In praise of AR3a's

[soundminded]

Yes, the circumstances of each seem quite similar, except that the wine researchers seem more willing to admit that they still haven't got it down right.

To hear the audiophiles and those who manufacture and sell to them tell it, you'd think they have all the answers when in fact they have none of them. Proof? They are constantly coming out with new and improved models. Last years whizbang is this year's dog. As for the reviewers, it's a case of the blind leading the blind. Measurements? What measurements? Take the case of the recently cited review of Bose 901 with measurements. The "reviewer" didn't even bother to make an attempt to install the speakers according to the manufacturer's explicit instructions which were to place them 6 to 24 inches in front of a wall with 12 to 18 inches being optimal. Instead these were 5 feet away. The results were worthless. What if Alison speakers were installed in the middle of a room, turned sideways, and upside down? What kind of measurements do you suppose you'd get then? The reviews in the magazine were hardly much better. With a vast sophisticated array of test equipment specifically for measuring the performance of audio equipment, the reviewers acknowledged that there were special qualities resulting from the novel direct/reflecting principle yet no one of them tried to devise a way to measure this aspect of its performance by for example using a shotgun mike to measure the energy as a function of angle of incidence upon the listener at various frequencies and compare it to other speakers or to musical instruments, not even one tried.

Tom Tyson said in a prior post on another thread that this industry died a long time ago. He's right of course, it's the natural result of those who make money out of it being brain dead themselves. They don't even know it. Their lack of real knowledge doesn't allow for the fact that as engineering is by definition applied science, when you don't have the scientific knowledge to understand a problem let alone solve it, you have to go back to the most basic assumptions and reassess them with a very critical eye. This should lead to entirely different understanding of new scientific knowledge but it is a clear risk of time an money with no guarantees. It is far easier and safer to find some quirky buzz-word or catch-phrase of the year, try a new wrinkle with it, and advertise the hell out of it with plenty of profit as reward to those who successfully sell it. Don't expect much at all from these people and you won't be disappointed. Just don't hand over much of your money to them and you won't feel cheated either.

[Mexicomike]

C'mon Pete, you can't make two belittling comments and then follow it with a "it's only music lighten up people." That doesn't really work, does it?

I go back to my primary argument...LISTEN to the speakers and make a decision. Don't pay ANY attention to what some theory guy has to say or what some marketing guy has to say or, to use your term, what some internet wanna-be expert has to say. Just listen and decide. That's the easiest, least expensive, and most productive solution. It is, as you (and others) have noted, ONLY about the music.

[soundminded]

"Objective truth" would be independent of subjective preference. If someone could accurately measure and characterize the performance of the AR-3a and the Bose 901, they could use that information to configure the performance of another speaker system, even one that utilized entirely different technical approaches in the design of its system and components, to exactly duplicate the performance of the 3a and 901. The listeners might still not agree on what sound they were hearing and whether they thought it was "good" or not, but they'd be unable to distinguish the duplicate from the original. IOW, their subjective like or dislike of the performance of the duplicate would be identical to their subjective like or dislike of the performance of the original.

If you ever find yourself in the San Diego area, look me up. I have wine, AR speakers and NO sound measuring equipment.

That is correct. If a method were found that could measue ALL of the audible aspects of those speakers accurately and other speakers engineered with the same measurements, they would sound identical to the originals. This alone points out the woeful inadequacy of the state of the art of speaker measurement since speakers that measure the same using current methods don't sound the same. And were one to develop such methods, they would be the result of far greater understanding of what we hear and what we can hear. This would allow other speakers to be designed that combine the best aspects of each of them and outperform either of them...or engineered to far better standards using entirely different criteria. That is the way the state of the art is advanced. But before you can do that, you must return to basics and see where the fundimental flaws and limitations are in the current understanding and correct them.

[genek]

Mike, if you ever find yourself in the San Diego area, look me up. I have wine, AR speakers and NO sound measuring equipment.

[Zilch]

I go back to my primary argument...LISTEN to the speakers and make a decision. That's the easiest, least expensive, and most productive solution. It is, as you (and others) have noted, ONLY about the music.

And IF, on that basis, we should choose something other than ARs, you're not gonna be all HURT, or nothin', are you; I mean, what you or anyone else likes or dislikes should not be of any consideration or consequence in this, right?

[Pete B]

I'm not trying to measure what I hear. For me, it's enough to hear it and to nag those of you who claim you can to back up your claims.

Would you be willing try an A/B or A/B/X test to identify the "better" speaker

with one being an AR-3a?

Any AR enthusiast willing to give it a try?

[Pete B]

C'mon Pete, you can't make two belittling comments and then follow it with a "it's only music lighten up people." That doesn't really work, does it?

I go back to my primary argument...LISTEN to the speakers and make a decision. Don't pay ANY attention to what some theory guy has to say or what some marketing guy has to say or, to use your term, what some internet wanna-be expert has to say. Just listen and decide. That's the easiest, least expensive, and most productive solution. It is, as you (and others) have noted, ONLY about the music.

Belittling? nah, it's only Hi-Fi and the internet, there are much more serious things in life.

[genek]

Would you be willing try an A/B or A/B/X test to identify the "better" speaker

with one being an AR-3a?

Any AR enthusiast willing to give it a try?

As long as I don't have to haul a pair of 60 lb speakers somewhere, sure. When are you coming over with yours?

[Pete B]

As long as I don't have to haul a pair of 60 lb speakers somewhere, sure. When are you coming over with yours?

You're a bit far, no I'm not hauling anything either, so good luck.

If you're ever in the North East let me know - I do have an A/B box.

[Mexicomike]

You misunderstand me...I don't care what speakers anyone prefers. Heck, one of my best friends (the jazz/blues guitar player) swears that his Heil AMTs driven by a pair of tube McIntosh amps are the best speakers in the world. He thinks my ARs or my Maggies or my Thiels or, until I sold them, my Nestorovics 5AS's sound like crap. But I feel the same way about his Heils. I'm just saying that speakers should be selected by listening to them, not by listening to somebody spouting specs or theories as to why a speaker "must" be better than another speaker.

I'm not saying that LSTs or whatever are the best speaker ever made. I clearly stated they were the best speaker reproducing a good recording of a Jazz/Blues quartet that I have ever heard. But I certainly haven't heard them all and even at that, I hear with MY hearing, not your hearing. I have sat at demos where one person said, "listen to those realistic cymbals," while I thought, "Geez, turn the treble down!" Obviously we were hearing totally different sounds out of the same loudspeaker.

Sure, I'd be glad to AB ARs with any other speaker with the same general range capability on any material - listening to speakers that way is what I used to spend a LOT of time doing. Would I bet they'd win? No, that's crazy, - I have no idea what the other person hears OR what they prefer soundwise. But I'd bet they'd be in the running. Again - speakers within the same range, no AR4x against Thiel 3.7s etc!

This whole discussion, if I remember, started with someone talking about how they liked their AR speakers and some folks have advocated that actually, they are not very good. So I guess everyone that thinks they DO sound good is wrong and those that have specs to explain that they don't sound good is right? That's like saying if you like the "Mona Lisa" you are seeing wrong because "Scream" is better looking.

It's TOTALLY SUBJECTIVE!

PS: GEN, If I ever make it to San Diego, count me in!!!

[Mexicomike]

Forgot to add...hope we're all still friends!

[genek]

And IF, on that basis, we should choose something other than ARs, you're not gonna be all HURT, or nothin', are you; I mean, what you or anyone else likes or dislikes should not be of any consideration or consequence in this, right?

The ARs that most of us prefer haven't been made for over 30 years now. If AR enthusiasts experienced pain every time someone chose a different speaker from the ones we prefer, we'd all be in wheelchairs by now.

[Carlspeak]

C'mon Pete, you can't make two belittling comments and then follow it with a "it's only music lighten up people." That doesn't really work, does it?

.............. It is, as you (and others) have noted, ONLY about the music.

Yup, that's my signature line. After starting this thread and lurking on the sidlines for a few days now, I'll add a post script:

IT'S ALL ABOUT THE MUSIC - NOT SO MUCH THE SOUND OF THE MUSIC

THE MUSIC TRUMPS THE SOUND OF THE MUSIC, HANDS DOWN! Just listen to the wonderful song in the video PB has a link to in a prior post.

Debating which design of transducer is better than the other from a technical perspective is great fun for the hobby we all enjoy. I imagine similar debates have been had already in other forums on the net. As was most likely the case with those, not much will be settled here either.

[genek]

Here's an amusing exercise for anyone who might be interested...

I snapped the photo from my desk chair while playing a recording of a solo piano CD, original environment unknown because the CD notes say nothing about where it was recorded but the piano sounds as if it's in the same room with me, so my guess is that it's a studio recording rather than one made in a concert hall. The red X marks the location of the left channel AR-3a (the desk and computer monitor block it from sight), the right channel AR-3a is visible. The green O marks the spot where my ears tell me the sound is coming from if I close my eyes for a bit. The AR-6's above are off. (This is not the intended home for the 3a's, I just haven't gotten around to rearranging the sound systems since I refurbished the 3a's).

So how much direct vs reverberant sound am I hearing from each channel...?

[RoyC]

THE MUSIC TRUMPS THE SOUND OF THE MUSIC, HANDS DOWN! Just listen to the wonderful song in the video PB has a link to in a prior post.

I'll second that!

I listened to Pete's link through a pair of 10 year old $10 computer speakers...Very enjoyable!

Thanks Pete.

Roy

[Zilch]

So how much direct vs reverberant sound am I hearing from each channel...?

It's all reflections from the left speaker, since you have no line of sight. You're listening from "around the corner," basically, to that side, with attenuated highs.

Your off-axis direct sound from the right speaker appears to be at ~30° horizontal, but oblique, since you're above the tweeter axis. I'd have to measure the direct response from that vector, but assuming that tweeter is on the outside, it's likely a combination of the first and last curves I posted at P. 5, #92, i.e., the highs are attenuated ~10 dB, with a deeper notch at 8 kHz.

To get an idea of the spectral content of the first reflection, place a mirror on the right wall (or have someone move one along there) until you can see the tweeter in it from the listening position, and then measure the off-axis angle (with a protractor or trigonometry) from the tweeter to the location of the image in the mirror, and look up the curve at that angle at #92.

Those are the two most direct paths to your ear(s), it appears, and all others are uniformly attenuated at -3 dB per doubling of distance relative to those path lengths.

In combination, then, I'd estimate that you've got substantially attenuated HF from where you're listening, but there's not a heck of a lot of "sizzle" in a piano's bandwidth to begin with, so it may sound fairly realistic.

As expected, your image, to the extent that there is one, is offset to the right from where it would be if the desk and monitor were not blocking the direct sound from the left speaker (you're closer to it and on its axis, basically,) and it is probably more diffuse than it would be if you were listening to the direct field from both speakers, instead.

Put on some jazz and stand up to hear the cymbals. If they are centered in the recorded soundstage, the image should move to the left. Switch to summed mono, if you have that capability, to hear these effects more precisely, as that will center everything....

[genek]

I'd have to measure the direct response from that vector, but assuming that tweeter is on the outside, it's likely a combination of the first and last curves I posted at P. 5, #92, i.e., the highs are attenuated ~10 dB, with a deeper notch at 8 kHz.

trigonometry) from the tweeter to the location of the image in the mirror, and look up the curve at that angle at #92.

Tweeter of right speaker is on the inside.

My impression of the sound from these speakers so far has been that there's a bit too much bass because they're on the floor, but I suppose it could actually be attenuated highs. I'll reevaluate that when I get them off the floor.

[tysontom]

Here's an amusing exercise for anyone who might be interested...

I snapped the photo from my desk chair while playing a recording of a solo piano CD, original environment unknown because the CD notes say nothing about where it was recorded but the piano sounds as if it's in the same room with me, so my guess is that it's a studio recording rather than one made in a concert hall. The red X marks the location of the left channel AR-3a (the desk and computer monitor block it from sight), the right channel AR-3a is visible. The green O marks the spot where my ears tell me the sound is coming from if I close my eyes for a bit. The AR-6's above are off. (This is not the intended home for the 3a's, I just haven't gotten around to rearranging the sound systems since I refurbished the 3a's).

So how much direct vs reverberant sound am I hearing from each channel...?

genek,

I know you said this is not the intended home for your AR-3as, but it is a good example of placement that might make them sound quite bass-heavy. This is a floor-corner π/2 arrangement, and you would have at least 6 dB bass reinforcement (if not more) -- not just in the deep bass, but actually from about 500 Hz on down to resonance (43 Hz), with the usual 12-dB/octave roll off below that point. This would make the speaker’s output very strong in the deepest bass, but also it would make the speaker sound “heavy” throughout the bass range and especially on music with lots of low-bass content. Normal placement of the speakers for a 2π “loading” would be up off the floor a foot or two and out from the wall about two feet or more. This would give you flatter (but thinner-sounding) bass than the corner placement, and it would be closer to the original recording’s level of bass.

I also think that, judging by the proximity of the speakers to your desk, you would be hearing a some direct (in relation to reflected) energy, but I can’t tell how far away from the speakers you would be sitting or the size of the room.

--Tom Tyson

[genek]

I know you said this is not the intended home for your AR-3as, but it is a good example of placement that might make them sound quite bass-heavy. This is a floor-corner π/2 arrangement, and you would have at least 6 dB bass reinforcement (if not more) -- not just in the deep bass, but actually from about 500 Hz on down to resonance (43 Hz), with the usual 12-dB/octave roll off below that point. This would make the speaker’s output very strong in the deepest bass, but also it would make the speaker sound “heavy” throughout the bass range and especially on music with lots of low-bass content. Normal placement of the speakers for a 2π “loading” would be up off the floor a foot or two and out from the wall about two feet or more. This would give you flatter (but thinner-sounding) bass than the corner placement, and it would be closer to the original recording’s level of bass.

I also think that, judging by the proximity of the speakers to your desk, you would be hearing a some direct (in relation to reflected) energy, but I can’t tell how far away from the speakers you would be sitting or the size of the room.

The room is 11 ft square, with a corner cut off just to the left of the speakers for the doors. The fronts of the speakers are about 2.5 ft from the wall, and when sitting at my desk I'm about 2 ft from the opposite wall. So I'm about 6-7 ft away from speakers that are about 6 ft apart, and as the red X depicts, I have no direct view of the left speaker.

If the 3a's were off the floor on that cabinet behind them where the 6's are and I rolled my chair about 2 ft to the right I'd be in a pretty good spot, except that the room is still awfully small for 3a's and the HK amp only puts out about 60 WPC @ 4 ohms (it's roughly equivalent to the old AR amp). These are eventually destined for the living room and a more powerful piece of electronics. I don't play anything very loudly in this space with either of these speakers, so 60 WPC appears adequate here.

As far as the possiblity of attenuated treble raised by Zilch, I have the mid and hi pots about halfway between "white dot" and full increase. This is working quite well with most LPs and tone controls at flat, but for anything on CD I seem to prefer the treble control turned up about 45 degrees. Some of my jazz CDs that are "remasters" of very, very old LPs and 78s need more like 90 degrees, but they have always needed some treble boost on any system I've played them on, so I think that's just the nature of the old source material.

BTW, when listening to the 6's, the stereo image is still reasonably well centered without playing with the balance control even though I'm sitting directly in front of the left speaker and the 6's tweeter is nowhere near the level of performance of the 3a's. Another headscratcher if anyone wants to tackle it, because the 6's never did this well in any other room they've been in during 30 years or so that I've owned them.

[onplane]

The proposition that in improving imaging and soundstage resolution, constant directivity sacrifices spaciousness and ambience like headphones, as speculated by others here, clearly has no basis in fact. If AR3a is set forth as a standard for comparison, constant directivity does it better....

Zilch, you can't have it both ways.

Horns "focus" the sound and I don't care what kind of horn. None of them propagate sound in a hemisphere like domes.

This focus or directivity is what creates that better image just as in headphones. What are you talking about?

Even Toole admits that most people generally prefer direct radiators over cd waveguides.

Regards,

Jerry

[Zilch]

Tweeter of right speaker is on the inside.

O.K., the middle and right curves at #92, then. I measured both.

Horns "focus" the sound and I don't care what kind of horn. None of them propagate sound in a hemisphere like domes.

This focus or directivity is what creates that better image just as in headphones. What are you talking about?

Wrong.

I've provided the measurements of dispersion for you. That little 90° waveguide, 4430, Geddes, and a bunch of others easily outperform the AR3a tweeter with respect to dispersion.

Geddes measured the E'Waveguide; it does so, too, even in the vertical, despite being axiasymmetric 50° there. Read up the experiences of E'Wave builders. They're all astounded by its dispersion outperforming any cone/dome they've ever heard.

You presume domes radiate uniformly in a hemispheric pattern; just measure a few, or look at their off-axis performance, if published. Show me ANY dome good to half that, even.

Allison shows it for the AR3A tweeter in his Fig. 3 (1972). It crashes beyond 30°, i.e., 60° beamwidth, and that's not even with the cabinet molding blocking it.

Look at Carl's measurements of the AR3A tweeter dispersion characteristics, also. We're talking mythology here....

[genek]

Horns "focus" the sound and I don't care what kind of horn. None of them propagate sound in a hemisphere like domes.

If the attempts Carl and Roy have made in this area are any indication, we can't even find domes that propagate sound in a hemisphere anymore.

[Zilch]

Allison's 1970 paper (particularly the longer preprint version) explains why this is the case in detail.

Allison thesis relies on the Beranek model. Toole devotes an entire chapter to demonstrating that it doesn't apply in typical home listening rooms, including measurements.

Why is it, do we suppose, that the Allison's "Acoustic power" curve rolls of 13 dB in the HF, even with the mid and high controls set to "Max?" Because a dominant homogeneous and isotropic Beranek reverberant field is propping it up?

It ain't happenin'....

Perhaps you listen that way or your audio rig is installed in a padded cell.

Surely you appreciate by now, Howard, that there's not a lot of traction to be achieved in presuming me to be a dolt....

[onplane]

O.K., the middle and right curves at #92, then. I measured both.

Wrong.

I've provided the measurements of dispersion for you. That little 90° waveguide, 4430, Geddes, and a bunch of others easily outperform the AR3a tweeter with respect to dispersion.

Geddes measured the E'Waveguide; it does so, too, even in the vertical, despite being axiasymmetric 50° there. Read up the experiences of E'Wave builders. They're all astounded by its dispersion outperforming any cone/dome they've ever heard.

You presume domes radiate uniformly in a hemispheric pattern; just measure a few, or look at their off-axis performance, if published. Show me ANY dome good to half that, even.

Allison shows it for the AR3A tweeter in his Fig. 3 (1972). It crashes beyond 30°, i.e., 60° beamwidth, and that's not even with the cabinet molding blocking it.

Look at Carl's measurements of the AR3A tweeter dispersion characteristics, also. We're talking mythology here....

Who said anything about the tweeter?? I was talking about the mid dome.

The range from 600 to 5000 Hz is where most of the instruments play. Further as has been point out, majority of instruments are direct radiators.

Regards,

Jerry

[soundminded]

Who said anything about the tweeter?? I was talking about the mid dome.

The range from 600 to 5000 Hz is where most of the instruments play. Further as has been point out, majority of instruments are direct radiators.

Regards,

Jerry

I have to disagree and even where it is true, it is misleading. Horns (brass and wind instruments) are direct radiators but they are rarely if ever aimed at the audience. The bell of a tuba is pointed skyward. French horns, sideways. Trumpets and trombones diagonally downward. Clarinets, oboes, english horns straight down. Bassoons, vertically. Unless the lid of a grand piano is propped up or you are in direct line of sight of the harp, strings, sounding board, a grand piano is an indirect radiator and rather omnidirectional. Spinets and upright pianos entirely indirect. String instruments are interesting. Their strings vibrate cylindrically but contribute very little to the total sound. This can be seen in a music store with an electric violin with the amplifier turned off. You can hardly hear it from a few feet away. The box vibrates mostly front and back and its internal air is a resonant chamber with sound exiting from the f holes, usually not pointed at the audience. In a symphony orchestra, cellos and basses face sideways, violins and violas have their f holes pointed upwards. Harps are rather omnidirectional. The human voice is a direct radiator. A guitar is fairly directonal. A cello can be a fairly direct radiator when heard facing you close up in a chamber ensemble such as a string quartet.

If you want to reproduce the sound of a musical instrument as it would be heard in your room, this presents a dichotomy. One choice would be to duplicate the radiating pattern of each instrument. But they are all different. No one speaker design could duplicate more than one of them. If you want to reproduce the sound as the microphone captured it, you need a speaker whose direct and indirect radiation arrives at the listener flat. One problem for recording a piano is where to place the microphones. Too far away, and because a piano is so omnidirectional, even a cardioid mike will capture a lot of room reverberation unless the room where it is recorded is fairly dead. Highly reverberant piano recordings blur the sound of the notes and the piano sound distant and bloated. Close miked, the microphone will pick up a lot of sound the audience doesn't normally hear much if at all. These include the damper action, resonances in the harp, and sympathetic resonances of strings on keys not struck. These are plainly audible on many cds of piano. CDs are the ideal medium for recordings of pianos because of their huge dynamic range. A large concert grand like a Steinway D or a Baldwin SD-10 can play at a whisper or at nearly deafening loudness in a small room. They are designed to be able to fill large concert halls. For much music, vinyl simply can't capture them without substantial dynamic compression.