Some speaker curves

[Howard Ferstler]

Zilch and a few others here have made comments about speaker dispersion and speaker sound as it relates to AR, Allison, and JBL systems and drivers. Some have indicated that wide dispersion from top to bottom is not an advantage, with directional sound being better at separating room effects from what is on the recording. I do not care to debate that issue any further, since it involves musical taste (Zilch has said elsewhere that nobody cares about live, acoustic-music, concert-hall sound any more) more than speaker technology. As tastes decline, speaker sound parallels the trend.

On the other hand, in some posts Zilch has indicated that speakers like the AR-3a do not have dispersion as wide as some of the JBL brands he lionizes. Zilch has also posted pictures of some JBL system response curves to support that view, including a polar curve. I will reproduce those here, and then submit some comparison shots of various AR driver and system curves. In addition, even though this is the AR section of the Classic Speaker site, I am going to also submit some Allison Acoustics driver and system curves, as well. I do this, because so much of what we find with "classic" AR speakers was the result of work by Roy Allison. Roy's approach was an outgrowth of what Edgar Villchur came up with at the very beginning, with both men moving on to refine those basic priciples. In many ways, Allison Acoustics was a continuation of the original AR approach to hi-fi sound.

Drawing 1 shows the on and off-axis response of the JBL AC16 system from 200 Hz on up. Zilch has posted this drawing elsewhere, and it shows readouts from on axis out to 80 degrees off axis.

Drawing 2 shows the AR-3a (and LST) woofer output on axis and at 30 and 60 degrees off axis. The diagram shows the crossover point, and you can see that from 200 to 575 Hz the response is certainly equal to what the AC16 exhibits over that range.

Drawing 3 shows the AR-3a midrange output on axis and at 30 and 60 degrees off axis. The response between 575 Hz and 5 kHz Hz is certainly wider in dispersion than the AC-16 operating over that range. Of course, the same midrange is used in the AR-LST, and since the four mounted on that system are on angled panels, the effective dispersion will be much better than the JBL model.

Drawing 4 shows the AR-3a tweeter output above 5 kHZ on axis at at 15-degree intervals out to 75 degrees off axis, and nobody is going to say that its output is inferior in terms of dispersion to the JBL AC16. (This is a special curve made by Roy Allison after he left AR, and he made a point of measuring beyond the usual 60-degree point.) The dispersion contest here is really no contest at all. Again, this tweeter is on the front and angled side panels of the AR-LST, and clearly the dispersion of that system will eclipse that of the JBL model. However, even the AR-3a, with a single tweeter, exhibits wider dispersion than that JBL model.

While diffraction effects and crossover artifacts will impact the direct-field output of the AR-3a and AR-LST, the very wide dispersion of those systems will insure that those direct-field artifacts will be submerged in the power-response output of the systems. The wide dispersion insures that the power response will stay uniform over the audible operating ranges of the two systems.

Drawing 5 shows the power response output of an AR-3a in a reverberant chamber (the curve was run by Roy Allison), with the mid and tweeter level controls all the way up. Zilch has said that the power response output of the system sags at higher frequencies, but you certainly cannot see that here.

Drawing 6 shows the power response output of the AR-LST with its level control set at the "flat" position. Obviously, you can get very flat power from this system.

Drawing 7 shows the output of the Allison midrange driver that Roy came up with for his company's systems. The crossover points are at 350 and 3750 Hz, and the readouts are at 15 degree intervals from on axis out to 90 degrees off. It is hard to see all of the lines, because the super-wide dispersion characteristics have them overlapping so much. This driver is almost omnidirectional within its operating range over a 180-degree arc. When mounted on the angled panels of the Allison Model One, Model Two, and IC-20 the same angular range is increased to 270 degrees.

Drawing 8 shows the output of the Allison tweeter driver that Roy also came up with for his company's systems. The crossover point is at 3750 Hz, and the readouts are at 15 degree intervals from on axis out to 90 degrees off. It is hard to see all of the lines up to 10 kHz, because the super-wide dispersion characteristics have them overlapping so much. This driver is almost omnidirectional within its operating range over a 180-degree arc. When mounted on the angled panels of the Allison Model One, Model Two, and IC-20 the same angular range is increased to 270 degrees. Even operating alone, the dispersion here eclpses Zilch's exampled JBL model.

Drawing 9 shows a polar response curve that Zilch posted to highlight the dispersion superiority of a favored JBL model above about 600 Hz. Anybody can see how much the upper midrange and treble beam in comparison to the lower frequencies. From 3 kHz on up this is a very beamy speaker. This beaminess tends to make the speakers have a boxy sound, and negatively impacts soundstage realism in typical listening rooms. It tends to make them sound impressive in large, open showrooms, however.

Drawing 10 shows the polar response of the Allison IC-20 system from 63 Hz on up to 16 kHz. The Allison Models One and Two have similar horizontal polar patterns. It is clear that this is a wide-dispersing speaker that has the upper midrange and treble outputs paralleling those of the lower midrange and even the bass. Note that the off-axis response is wide well beyond this 180-degree plot.

Drawing 11 shows the polar response of the AR-LST, as made by a British review magazine some time ago. Since it was made in the direct field, it shows a lot of inter-driver lobing in the tweeter's range, due to the overlapping outputs of the four drivers, but it is clear that this is a very wide dispersing speaker, much more so than any JBL model Zilch has illustrated.

Now, some might decry this kind of wide dispersion, but whatever one might think of it, it certainly eclipses the varying broad-bandwidth dispersion characteristic of the kind of speakers Zilch prefers, and pretty much proves that the AR-3a (let alone the AR-LST and any Allison models) is a wider-dispersing model, particularly in the upper midrange and treble, than the JBL AC16.

Because of this, those speakers are much better at simulating concert-hall sound from typical classical and acoustic-instrument jazz recordings than beamier versions.

Howard Ferstler

[Zilch]

As tastes decline, speaker sound parallels the trend.

We kinda know what tastes have declined, Howard

On the other hand, in some posts Zilch has indicated that speakers like the AR-3a do not have dispersion as wide as some of the JBL brands he lionizes.

Never happened. I said we don't know the actual dispersion characteristics of AR3a. It's kinda hard to tell from the off-axis measurements, but the Allison (1970) PDF may afford some clues, once it is published. We have enough data already, actually, but you assert the 1972 reference is incomplete, so we'll have to wait.

http://www.classicspeakerpages.net/IP.Boar...ost&id=4590

Zilch has also posted pictures of some JBL system response curves to support that view, including a polar curve.

Nope, I posted that, and several others, actually, to illustrate that much of what you had said about the nature of constant directivity, and the inability of anechoic measurements to characterize it, was invalid.

Drawing 1 shows the on and off-axis response of the JBL AC16 system from 200 Hz on up. Zilch has posted this drawing elsewhere, and it shows readouts from on axis out to 80 degrees off axis.

Yes, normalized to show measured directivity, assuming flat on-axis response. It is not the the system frequency response, which is shown elsewhere on the spec sheet:

http://www.jblpro.com/catalog/support/getf...&docid=1078

Drawing 2 shows the AR-3a (and LST) woofer output on axis and at 30 and 60 degrees off axis. The diagram shows the crossover point, and you can see that from 200 to 575 Hz the response is certainly equal to what the AC16 exhibits over that range.

Not bad for a 6.5" woofer, huh?

Drawing 3 shows the AR-3a midrange output on axis and at 30 and 60 degrees off axis. The response between 575 Hz and 5 kHz Hz is certainly wider in dispersion than the AC-16 operating over that range. Of course, the same midrange is used in the AR-LST, and since the four mounted on that system are on angled panels, the effective dispersion will be much better than the JBL model.

"Bait and Switch," Howard; that's the response of the driver on a test baffle, not in the system.

Compare to the AR3a curves from Allison (1972) here:

http://www.classicspeakerpages.net/IP.Boar...ost&p=78713

Drawing 4 shows the AR-3a tweeter output above 5 kHZ on axis at at 15-degree intervals out to 75 degrees off axis, and nobody is going to say that its output is inferior in terms of dispersion to the JBL AC16. (This is a special curve made by Roy Allison after he left AR, and he made a point of measuring beyond the usual 60-degree point.) The dispersion contest here is really no contest at all.

Indeed not, Howard, sleight of hand again, just as AR marketing did originally; it's NOT in the cabinet.

ibid.

While diffraction effects and crossover artifacts will impact the direct-field output of the AR-3a and AR-LST, the very wide dispersion of those systems will insure that those direct-field artifacts will be submerged in the power-response output of the systems. The wide dispersion insures that the power response will stay uniform over the audible operating ranges of the two systems.

If so, it's certainly not evident in Allison's in-room power response curves.

[Zilch]

Drawing 5 shows the power response output of an AR-3a in a reverberant chamber (the curve was run by Roy Allison), with the mid and tweeter level controls all the way up. Zilch has said that the power response output of the system sags at higher frequencies, but you certainly cannot see that here.

'Cause we're looking in the wrong place. AR's reverberant chamber was carpeted, right? With furniture, doors and wall openings? Listening rooms are not reverberant chambers.

Drawing 6 shows the power response output of the AR-LST with its level control set at the "flat" position. Obviously, you can get very flat power from this system.

Nice one Howard. Are we still in the shower here? AR sang the same song there when CBS Labs actually measured them. How come the VHF is rolled off?

Drawing 7 shows the output of the Allison midrange driver that Roy came up with for his company's systems. The crossover points are at 350 and 3750 Hz, and the readouts are at 15 degree intervals from on axis out to 90 degrees off. It is hard to see all of the lines, because the super-wide dispersion characteristics have them overlapping so much. This driver is almost omnidirectional within its operating range over a 180-degree arc. When mounted on the angled panels of the Allison Model One, Model Two, and IC-20 the same angular range is increased to 270 degrees.

Cool! Test baffle or in-cabinet?

Drawing 8 shows the output of the Allison tweeter driver that Roy also came up with for his company's systems. The crossover point is at 3750 Hz, and the readouts are at 15 degree intervals from on axis out to 90 degrees off. It is hard to see all of the lines up to 10 kHz, because the super-wide dispersion characteristics have them overlapping so much. This driver is almost omnidirectional within its operating range over a 180-degree arc. When mounted on the angled panels of the Allison Model One, Model Two, and IC-20 the same angular range is increased to 270 degrees. Even operating alone, the dispersion here eclpses Zilch's exampled JBL model.

Alone, indeed.

Drawing 9 shows a polar response curve that Zilch posted to highlight the dispersion superiority of a favored JBL model above about 600 Hz. Anybody can see how much the upper midrange and treble beam in comparison to the lower frequencies. From 3 kHz on up this is a very beamy speaker. This beaminess tends to make the speakers have a boxy sound, and negatively impacts soundstage realism in typical listening rooms. It tends to make them sound impressive in large, open showrooms, however.

Say "Never mind" to the readers, Howard, 'cause it's a 10" midwoofer I showed to illustrate what DI meant for Carl:

http://www.jblpro.com/pub/obsolete/2123.pdf

[Guess you missed that part.]

Drawing 10 shows the polar response of the Allison IC-20 system from 63 Hz on up to 16 kHz. The Allison Models One and Two have similar horizontal polar patterns. It is clear that this is a wide-dispersing speaker that has the upper midrange and treble outputs paralleling those of the lower midrange and even the bass. Note that the off-axis response is wide well beyond this 180-degree plot.

They'd make good HT surround speakers, maybe, where that might matter.

Drawing 11 shows the polar response of the AR-LST, as made by a British review magazine some time ago. Since it was made in the direct field, it shows a lot of inter-driver lobing in the tweeter's range, due to the overlapping outputs of the four drivers, but it is clear that this is a very wide dispersing speaker, much more so than any JBL model Zilch has illustrated.

Let's pretend the lobing isn't audible? Or the 20 dB notches at 10 kHz? The trend is clear -- that's all gonna smooth out above 10 kHz, right?

Now, some might decry this kind of wide dispersion, but whatever one might think of it, it certainly eclipses the varying broad-bandwidth dispersion characteristic of the kind of speakers Zilch prefers, and pretty much proves that the AR-3a (let alone the AR-LST and any Allison models) is a wider-dispersing model, particularly in the upper midrange and treble, than the JBL AC16.

AC16's spec'd at 90° x 90°. It's indeterminate whether AR-3a can accomplish that, even, and AR clearly had justification for installing multi-mid/tweeter combinations on angled baffles to achieve the "max dispersion" objective. Still not good enough to meet the target, apparently; it took three to do it in LST-2, no?

Because of this, those speakers are much better at simulating concert-hall sound from typical classical and acoustic-instrument jazz recordings than beamier versions.

Speculation on many scores, obviously. Indeterminate from the data you've provided.

Thanks for doing this, though, Howard; it has revealed far more than you intended....

[Zilch]

In summary:

1) It's apples to oranges you're doing here, Howard. While measurements on a test baffle of each individual driver are informative, it's how they operate together in a cabinet with a crossover that most counts, and in a room, the game changes substantially again. That's why anechoic system response measurements give the clearest picture, and here is apples to apples, as best we have them to date; open the links in separate windows or tabs to compare:

JBL AC16

vs.

AR3a

2) Classic AR (and Allison also, apparently,) design relies upon a presumption that a uniform, diffuse, integrating, reverberant field is dominant in typical listening spaces, and thus, power response most accurately characterizes what we hear therein. True for low frequencies, in terms of knowing what's being "delivered" to the room, but even then, what we hear depends upon where we sit. Not true above the transition frequency at which the room's dominance transfers to the speaker. Flip the phase on the midrange or face the speakers to the wall, or backwards, in the extreme, and they do not sound the same; toeing in versus out clearly matters, as well.

3) The theory has long since been discredited by research, which has also revealed many of the factors to which what we find appealing about how these designs sound may be attributable. That knowledge defines a clear path toward achieving the same or a similar result using modern loudspeaker technology, but without the compromises in sonic quality so apparent in these early attempts at bringing the advantages of constant directivity to listening at home....

[Zilch]

Counterpoint:

While baffle mounting and driver interactions would impact the direct-field output, those characteristics would have only a slight impact on the actual wide-angle power output of those speakers. Reflected energy may be redirected, but it certainly is not lost. Allison measured only a handfull of angles to make his point. However, if one measures the multitude of angles involved the overally pattern that results resembles the power response. Your problem is that you take a handful of curves and assume they show the strong points or limitations of a speaker when it comes to its direct-field response. To do the job right, you would need to do a million such measurements, and the net result would be a power curve.

Surely, you're not saying the 22 points in 8 living rooms Allison used are insufficient, I hope. If so, we are RUINED!

Reversing the leads (reversing the phase) on a midrange driver will impact the power response as well as the direct-field response, so, yes, the result will be audible. So what?

WAIT now, Howard, you just told us that no energy is lost, rather, it is integrated by the reverberant field. Surely the midrange puts out no less energy merely because its polarity is reversed. You're not telling us that phase matters, I hope?

1. Facing the front wall introduces a delay that is differerent from what we have with facing the speaker forward. It also enlarges the apparent sound source.

And delay, too? AARRGH!

2. Typical wide-dispersing speakers are still not fully omnidirectional, so obviously facing them towards the front wall will screw up their radiation pattern behavior in relation to the listener out in the room.

AR put the mic behind the speaker in the reverberant chamber. No good, huh?

3. Typical walls will absorb some energy, so facing the speakers at them will cause a loss in upper-midrange and treble energy even in the reverberant field.

Which doesn't happen when they're facing front, right?

4. Even Allison will admit that the direct-field output controls soundstaging and imaging, and obviously facing a speaker towards the front wall will undermine that function.

5. Allison does not say that the direct field is inconsequential. However, its impact on the spectral balance of a wide-dispersion speaker is minimal.

But the phasing, delay, and directivity DO affect it? Say it ain't SO, Howard.

[Zilch]

1. The curves I posted, as well as the one you posted showing the cumulative direct-field curves Allison measured at a variety of off-axis angles, are obviously direct-printed jobs that show every nuance of the outputs. On the other hand, the curve you posted is anything but a machine-run curve that illustrates exactly what the speaker is doing. Rather, it is a hand-drawn job designed to illustrate points, probably in a sales brochure.

2. The AR-3a group of curves you ran (from the Allison JAES article) are probably rougher in the midrange than those one would run with the older AR-3 model. Indeed, Toole ran such curves in his book to illustrate just how poor the AR-3 performed.

YOU provided the JBL product curve for comparison; that's no longer good enough now for you to make your point that AR3a trumps it? O.K., here's the $9.90 JBL EconoWaveguide, which I also posted earlier, as measured by Geddes and posted on his website to demonstrate that it rather sucked in comparison to his design, 0-90° (180° beamwidth) in 7.5° increments, in comparison to my own measurements of AR3a, which are remarkably similar to those published by Allison 40 years ago, and those shown in Toole's book Fig. 17.2, p. 341. Again, open in separate windows or tabs to compare:

E'Wave

vs.

AR3a

From those curves one would get the impression that the speakers were outrageously bad sounding. Yet, with the AR-3 models Villchur managed to impress a large number of audio buffs (and "looking for faults" audio journalists) enough for the speakers to, at the very worst, be able to very closely mimic the sound of a live ensemble. Either those listeners were naive (or borderline deaf) or else your measurement analysis, as well as Toole's, are in error. And neither you nor Toole ever bothered to put your contentions about what does and does not matter with loudspeaker performance to the live-vs-recorded test as Villchur did. You may claim that your analysis is proper, but you are basically just guessing.

Contrived, Howard, no less so than Edison's "performances," including before a full house of "musically cultured and musically critical" listeners in Carnegie Hall fully 50 years earlier, which similarly impressed the audience. RCA did the same in 1947 using a full symphony orchestra.

Finally, Allison's article draft more or less proves that the reverberant field dominates with wide-dispersion speakers in typical home listening rooms. He did this by actually going out and measuring speakers in a number of different real-world listening rooms, which is more than, you, JBL, or Toole ever did.

We shall see, Howard. Toole and Olive did it under controlled conditions, with hundreds, perhaps thousands, of subjects, over the course of 20+ years.

All one needs to do to evaluate how carpeting and furnishings impact the power response is look at the various in-room curves of the AR-3a performance that Allison ran. (Geeze, I wish the webmaster would get that more expansive material I sent to him posted.) Obviously, such furnishings impact the power response, because of absorption. So what?

So it's refreshing to see you concede that AR's reverberant chamber curves, which they published in lieu of the anechoic system frequency response then (and still today) standard throughout the industry, are not representative of how the system "plays" in normal listening spaces.

As for the CBS Labs data, I am including both the AR reverberant-field measurements (again), plus a copy of the actual CBS Labs draft report measurement of the speaker. Sure, the average omnidirectional output will roll off compared to the front hemisphere response, simply because of attenuation at very wide angles. Unless a speaker is completely omnidirectional, the power response will always roll off compared to any direct-field or narrower angle measurements.

Cool. Looks like what Toole shows throughout his book, on-axis, listening window, and power response. Only thing missing is the Directivity Index, which may be derived from those curves. Alas, it took AR four tweeters and four midranges on angled baffles to achieve those results. You've given up on contending that AR3a can match that performance, apparently; most gratifying.

As for the AR data not exactly matching that of CBS Labs, I think we can write that off to measurement technique and microphone differences.

Not to the "special" reverberant chamber AR used to cast the best possible light on the performance of their designs, of course....

[Zilch]

Footnote:

Most good speakers over the years could, if adjusted properly, do pretty much what the AR-3 pair did with those Villchur demos.

[soundminded]

Footnote:

Then why don't they prove it?

Most users of modern speakers place their speakers and their head for listening with a tape measure to sit in some golden sweet spot. Yet AR did it simultaneously for about two or three dozen people at a time. What good are speakers if you have to be a nuclear physicist to install them properly and they don't send one along with the equipment to do it for you? One guy who will go unnamed went to Stereophile's facility to set up the speakers he manufactured reportedly to within about an inch of optimal and evidently it took some considerable time and effort. Do you think he'd have done that for me too if I'd have bought a pair of them?

[Zilch]

Then why don't they prove it?

They do; at virtually every live performance you've ever heard outside the acoustic concert hall, and also at some of those, too.

It's called "Sound Reinforcement," is mostly done with constant directivity loudspeaker systems these days, and there are several professions devoted to the art and science of doing it well. You REALLY should get out more....

[genek]

They do; at virtually every live performance you've ever heard outside the acoustic concert hall, and also at some of those, too.

It's called "Sound Reinforcement," is mostly done with constant directivity loudspeaker systems these days, and there are several professions devoted to the art and science of doing it well. You REALLY should get out more....

Sound reinforcement isn't reproducing an original performance in such venues, it is the original performance. Does anybody at one of those concerts believe that what's coming out of those speakers is the unreinforced sound of the performers?

I expect the task of convincingly reproducing the sound of a reinforced performance in its original environment in someone's living room has just as many issues to addess as reproducing the sound of an unreinforced performance.

[genek]

Then why don't they prove it?

I have an admittedly limited amount of experience with today's "high end" audio owing to a dearth of local dealers available to me, but from what few observations I have been able to make, it strikes me that since the demise of mid-fi, the few remaining manufacturers catering to those who make the jump from cheap HT to the stratospheric prices of "high end" have it pretty good. Customers buy all sorts of expensive speakers and amplifiers on faith in response curves and after very limited listening, and when they find they dislike the sound coming out of their systems and can't adjust it because they have no tone controls, they complain not about the systems, but about the poor quality of their sources, which leads to them spending even more money on megabuck turntables and CD players, exotic cables and speaker wires, CD and vacuum tube "stabilizers," turntable platter mats cut from foam toolbox liners, etc.

I wonder, even if a manufacturer could prove its equipment can fool listeners into thinking they're hearing live, might it not be a bad thing to do? How would they get their limited number of niche market customers to buy anything new next year?

[Zilch]

In Summary #2:

1) In pursuit of maximum dispersion to achieve flat power response in the reverberant field, early AR design incorporated constant directivity, most likely without knowing what it was, conceptually, or how optimally to achieve it, at least not in modern terms nor with contemporary understanding and appreciation. What we most like about AR3a may well be this artifact rather than the intended objective, the benefits of which become apparent once the erroneous concept of a dominant reverberant field in normal listening spaces is abandoned and better assigned its actual role, as do opportunities for improvement upon these pioneering, albeit inadvertent, innovations.

2) AR3a was not fully up to the task. Despite the widespread (and also erroneous) perception that the dispersion characteristics of the tweeter have never been surpassed, its performance has long since been eclipsed by technological advances; it may also be shown that alternative approaches bettered it, even in its own time.

3) Most apparently, the transducers themselves, good as they may have been, were substantially compromised once incorporated into system designs which ignored the consequences of driver/cabinet interactions, largely based upon the (once again, erroneous) presumption that the reverberant field would fix all of them. It doesn't; directivity issues remain evident, even in the summed and highly averaged and smoothed in-room power response measurements presented by Allison (1970, 1972).

4) Can AR3a's legendary dispersion be matched? Easily, as evidenced above, and by a $6 entry-level waveguide no less. It doesn't require the 9 drivers AR (and others) used to best it anymore; two will do.

5) Alas, the hidden agenda here demonstrating that Allisons are better'n ARs torpedoed the case....

A major design goal was to achieve constant directivity, and they did well; the only minor exception 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 [AR-3] is its frequency response, which significantly rolls of toward both low and high frequencies.

[Shacky]

In Summary #2:

1) In pursuit of maximum dispersion to achieve flat power response in the reverberant field, early AR design incorporated constant directivity, most likely without knowing what it was, conceptually, or how optimally to achieve it, at least not in modern terms nor with contemporary understanding and appreciation. What we most like about AR3a may well be this artifact rather than the intended objective, the benefits of which become apparent once the erroneous concept of a dominant reverberant field in normal listening spaces is abandoned and better assigned its actual role, as do opportunities for improvement upon these pioneering, albeit inadvertent, innovations.

2) AR3a was not fully up to the task. Despite the widespread (and also erroneous) perception that the dispersion characteristics of the tweeter have never been surpassed, its performance has long since been eclipsed by technological advances; it may also be shown that alternative approaches bettered it, even in its own time.

3) Most apparently, the transducers themselves, good as they may have been, were substantially compromised once incorporated into system designs which ignored the consequences of driver/cabinet interactions, largely based upon the (once again, erroneous) presumption that the reverberant field would fix all of them. It doesn't; directivity issues remain evident, even in the summed and highly averaged and smoothed in-room power response measurements presented by Allison (1970, 1972).

4) Can AR3a's legendary dispersion be matched? Easily, as evidenced above, and by a $6 entry-level waveguide no less. It doesn't require the 9 drivers AR (and others) used to best it anymore; two will do.

5) Alas, the hidden agenda here demonstrating that Allisons are better'n ARs torpedoed the case....

Looks like a bunch of BS to me.

Graphs don't translate exactly to human hearing. I haven't heard any speaker sound so full, so musical, so non-fatiguing as my AR's. I listen at low to moderate sound levels. So many new speakers only open up after cranking them. So how do you measure these AR characteristics? By listening to them with your favorite music.

[Zilch]

Looks like a bunch of BS to me.

Graphs don't translate exactly to human hearing. I haven't heard any speaker sound so full, so musical, so non-fatiguing as my AR's.

Can y'all say "Transparent?"

As any of those who know me, I love AR speakers. My older 2AX, my 3a's. All recapped and refurb'ed. Driving them with Sansui 890DB or 9090DB fleshes out the bass potential in these Acoustic Suspension speakers.

Well last week I hooked up my Fisher console tube amp. and since that isn't a good combo for my AR speakers, thought I'd give my JBL L36 Decades another spin. I liked them a lot last itme I used them with the Fisher tube amp.I gotta say this is my favorite setup so far. No way am I getting rid of the AR's. But I'm keeping the JBL's in place for a long time.

How 'bout "Disingenuous?"

Or " Subjectivist Thread Crap," maybe?

[Shacky]

Can y'all say "Transparent?"

How 'bout "Disingenuous?"

Or " Subjectivist Thread Crap," maybe?

Disingenuous because I can also appreciate JBL L36's with tubes? Come on Zilch. You can do better than that. You should be celebrating fact that I like my JBL's too. I also like my Dynaco A25's, KLH 6's, Wharfedale W70D's. So go ahead. Lay it on.

You know - the easiest thing in the world to do is say somthing negative. How about you put a liitle work into your next post and say something positive about an AR 3a's performance, Have you listened to any music through them?

[Zilch]

You know - the easiest thing in the world to do is say somthing negative.

Like this, Shacky?

Looks like a bunch of BS to me.

This forum knows you full well as a proponent of ignorance; there seems little point in reiterating this at every available opportunity....

[Shacky]

Zilch,

You are far from being in a position of speaking fon behalf of this forum. Seems to me most have questioned why you even bother to be here.

So please speak for yourself when trying to insult me.

I'll ask again. Have you listened to any music ( and what type would also by nice to know ) through your or another's AR 3a? Or would that strike of acting subjectivist?

Oh wait! I'm sorry. Speakers are made to be measured by non-carbon equipment. Not human ears. How foolish of me to keep coming back to how speakers actually sound. Shame on me. I'll go back in my ignorant corner.

[Zilch]

I'll ask again. Have you listened to any music ( and what type would also by nice to know ) through your or another's AR 3a? Or would that strike of acting subjectivist?

Oh wait! I'm sorry. Speakers are made to be measured by non-carbon equipment. Not human ears. How foolish of me to keep coming back to how speakers actually sound. Shame on me. I'll go back in my ignorant corner.

Look at the title of this thread, please, Shacky, and recognize that the subject is measurements. I have suggested at least twice here how these might relate to what we like about AR3a's. Still, apparently because you don't understand it, and have no desire to, you feel somehow compelled to embellish the discussion with a gratuitous and insulting off-topic thread crap.

Thank you for contributing that you think measurements, and by implication, this thread itself, are BS. Did you have something of substance to further the discourse here? Because what you have thus far offered is tired, old, news with respect to which I believe it safe to suggest that neither Howard nor I give a flying whit....

[Shacky]

Look at the title of this thread, please, Shacky, and recognize that the subject is measurements. I have suggested at least twice here how these might relate to what we like about AR3a's.

Ok Zilch. Are all these coments from you measurements or subjective fodder?

improvement upon these pioneering, albeit inadvertent, innovations

AR3a was not fully up to the task

long since been eclipsed by technological advances

Allisons are better'n ARs torpedoed the case

Lots of subjective extrapolations there. Maybe you oughta direct that crtical eye of yours on your own material for a change. Instead of making such subjective comments then quickly hiding behind the oft missing data or worse a "reference" to a post here or there from "experts" as world renowned as Shacky or Onplane.

You constantly remind me of Bill Murray's character in the movie Ghostbusters and the line I love:

"Back off! I'm a Scientist!"

[Zilch]

You constantly remind me of Bill Murray's character in the movie Ghostbusters and the line I love:

"Back off! I'm a Scientist!"

The one I like is "Who ya gonna call?"

Am I to trust you or Tom Brennan?

Have you looked up the quote yet? What'd he say?

The answer is, "Neither;" both are equally irrelevant....

[Zilch]

RE: the 22 measurement points. Allison was measuring the reverberant field outputs, and so doing a small number of measurements were not as restrictive as what would be the case if he were measuring in the direct field.

Thank you for affirming your confidence in Allison's measurement of power response. We'll be coming back to that at a later date.

RE: reversing the midrange driver phase. If the midrange driver's polarity is reversed there can be energy losses at the woofer and tweeter crossover points with the midrange, due to the driver outputs being out of phase with each other. This can cause dips in the response, or peaks if the in-phase overlaps are large.

However, when such dips are instead derivative of suboptimal driver alignment, they are inconsequential?

1. Certainly it introduces a delay. Depending upon the driver (especially the woofer driver) distances to the wall in either orientation it will impact the mid bass.

But the reverberant field mitigates that when it's coming in laterally?

2. In a reverberant chamber you can get away with that, because so much of the sound is reflected. It is not so much reflected in typical listening rooms, although it is reflected enough to have the reverberant field dominate with speakers having reasonably wide dispersion. Of course, many speakers begin to beam in the upper midrange and treble (and some even beam at the tops of the operating ranges of their woofers, depending upon woofer size and crossover points), and this could allow the direct field to gain relative strength.

It's beginning to become tortuous, now.

3. Drivers all beam to some extent, so aiming them at the wall will increase the amount of energy absorbed. With the speaker turned around the treble and upper midrange energy hitting the wall is far less, and so less is absorbed.

Which is why the direct field dominates, of course.

5. Re: spectral balance. Phasing can effect it, for the reason I mentioned in the midrange driver phase comment, above. Directivity can impact it, at least at frequencies where the driver size is causing beaming, at least with speakers that have to be aimed to sound good. With good, wide-dispersing speakers directivity is not really an issue, assuming the speakers are at least aimed nicely into the room area. As for delay, no. Delay has an impact on other things besides spectral balance, and that impact can be considerable.

Thank you, Howard, for supporting my case.

PS: your "pen" name, Zilch, fits you to a T.

Losing your resolve again, are you Howard?

[Zilch]

Of course, one can see driver interference and diffraction effects in your AR-3a measurement, but whatever those effects, one would have to be blind to think that the waveguide has dispersion as wide as that of the AR-3a.

Look again; it's obviously wider and also largely devoid of the response anomalies so apparent in the AR3a.

It is preposterous to say that the people who attended those AR demos, some of whom have posted here about how effective they were, with some of the other attendees also being astute audio buffs and audio journalists looking for flaws and not just in the mood to be overawed, would have been so naive as to be fooled by the Villchur demo.

You have yourself declared that many quality speakers would perform as well. It's just not relevant, is all.

Toole's measurments and comments about the AR-3, not to mention yours, make the speaker out to sound not much better than a table radio, which is idiotic.

Many table radios have better HF response, actually. I have measured the Bose Wave Radio, and Toole shows results for inexpensive consumer mini-systems in Fig. 2.5.

As I have said elsewhere, Villchur's demos puts to the lie both Toole's and your primary theses about what matters with speaker sound. Obviously, you both have a very strong vested interest in downgrading what Villchur did.

Please, Howard, find some basis more credible to support your subjective findings of merit....

[bitbandit]

It is quite common on forums to have a troll post, one that says something controversial and then sits back and enjoys the arguments it provokes.

It reminds me of the old saying, "never wrestle with a pig, you'll end up covered in mud - and the pig enjoys it!"

It was an interesting discussion early on, but come on guys, when you have got to the level of quoting each other's posts and then attempting to disprove them line by line, it is no longer interesting and maybe time to stop and agree to differ.

Just my tuppence worth. ;-)

[Shacky]

It reminds me of the old saying, "never wrestle with a pig, you'll end up covered in mud - and the pig enjoys it!"

I'm finally coming to see this. I don't enjoy it so hope that means I'm not a pig ; )

[Shacky]

Me, too, Shacky. The whole issue has become depressing. Zilch is obviously having a lot of fun pulling multiple tails around here, and I rather doubt that any level of contrary expertise or evidence is going to turn the guy off. Best to just give up on him and find something else to do.

I believe that the Villchur concerts put his argument to rest before it even got off of the ground, but since Zilch thinks they were bogus showboating, their value as a debating tool is, well, zilch.

Howard Ferstler

Were those demos done with the 3 and 3a's? I have a pair of 3's awaiting restoration. I'm looking forward to it as I do prefer the early 2AX with same tweeter as the 3 more so than the later 2AX. Course it could just be me and/or my listening space. But the early 2AX are magincal to me. Darn but didn't one of my tweeters go out a few weeks ago. I have a spare. Thinking about going back in and replacing all the AP rhesotats with new Ohmites I picked up. I may also touch up the doping on the cloth seurrounds as there's not much delay with the finger push test.