflected pathways [of surround speakers] to all of the listeners, not just the sweet spot in the center of the audience."

My suggestion in Howard's measurements thread that LSTs or Allisons might be worthy candidates was neither capricious nor gratuitous....

[soundminded]

I listened to some in showroom and at home about 10 years ago. At the time I had only my 2ax's and 6's to compare them to, no 3a's, but I thought the sound was every bit as good as I remembered from hearing 3a's back in the 70's. We have since learned that the tweeter dispersion is not quite the equal of the previous model in a direct comparison, but perhaps the redesigned grill compensated by reducing the impact the 3a's picture frame front had on its tweeters. What ultimately deterred me from buying that pair and put me on the path toward acquiring a pair of 3a's was the fact that the only 303 finish available to me at that time was the black laminate; if they had been the rosewood veneered version I never knew existed until after the model was discontinued I probably would have gone for them.

How fortunate we are that a thin strip of wood molding around the outside of a speaker baffle can't really affect what people hear from it. If it could, just imagine what the effect would be on the sound of a violin or a cello by the presence of something as acoustically reflective as.........a music stand

[genek]

How fortunate we are that a thin strip of wood molding around the outside of a speaker baffle can't really affect what people hear from it. If it could, just imagine what the effect would be on the sound of a violin or a cello by the presence of something as acoustically reflective as.........a music stand

A 1.5" deep recess is not exactly what I would call "thin," but it was just a wild guess, since I didn't have a 3a at the time to compare the 303 to.