Some AR2ax measurements

[Zilch]

I continue to remind the world that if the AR-3 speaker was as "dull" and inadequate sounding as some say (this includes Toole in his book, and one or two posters here) it would not have been able to do the work it did at those demos. Even those who might have heard differences at times would have to say that those 40+ year old speakers were remarkably good at replicating the sound of the live performers, even by modern standards. I think that now a number of other speakers could do as well, but in that case what we must say is that the art of speaker design and construction has not really progressed all that much since the AR-3/AR-3a era. There have been improvements, but none that are overwhelming.

Spoken like a true antiquarian. The same could be said of the the AR live demos vs. Edison's, 50 years earlier:

Live versus reproduced comparison demonstrations were also conducted by RCA in 1947 [using a full symphony orchestra (Olson, 1957, p. 606)], Wharfedale in the 1950s [briggs, 1958, p. 302], Acoustic Research in the 1960s, and probably others. All were successful in persuading audiences that near perfection in sound reproduction had arrived. Based on these reports, one could conclude that there had been no consequential progress in loudspeaker design in over 50 years.

It was grandstanding -- a contrived marketing ploy -- and citing it as definitive evidence of anything substantive is tenuous, at best....

[Shacky]

It was grandstanding -- a contrived marketing ploy -- and citing it as definitive evidence of anything substantive is tenuous, at best....

I wouldn't go that far. It certainly was substantive in that it showed what was possible with audio equipment. Regardless of source recording and or treble boost. Fact that an audio system could recreate sound that aproximated the real thing was a technological feat. Just like the earlier examples, bringing anything close to the sound of the original to the masses was a huge step forward for society - delivering experiences once only available to kings.

Regardless of whether or not you think AR speakers are the bee's knees, you have to admit impressing the critics in that audience was a job very well done. And their speakers sounded good enough to garner one third of the speaker marker at their height.

Hard to argue with that success. Will never be repeated.

[genek]

Roy Allison mentioned to me that the EQ Villchur used generally was a roughly 5 dB boost from the treble tone control of the Dynaco preamps that were used.

Sounds perfectly consistent with the advice provided in the mimeographed AR instructions on how to set your 3a's and 5's for "uniform energy output" if desired.

[Zilch]

Baloney. The padding, by absorbing more at higher frequencies than at lower frequencies, reduces the total-system treble and upper-midrange energy reaching the listener, while not reducing the bass and lower-midrange energy reaching them, which means that the speaker will have a more muted sound at higher frequencies. The spectral balance tilts downward.

I am merely reciting your own earlier argument; damping moots the wide dispersion and kills the very reverberant field upon which your thesis relies, effectively narrowing the directivity, and allowing the direct field to become more dominant in what the listener hears. Is the direct field "brighter" or "duller" than the reverberant one?

Tell us, please, Howard, how the reverberant field spontaneously generates augmented high-frequency response and shifts the spectral balance above the transition frequency? It's quite possible to have more prominent bass concurrent with highs that are less rolled off, and thus "brighter." Is your hypothetical damped listening space frequency selective?

[Zilch]

Regardless of whether or not you think AR speakers are the bee's knees, you have to admit impressing the critics in that audience was a job very well done. And their speakers sounded good enough to garner one third of the speaker market at their height.

Yes, of course, as marketing, but AR enthusiasts ascribe far more significance to this than it deserves:

Of a 1916 demonstration [by Edison] in Carnegie Hall before a capacity audience of "musically cultered and musically critical" listeners, the New York Evening Mail reported that "the ear could not tell when it was listening to the phonograph alone, and when thto the actual voice and reproduction together. Only the eye could discover the truth by noting when the singer's mouth was open or closed" (quoted in Harvith and Harvith, 1985, p. 12).

Edison's phonographs were wildly successful, as well, of course. Is that testament to their "accuracy" in any absolute sense?

Were the "Is it live or is it Memorex?" demos with Ella Fitzgerald shattering wine glasses somehow less convincing?

Is AR still making these speakers today like Bose is the 901? How long has it been since speakers designed according to these principles, let alone any employing the 40-year-old technology they incorporate, were even viable in the marketplace?

Again, I'm NOT bashing ARs here, rather, merely demonstrating that the common rationale for liking them is fundamentally flawed. Does that mean that we should not like them? Not at all, but instead, that there is value in ascertaining why this is so, for all that might be worth, in which the prevailing mythology counts for nothing until proven, and I can guarantee that "magic" is not among the essential elements, either....

[soundminded]

Yes, of course, as marketing, but AR enthusiasts ascribe far more significance to this than it deserves:

Were the "Is it live or is it Memorex" demos with Ella Fitzgerald shattering wine glasses somehow less convincing?

Is AR still making these speakers today like Bose is the 901? How long has it been since speakers designed according to these principles, let alone any employing the 40-year-old technology they incorporate, were even viable in the marketplace?

Principles pioneered by AR are still in widespread use today. The acoustic suspension principle is quietly used by many of the best speaker manufacturers while ferro fluid cooled dome tweeters are dominant at all price levels. It's easy to forget that before Acoustic Research came along, in order to produce deep bass you needed a gigantic cabinet. Even the giant Klipschorn was no match when it came to the lowest octave compared to AR1 and AR3. I'll bet virtually all fine turntables use the same clever suspension to mechanically isolate them from shock and feedback pioneered by AR and many use the split ring platter AR used as well (I don't know if Thorens had that first.) Certainly Empire copied both.

AR's demise came about for many reasons. Arrogance towards its distributors was one. Being taken over and run by a series of conglomerates was another. Hiring engineers and managers with an entirely different notion of what kinds of product to design was yet another.

Bose Corporations stores by and large don't sell many 901s anymore, most of them don't have them on display to demonstrate and have to special order them. It isn't considered one of the big money makers for them anymore. I heard a pair about a year ago at the Bose store in Fashion Island Shopping Center in Newport Beach Ca. They had the same dull rolled off treble IMO the prior versions had. Unfortunately, I didn't have time to test the bass with an organ disc the store had on hand.

Large horn speakers are used for theaters because they can produce very loud sound before they distort filling up a theater. Power and efficiency are not the consideration they once were for obvious reasons. The use of constant diretivity speakers which horn design lends itself to allows speakers to uniformly cover areas reducing loud and soft spots in the audience with the fewest number of units and if used for sound reinforcement, the most gain before feedback possible. The design using them in arrays has been reduced to a fairly exact science and is usually performed using canned software. I think it's sad that even at many classical music concerts, sound reinforcement systems are used. You might just as well stay at home and hear a recording. You aren't really hearing music anymore. It's even sadder to me that listening to recordings has become a subsitute for listening to music to the point where most people thinK that recordings are music.

[Shacky]

Is AR still making these speakers today like Bose is the 901? How long has it been since speakers designed according to these principles, let alone any employing the 40-year-old technology they incorporate, were even viable in the marketplace?

Marketplace viability is complex. First there's the sellout ~67 IIRC. Then there's the matter of the original AR's at their hieght being the medium cost supplier providing performance way above their price point. That could not continue with competitive pressure and fact that they were hand made. Even if AR didn't sell out, they would have had to make a choice. Decrease quality/cost to compete or gravitate towards higher end users willing to pay high price for hand made speakers.

I don't think it's fair to say "they aren't making them today so that means their characteristics are no longer marketable". If the auto industry re-made the '64 GTO would it sell today? It's price point - if you could get it past all new regs - would be prohibitive. One of the reasons the GTO was so successful was it's relatively low price point.

I'm not going back to our old argument. I just think this statement doesn't hold water..

[Zilch]

Large horn speakers are used for theaters because they can produce very loud sound before they distort filling up a theater. Power and efficiency are not the consideration they once were for obvious reasons. The use of constant diretivity speakers which horn design lends itself to allows speakers to uniformly cover areas reducing loud and soft spots in the audience with the fewest number of units and if used for sound reinforcement, the most gain before feedback possible.

And having heard constant directivity in your own listening room, you know it to be all but worthless, right?

The Acoustic Research AR-3 [Figure 17.2b] was famous for its novel acoustic suspension woofer, and it came to be one of the reference loudspeakers of that generation. Its acoustic performance was well documented in the literature (e.g., Allison and Berkovitz, 1972), which was a great credit to the company. A major design goal was to achieve constant directivity, and they did well; the only minor exceptions are the woofer beginning to beam as it approaches the crossover to the midrange and the tweeter at very high frequencies. The essential issue with this product is its frequency response, which significantly rolls off toward both low and high frequencies. Low-frequency output would be aided by boundaries (see Figure 12.5 for boundary interaction data). Almost all listeners found it to be slightly "dull" sounding, and some identified coloration around 1 kHz. [Emphasis added.]

Of note: Howard has stated that he will not read any Toole quotes I post because of these awful things Toole said about ARs....

Since, at least with wide-dispersion speakers, the bulk of the energy one hears with good speakers exists in the reverberant field (that you miss this point shows that you absolutely do not understand what Roy Allison said in his speaker/room performance papers) any reduction of energy in that field (in this case at upper midrange and treble frequencies) is audible - even if the energy in the direct field remains the same.

I do understand it to be as Speaker Dave pointed out, an "Old School" design principle long since discredited, the invalidity of which you can yourself prove with marshmallows.

[genek]

Yes, of course, as marketing, but AR enthusiasts ascribe far more significance to this than it deserves:

Edison's phonographs were wildly successful, as well, of course. Is that testament to their "accuracy" in any absolute sense?

Were the "Is it live or is it Memorex?" demos with Ella Fitzgerald shattering wine glasses somehow less convincing?

Is AR still making these speakers today like Bose is the 901? How long has it been since speakers designed according to these principles, let alone any employing the 40-year-old technology they incorporate, were even viable in the marketplace?

Again, I'm NOT bashing ARs here, rather, merely demonstrating that the common rationale for liking them is fundamentally flawed. Does that mean that we should not like them? Not at all, but instead, that there is value in ascertaining why this is so, for all that might be worth, in which the prevailing mythology counts for nothing until proven, and I can guarantee that "magic" is not among the essential elements, either....

You're not positing that "the common rationale for liking them is fundamentally flawed" because they're not viable in the marketplace are you? Because if you're arguing that popularity is a valid rationale for liking something I only have one thing to say in reply:

"Laverne and Shirley."

[Zilch]

As for the AR approach no longer being viable in the marketplace, this has more to do with modern musical tastes than technology, per se. When AR was an expanding concern (the Villchur/Allison era) magazines like Stereo Review regularly reviewed a large number of classical recordings in each issue of the magazine. This is no longer the case (the magazine itself is now gone), because musical tastes have warped to a point where serious accuracy with loudspeakers is no longer important. Instead, given the music most people listen to, the kind of speakers you prefer are preferred. So, it is not that the AR model is flawed, it is that listener tastss now dictate a lower-common demoninator kind of sound that suits you but does not suit those who prefer live acoustic sound to an electronic blast wave.

We are getting to the truth here, in more ways than one.

Once again, you confuse realism with accuracy. Accuracy is well-defined, and measurable -- input = output.

Realism is subjective, though in broad terms, it's still determinate, and the reality of the concert hall, which for you is apparently the only "worthy" objective, the one Villchur and Allison were attempting to simulate, is inherently inaccurate, and no less so than that a rocker most enjoys. If a loudspeaker incorporates a sonic coloration to optimize reproduction of any particular perception of realism, it is, virtually by definition, ill-suited for all others, and thus, ARs are generally known to be "Classical music" speakers, and West Coast designs, "Rock" speakers; both are inaccurate, each in their own way.

Accurate speakers, on the other hand, are equally suited to both, and the entire spectrum between, key to their inherent versatility being smooth, flat frequency response and uniform power response -- EQ to taste....

[genek]

Accurate speakers, on the other hand, are equally suited to both, and the entire spectrum between, key to their inherent versatility being smooth, flat frequency response and uniform power response -- EQ to taste....

I just hope that "equally suited to both" isn't really "equally unsuited to both."

But if the only thing that makes one type of speaker better suited to a particular type of music when at its default setting is its frequency balance (which would seem to be your arguement, since you dismiss the effects of reverberant fields), then it should be feasible to EQ either type of speaker to suit either either type of music. Assuming, of course, that one hasn't painted oneself into a corner by refusing to buy anything with tone controls.

[Zilch]

You're not positing that "the common rationale for liking them is fundamentally flawed" because they're not viable in the marketplace are you?

No, I cite it as counterpoint to the argument that they are good because they were once a major presence in the marketplace.

Frankly, what I find intriguing about this debate is that subjectivists should require any rationale at all; Shacky has it right from that perspective, in my view: a resolute "I like 'em" is sufficient.

[Apparently not.... ]

But if the only thing that makes one type of speaker better suited to a particular type of music when at its default setting is its frequency balance (which would seem to be your argument, since you dismiss the effects of reverberant fields), then it should be feasible to EQ either type of speaker to suit either either type of music. Assuming, of course, that one hasn't painted oneself into a corner by refusing to buy anything with tone controls.

There's more to it than that, of course, but once the variables are understood, it's certainly more efficient than a lifelong quest for the "perfect" system/room synergy to satisfy a particular preference when the medium itself falls inherently well short of the goal, and our preferences are themselves subject to change as early as the next time we listen....

[genek]

No, I cite it as counterpoint to the argument that they are good because they were once a major presence in the marketplace.

I think the arguement is actually the other way around, that they became a major presence in the marketplace because they were good. But of course, the counter to that one is still "Laverne and Shirley."

Frankly, what I find intriguing about this debate is that subjectivists should require any rationale at all; Shacky has it right from that perspective, in my view: a resolute "I like 'em" is sufficient.

[Apparently not.... ]

Covered that one already (see the post above on marketing brainwashing and "good sound must equal accurate"). Nevermind the fact that Villchur and Co told us a long, long time ago what the required settings for "accurate" were and warned us that we probably wouldn't want to use them until "someday" in the future. Which, of course, it now is.

There's more to it than that, of course, but once the variables are understood, it's certainly more efficient than a lifelong quest for the "perfect" system/room synergy to satisfy a particular preference when the medium itself falls inherently well short of the goal, and our preferences are themselves subject to change as early as the next time we listen....

I think my "perfect" speaker would be one that has the wide dispersion (for those who want it; those who think it's irrelevant can just ignore it) and has a wide enough range of MF and HF adjustments to enable it to switch back and forth between "live classical" and "in your face rock" settings (didn't somebody already make a speaker at one time whose balance control switch said "Classical/Rock" rather than "Low/High?")

So when are the WG trials?

[Zilch]

Sounds perfectly consistent with the advice provided in the mimeographed AR instructions on how to set your 3a's and 5's for "uniform energy output" if desired.

Only in AR's indeterminate reverberant chamber, not in actual home listening spaces, and forunately so, as with rising directivity, flat power response yields rising frequency response, which is decidedly too "hot."

Though the mimeoed supplemental instruction sheet speaks of both flat anechoic frequency response and uniform energy output, the two may be related, but are not equivalent. That sheet DOES clearly indicate, however, that constant directivity was in fact a prime AR design objective, as Toole observed, above.

http://www.classicspeakerpages.net/library...vel_contro.html

>*flush*<

I think my "perfect" speaker would be one that has the wide dispersion (for those who want it; those who think it's irrelevant can just ignore it) and has a wide enough range of MF and HF adjustments to enable it to switch back and forth between "live classical" and "in your face rock" settings (didn't somebody already make a speaker at one time whose balance control switch said "Classical/Rock" rather than "Low/High?")

Easily accomplished, if we start with "accurate" as the baseline....

[Zilch]

But of course, the counter to that one is still "Laverne and Shirley."

Edison's not bad, either.

So when are the WG trials?

The E'Waved AR2xs are spot on this thread for comparison, but I didn't build them, so that'll have to wait 'til they come in for testing.

Parts to update the AR3s are here, so I'll know how the pair sounds and measures soon, maybe this weekend. Even though that's not relevant, I'll link to it when it happens as a matter of potential interest.

I can probably do some preliminary polars on the A60's this weekend, too, and I'll link that in Carl's waveguide thread and y'all can compare those results, to the AR3a's as well as Speaker Dave's AR2ax's, at least insofar as directivity is concerned. That's the $6 90° x 90° waveguide in the $600 (per each, I believe) JBL "Ultra Compact" 2-way I earlier linked here.

[soundminded]

We are getting to the truth here, in more ways than one.

Once again, you confuse realism with accuracy. Accuracy is well-defined, and measurable -- input = output.

Realism is subjective, though in broad terms, it's still determinate, and the reality of the concert hall, which for you is apparently the only "worthy" objective, the one Villchur and Allison were attempting to simulate, is inherently inaccurate, and no less so than that a rocker most enjoys. If a loudspeaker incorporates a sonic coloration to optimize reproduction of any particular perception of realism, it is, virtually by definition, ill-suited for all others, and thus, ARs are generally known to be "Classical music" speakers, and West Coast designs, "Rock" speakers; both are inaccurate, each in their own way.

Accurate speakers, on the other hand, are equally suited to both, and the entire spectrum between, key to their inherent versatility being smooth, flat frequency response and uniform power response -- EQ to taste....

"Accuracy is well-defined, and measurable -- input = output."

By that definition which is a very good one IMO, there are no accurate speakers or sound systems for reproducing musical recordings in the home. And that is not surprising. When one of our most prominent contributors who works in this field and was educated in one of the most prestigious engineering schools in the world does not know that sound fields are vector fields and not scalar fields, it gives great insight into the lack of understanding of how sound physically works. Accuracy is far more complex an issue than on axis frequency response. It is being able to measure the source field in all of its significant parameters where the listener in the audience hears it, devise a system for reproducing that exact sound field somwhere else, and being able to measure the resulting sound field where the listener hears it in his home or wherever it is being reproduced. The capability to do every one of those tasks is well beyond the current state of the art. There is no known method for even measuring it. Look at Leo Beranek's web site and read his white paper comparing 59 concert halls trying to correlate 20 measurements with the preferences of golden eared listeners including conductors of symphony orchestras. His measurements are a hodgepodge of bits and pieces which cannot be assembled to combine a coherent theory or model of sound. At least he is past the stage where he and every other acoustical scientist used a blank shot fired from a starter's pistol as a source.

You are right about the interactions of the listening room whether in the live performance venue where music is heard or the one in the reproduction space affecting the resulting field heard there as well. Those effects are poorly understood, the one in the live space has never been successfully captured, and the effect of the one in the home listening environment has not even been attempted to be taken into account by all but a handful of designers and always unsuccessfully. As an owner of Bose 901 for 40 years, I will say categorically that that speaker like all others cannot even in someone's wildest dreams come remotely close to reproducing the effect of concert hall acoustics. It simply cannot generate that type of sound field. It's other limitations as engineered also preclude it from reproducing the sound of musical instruments accurately as they would be heard in the listener's home. I've studied this and related problems for 35 years and knowing how to re-engineer them, it still took me four years to get as close as I have so far.

"Realism is subjective, though in broad terms, it's still determinate, and the reality of the concert hall, which for you is apparently the only "worthy" objective"

That is also a good definition IMO although another worthy objective is to reproduce musical instruments as they would be heard if the performers were in your own room as well. This restricts the kind of music you would hear to soloists or small ensembles. A symphony orchestra cannot be heard in a home. If the musicians and their instruments could be shrunk in size to fit physically but remain unaffected acoustically, the sound would be deafening. Reducing the loudness to within tolerable limits is a severe distortion. The only safe and suitable place to hear them is in a concert hall at a safe distance. Conductors often develop hearing loss over years as the result of being exposed to loud noise. This does not necessarily affect their ability to conduct. Nor can you realistically reproduce the sound of a symphony orchestra or any other sound as it is heard at a concert hall when your recording only contains five or ten percent of the sound field heard there by playing it ten to twenty times louder to make up for the missing sound which is qualitatively different from what you have.

To understand "realism" you'd need a good model of hearing. We don't have one. Measuring what the human ear can hear in terms of loudness and frequency range, sensitivity to differences in loudness and pitch are one thing. That and more are the easy part. That can be done with good accuracy. Understanding how the human brain interprets what it hears, draws conclusions about it is quite another. That is not understood. I've given as much thought to this problem as I have to the physical problem of engineering sound fields. I've drawn some tentative conclusions that are surprising even to me. (I've tried ploughing my way through Dr. Oliver Sach's book Musicophilia which is mostly anecdotal accounts of the medical effects of music but only touches briefly and inconclusively on the subject I'm interested in.) The question of reality depends on understanding which aspects of sound can and can't be heard and to what degree. Reproducing reality depends on reproducing with accuracy to a degree those aspects of sound which can be heard and to a degree beyond which differences can be detected. That is also beyond the state of the art.

The fact that AR3 was able to reproduce the sound of a guitar, a string quartet, that AR1 could reproduce a pipe organ, and AR 10 pi could reproduce the sound of drums all in the same venue as the sources they were compared to under highly contrived and controlled conditions does not imply that you can generalize that the accuracy of those speakers allows them to reproduce other musical instruments in those or in other venues such as a living room under less controlled conditions accurately or realistically, nor does it imply that those speakers can reproduce concert hall acoustics. In fact they can't. This is analogous to taking an accurate photograph of something illuminated under red light in an otherwise dark room and extrapolating to the concludion that the film and lens are so accurate, they can photograph the grand canyon at noon and looking at the photograph you'd believe you were there.

Contrary to the naive belief of many untrained audiophiles, electrical engineers know that equalization is a valid tool which is necessary in practically every type of analog communications system. Without equalization, long playing phonograph records, magnetic tape recording, FM radio, and color television would have been impossible. To those who reject the notion of equalization, DolbyA professional noise reduction if they understood it is a living nightmare of a concept. Loudspeakers do not live in a vacuum, they are part of an integrated system. That one component of that system should be equalization to flatten the overall system FR is virtually a given. Small wonder with such elementary concepts being rejected, this industry has reached a dead end.

[Shacky]

Edison's not bad, either.

The E'Waved AR2xs are spot on this thread for comparison, but I didn't build them, so that'll have to wait 'til they come in for testing.

Parts to update the AR3s are here, so I'll know how the pair sounds and measures soon, maybe this weekend. Even though that's not relevant, I'll link to it when it happens as a matter of potential interest.

I can probably do some preliminary polars on the A60's this weekend, too, and I'll link that in Carl's waveguide thread and y'all can compare those results, to the AR3a's as well as Speaker Dave's AR2ax's, at least insofar as directivity is concerned. That's the $6 90° x 90° waveguide in the $600 (per each, I believe) JBL "Ultra Compact" 2-way I earlier linked here.

Zilch,

Sounds like you are planning an EW project with your 3a's. I'm just curious, have you refurbed them to original specs to take your pre-EW measurements?

I do have to say - from my very subjectivist viewpoint - that modifying a valuable 3a to make it measure better does not mean it will subjectively sound better. It will certainly not be as valuable in resale market as an originally restored 3a which go for quite a lot - at least when ecoenomy turns around.

To me it's like take a 1962 Corvette and deciding to either restore it to its original beauty or making a hot rod out of it. Sure you could take a plain Jane Chevy small block 350 with modern fuel injection, upgraded transmission, suspension, plastic body panels... and prove that it's faster in quarter mile than the original and pulls more G's in the slolum. But does that make it "better" than the original?

I'd come down on the side of no absolutely not. You could take a much less valuable starting point than a '62 Vette and accomplish the exact same thing. So why would you want to take a very limited population '62 Vette and turn into something completely different and likely irreversable so?

Then it comes down to taste or sense of social responsibility. Nobody can stop anyone from taking a '62 Vette and doing whatever they want with it - even making it a crash car dummy test vehicle. But there are those of us that would never think of doing such a thing.

Do as you wish - you will anyway. But if you try to convince those that love '62 Vettes that modifying them is the "right" thing to do, you're not going to win. I see the same analogy with a pair or AR 3a's. I'm sure others here do as well.

But what's it all for? General knowledge? Not a good reason in my book. There are so many speakers out there one doesn't need to take on the venerable AR 3a and challenge it per my '62 Vette analogy. One does have to question the motivation behiond such a task. Is it to "prove" that they can be "improved upon". I think most hear would give you that - yes the AR 3a measured performance vectors can be improved upon. Will it be better than the original - I think not. But they are not my AR 3a's to decide.

[Steve F]

Two points, as comment to the L v R references on this topic:

1. The 1976-77 10 Pi vs. Neil Grover jazz drumming L v R demo was light years ahead of the earlier AR-3 vs. string quartet demo in terms of difficulty for the speakers to replicate the muscians' sound and also light years ahead in terms of how much better the associated electronics were in 1976 than in the early '60's.

The standard criticism of the older AR's being given "softball" string quartet material to replicate evaporates completely with the 105dB+ peaks of Grover's drum set. Additionally, the drum set demo was spectrally 'hot' with lots of close-field, brilliant cymbals being played and reproduced. The old "not enough highs" complaint against the 3/3a held no water with the 10 Pi/11.

I was there for several of these demos. I am (or more accurately, was) a professional jazz drummer. I was in my early 20's at the time of these demos and I was quite impressed with the demo. Perfect? No. I could detect a few shifts and switches. Outstanding nonetheless? Absolutely. 'Remarkable' was the word that comes to mind. There was a trace of audible tape hiss, which of course wouldn't be an issue today.

The 10 Pi had the exact same radiation pattern as the 3a (not the same spectral balance, but the same radiation pattern), and the 10 Pi did a superb job in these demos. Draw your own conclusions.

2. I asked Victor Campos (the head AR engineer who produced the 10 Pi/Grover demos) what about the LST? Even though the LST had just been discontinued, they did give it a try during the background work of this project. It didn't do as well as the 10 Pi. Campos said he thought it might have been due to the overlapping driver radiation of the LST's multiple mids and tweeters. The single wide-dispersion MF and HF units of the 10 Pi proved to be just right, for a demo group of about 20-30 listeners in a fairly big room.

Interesting stuff.

Steve F.

[genek]

Sounds like you are planning an EW project with your 3a's. I'm just curious, have you refurbed them to original specs to take your pre-EW measurements?

I do have to say - from my very subjectivist viewpoint - that modifying a valuable 3a to make it measure better does not mean it will subjectively sound better.

But what's it all for? General knowledge? Not a good reason in my book. There are so many speakers out there one doesn't need to take on the venerable AR 3a and challenge it per my '62 Vette analogy. One does have to question the motivation behiond such a task. Is it to "prove" that they can be "improved upon". I think most hear would give you that - yes the AR 3a measured performance vectors can be improved upon. Will it be better than the original - I think not. But they are not my AR 3a's to decide.

I think the update parts that Zilch is waiting for to complete his measurements of the original 3a configuration are new level controls.

As for whether a WG can be made to reproduce the original sound or make it "sound better," that is the question. Good-working original 3a drivers have become increasingly difficult to find to the extent that people like Roy and Carl who we know like the original AR sound enough to try to preserve it are experimenting with non-original domes and putting the original dome behind small WGs. IMO, we have reached the point where any replacement, of any technology, that can be made to reproduce the original performance, or widen its range of adjustability to suit modern needs if desired, is worth testing to add to the "general knowledge" of what alternatives are available when/if the original parts die. Even owners of '62 Vettes frequently modify them to run on unleaded gas and install seat belts and CD players in them. Museum-quality restorations are great if you can maintain them, but we want our speakers to work, not just look right behind glass.

[Zilch]

Sounds like you are planning an EW project with your 3a's. I'm just curious, have you refurbed them to original specs to take your pre-EW measurements?

The purpose in acquiring them was to shake out RoyC's L-pad pot substitution, and I have done that. There is more shaking yet to come, some of which will be of considerable interest to you.

I do have to say - from my very subjectivist viewpoint - that modifying a valuable 3a to make it measure better does not mean it will subjectively sound better.

If not, it only means we haven't measured the right "stuff." I believe we now know enough to make them sound better.

It will certainly not be as valuable in resale market as an originally restored 3a which go for quite a lot - at least when economy turns around.

I have no illusions that Zilchification of anything makes them worth more, particularly since I provide full DIY detail, often painstakingly and to some, excessively, for just about everything I do.

In this instance, the bar is pretty low, but as I stated earlier, I am not stupid. Bottom line, it's not about money, rather, knowledge, in which respect this pair of AR3a's have already paid for themselves many times over.

[A '53 'Vette, I would meticulously restore.... ]

[genek]

In this instance, the bar is pretty low

It doesn't necessarily have to be. It might be fairly easy to make the speakers "sound better" than the originals based on your and other peoples' preferences, but I would guess that trying to make them sound identical to the originals, regardless of what you think of them, would be more difficult, especially if fitting behind the original grill is also a requirement.

[Zilch]

I never stated what you indicate.

You are correct, Howard, and I apologize for the misattribution. It was another player in a different forum:

Yes, which is why I've taken to just not bothering to respond to any statement that falls back on quotes from Toole. Once you realize that everything Toole says about the subject proceeds from his belief that AR became more successful than any speaker company he ever worked for by swindling the unsophisticated and stupid because no speaker that failed to conform to his concept of correct design could possibly have been as good as its customers believed it to be, there's really no point in trying.

http://www.audiokarma.org/forums/showthrea...572#post2625572

********

Actually, as you define it, accuracy is also subjective, since the debate tends to involve whether we want a speaker that has flat input to a room over a wide angle (flat reverberant-field response), or if we want a speaker to deliver flat input to a room over a narrower angle (flat direct-field response).

Loudspeaker accuracy is well-defined and ascertainable by methodology that has basically been standard for 50 years. Once the room is brought in, all bets are off, though less so than many presume. It's you and others here who desire to redefine accuracy to encompass far larger scopes, up to and including realism, in support of theories as to how speakers perform, and what we like or dislike about them. The speaker must first be analyzed in isolation, and that information THEN combined with what we know about rooms and realism for a look at the "big picture." When we do that, the issues and parameters come into clear focus:

Frequency response is the single most important aspect of the performance of any audio device. It it is wrong, nothing else matters.

********

Anybody who says that somewhat later AR speakers like the LST had anything but flat frequency response needs to get out more.

That AR was compelled "smooth over" CBS Labs's measurements of LST, as well, clearly belies that contention:

http://www.classicspeakerpages.net/library...ort_on__12.html

[soundminded]

Two points, as comment to the L v R references on this topic:

1. The 1976-77 10 Pi vs. Neil Grover jazz drumming L v R demo was light years ahead of the earlier AR-3 vs. string quartet demo in terms of difficulty for the speakers to replicate the muscians' sound and also light years ahead in terms of how much better the associated electronics were in 1976 than in the early '60's.

The standard criticism of the older AR's being given "softball" string quartet material to replicate evaporates completely with the 105dB+ peaks of Grover's drum set. Additionally, the drum set demo was spectrally 'hot' with lots of close-field, brilliant cymbals being played and reproduced. The old "not enough highs" complaint against the 3/3a held no water with the 10 Pi/11.

I was there for several of these demos. I am (or more accurately, was) a professional jazz drummer. I was in my early 20's at the time of these demos and I was quite impressed with the demo. Perfect? No. I could detect a few shifts and switches. Outstanding nonetheless? Absolutely. 'Remarkable' was the word that comes to mind. There was a trace of audible tape hiss, which of course wouldn't be an issue today.

The 10 Pi had the exact same radiation pattern as the 3a (not the same spectral balance, but the same radiation pattern), and the 10 Pi did a superb job in these demos. Draw your own conclusions.

2. I asked Victor Campos (the head AR engineer who produced the 10 Pi/Grover demos) what about the LST? Even though the LST had just been discontinued, they did give it a try during the background work of this project. It didn't do as well as the 10 Pi. Campos said he thought it might have been due to the overlapping driver radiation of the LST's multiple mids and tweeters. The single wide-dispersion MF and HF units of the 10 Pi proved to be just right, for a demo group of about 20-30 listeners in a fairly big room.

Interesting stuff.

Steve F.

I think LST would have been an even better performer if it had two midrange drivers on the front panel to pair with the two front firing tweeters...or one midrange and one tweeter on each of thee panels...or on four at equal angles to each other. I think the geometry of the arrangement was not the most favorable one AR could have picked. Just a hunch. I'm sure they tried many combinations, performed many tests before the settled on the one they did....but still....

String instruments are not necessarily softball to reproduce. If you are familiar with the sound of real ones, that is something you'd probably agree with. A cello is a very difficult instrument to reproduce convincingly. If you ever hear one played in someone's home, you'll know what I mean. Bass output is phenomenal, beyond the capabilities of most speakers. As for reproducing the sound of a Guarnari del Jesu violin, no commercial speaker I ever heard came remotely close. Usually you're just lucky if reproduction of massed strings doesn't sound like fingenails scraping on a slate blackboard.

[Zilch]

Flat power response will only yield rising frequency response as the frequency climbs (and here I am assuming that you mean direct-field response, since the term "frequency response" can mean any number of things) IF the speaker is highly directional. With truly wide-dispersion speakers the direct-field output and the power response will be more similar than what we have with more directional speakers. If a speaker has flat anechoic response over wide angles, then the uniform energy output will indeed be pretty much the same. Wide-dispersion speakers will be as able to output uniform energy at high frequencies as at lower frequencies. It is only directional speakers that have this discrepancy between the direct and reverberant field outputs. AR was after constant directivity, in that they wanted uniform, broad-bandwidth output over a very wide radiating angle. If a speaker is going to be directional, lets make it that way over the widest angle possible and get maximum energy at all frequencies into the reverberant field.

You've come a long way, Howard; I am proud of you.

Now, despite the faulty reverberant field premise, the question becomes just how successfully vintage ARs such as AR3a actually met that now acknowledged constant directivity objective, and with factual knowledge of how loudspeakers and rooms interact, perhaps an estimate of what is optimum might be in the offing here.

You talk a lot about accurate, and I am curious about what particular speaker brands (other than the GedLee line) you consider as being "accurate." Certainly, even you have some favorites, and by listing those we can get an idea of what you are driving at when it comes to examples of solid performance.

Once the parameters are defined and understood, anyone can make that determination for themselves; nobody needs my "favorites" to dissect....

[Shacky]

I just looked at Stereophile on line. April columns not up yet. I encourage all on this thread to read the As We See It - I think I've already mentioned it earlier.

I think we are running around in circles - or at leat Zilch is - when talking about accuracy vs realism vs what speaker sounds best in my listening room. I'm of the oprinion and it's a novice opinion at best that a perfectly accurate speaker would become non-listenable on all but the best of recordings. I don't want that. And I don't want something that will develop listener fatigue.

I'm currently enjoying my Fisher Tube Amp with my JBL L36 Decades - refoamed, recapped and new binding posts. Yes I do listen to more than AR speakers. With my nice warm tube amp my JBL's are spectacular. Rich deep, tuneful bass - showing both ported and AS designs can work if done right - and I'm not saying the Decades are the best JBL but their the best I've had. I have the mid and hi attenuators turned down to almost -3. Probably a big reason why the bass sounds so prominent. I love this sound.

I tried my EPI 202's with same amp. Don't like them. Someone else could come in my room and say the love them or show me their output is more "accurate". Doesn't matter to me. I don't like them.

My AR 5's also sound great with my tube amp. AR 3a's - not so great. Just can't handle those AS woofers.

I'm sure Zich is right in that there is a good definition of accuracy - measuring what goes in vs what comes out of the speaker. But I don't listen to test tones. I listne to music. I want what sounds to me like a live exhilarating performance. And I do not believe that kind of presentation can be measured.

But what amp are you testing them with? What signal? Is it measuring how they play the music we listen to or a test signal? There is nore measuring equipment that can replicate our ears. Measuring is great, it's crucial in design and QC. But it's never going to be the end all untill we have electronic equivaent of ears and the brains procesing them.

So sharing what our favorites are to me means a heck of a lot more than scientific, philosophical, or just plain argumentative banter.