179 replies to this topic

[speaker dave]

Nearly 90 degrees off the speaker's axis in an AR3/AR3a, there is a reflector around the perimeter which in aggregate is approximately 1 square foot in area (4 foot perimeter and no more than about 3" wide) and is evenly distributed. It reflects at what appears to be about a 30 degree angle. Are you saying this reflector adds 10 db to the speaker's output at any frequency compared to the drivers without this reflector? I find that hard to believe. Certainly the tweeter's output 90 degrees off axis is relatively small compared to its forward output. To the degree that these reflections exist, they will create multiple comb filtering, so many because each point along the reflector is another source that the interval of frequency nodes and number of nodes at any given point in space are so numerous that will tend to cancel each other out. At least that is the theory. That is why I asked for any reference to measured data. It strikes me that like many audiophile notions, this type of generalization is not only theoretically flawed, it is not proven out by experience. This is typical for the arguments for wires, vacuum tubes, vinyl phonograph records, and many other audiophile notions that run contrary to factual knowledge. In fact audiophile notions generally fly directly in the face of real scientific knowledge and the experience of engineers even if the audiophiles have gone into the manufacturing business and have products to sell.

You need to read what I said again. I was refering to Olson's study and his measurements of a small driver on the face of a variety of cabinets. Worst combinations, with the unit in the center of a highly symmetical cabinet, gave 10dB peak to trough. More typical cases gave 2-3dB. These are diffraction related edge reflections alone. Perimeter dams would be a seperate additional issue. I'm not sure I would lump these in with cables and vacuum tubes. Olson didn't strike me as a tweak..

Reflections off the floor are critical to the sound of many musical instruments. For example, many wind instruments like clarinets and oboes direct their sound almost exclusively at the floor. The way in which the performing stage is constructed can directly affect the bass response of an auditorium as it is a critical reflector of large area near the sources of sound. Cancellations due to the Allison effect can easily be equalized out.

Which is why the oboes are at the front of the Orchestra....oops. Yes, musicians need to play in reverberent space. That doesn't mean that similar reflections are good for the accurate reproduction of sound. I like the studies of Soren Bech where he showed that the floor bounce is the most likely, after the back corner reflections, to be audible, i.e.detrimental.

One Allison like effect is the odd absorption characteristic of the audience to grazing waves across. I'm sure you have read about that. Within the orchestra shell don't you have a large quantity of sources all at varied distances from the closest boundary? Hard to see how that would give any uniform cancelation.

Diffraction around the back and sides of cabinets is of far greater concern at low frequencies than at high frequencies. This is only one reason why a forward firing speaker will sound muffled from behind. Only the lower frequencies radiate omnidirectionally to the sides, top, bottom and back of the cabinet and create reflections within the room that drastically alter the tonal balance of the speaker. This is why there is little correlation between the on axis anechoic FR and the real room FR when all of the sound reaching the listener including the reflections are collected. Moving the speakers away from the walls and adding sound absorbing material only marginally mitigates the problem.

"We tend to be a little sloppy with our terminology."

I've noticed. It hardly stops there though.

Depends on what you mean by "greater concern" we were talking about reflection related response aberations. The general directivity rise of speakers with frequency is another debate. If you want to see and hear the effects of edge reflections then they are most likely to be visible at mid and high frequencies. Distance of the driver to the edge determines the delay time for alternate paths and hence the wavelengths of cancelation or addition. Low frequency wavelengths are too long, the waves diffract but the reflections don't cause cancelation.

[soundminded]

I hadn't commented on diffraction, but since you ask...rather than wild conjecture, let me suggest an experiment. Have a friend hold a ruler with its long edge on the surface of you favorite system (single tweeter, in this case). The ruler is perpendicular to the surface and will create a reflective barrier. Play pink noise and have that person slide the ruler a little closer and a little farther from the tweeter. I guarantee you will hear a comb filtering effect that will go up and down in pitch with as the distance to the tweeter is varied. Now the remaining question is what reflection levels and what time delays are audible? Does the direction of arrival of a reflection have and effect? Let me recommend Toole's book, which has over a chapter on the subject, with his and others findings.


Yes, most baffles aren't truly infinite as that tends to take up a lot of space. A large enclosure is typically called an infinite baffle in our industry. I suppose you could qualify that as a low alpha system or one that had a small effect on the free air resonance of the driver.

You really are confusing breakup modes with nonlinear distortion elements. The primary elements of driver distortion are coil length vs. gap length (Bl vs. x), spider and surround nonlinearity and magnetic hysteresis effects. With those alone you can do a pretty realistic model of woofer nonlinearity. Klippel is very good in this area, if you want to read something.

I really don't understand the rest of your argument, sorry.

Regarding the light bulb: a bit of hyperbole on my part. You are correct in that he added a third element to a diode, not that you couldn't read a book from the light of some bright emitter triodes.

David

"Play pink noise and have that person slide the ruler a little closer and a little farther from the tweeter. I guarantee you will hear a comb filtering effect that will go up and down in pitch with as the distance to the tweeter is varied."

Pink noise in itself does not have an identifiable pitch. Pink noise is similar to white noise which is a combination of all frequencies at equal amplitude analagous to white light being a combination of all colors or light frequencies in the visible spectrum. Pink noise puts white noise through a low pass filter that reduces output as frequency rises under the rationale that this somehow mimics sensitivity of human hearing or music. The tilt of the amplitude versus frequency curve or rate of decrease is the weighting. Comb filtering of pink noise unless the filtering action is so sharp and the peaks are harmonically related will not yield an identifiable pitch, therefore unless the response is measured during this test, it will not be noticable. Such changes in pitch if they occur are due strictly to the doppler effect and will revert to their original frequency once the ruler stops moving. Changes in spectral balance are a different matter. As the reflective surface may selectively reflect some frequencies to the preference of others due to the nature of the material, the brighness or timbre of what is no longer truly pink noise can subjectively change. However, when the reflection is a relatively large surface compared to the dimensions of the wavelengths at issue such as a foot or two of reflective surface where the wavelength is a few inches, the number and distribution of cancellations and reinforcements at any point in space will be so numerous as to make them in all likelihood inaudible.

While I don't discount all of Olsen's contributions, he's made mistakes. For example, his model of a pulsating sphere as the ideal loudspeaker in a real room is wrong. He is also not the inventor of the acoustic suspension loudspeaker system, the court decision notwithstanding. They got it wrong. I'd still like to see how the perimeter molding of an AR3 can increase or decrease the sound field of any of the drivers at any point in space at any frequency by 10 db. I'd like to see how it could change it by even 1 db.

[speaker dave]

Pink noise in itself does not have an identifiable pitch. Pink noise is similar to white noise which is a combination of all frequencies at equal amplitude analagous to white light being a combination of all colors or light frequencies in the visible spectrum. Pink noise puts white noise through a low pass filter that reduces output as frequency rises under the rationale that this somehow mimics sensitivity of human hearing. The tilt of the amplitude versus frequency curve or rate of decrease is the weighting. Comb filtering of pink noise unless the filtering action is so sharp and the peaks are harmonically related will not yield an identifiable pitch, therefore unless the response is measured during this test, it will not be audible. Such changes in pitch if they occur are due strictly to the doppler effect and will revert to their original frequency once the ruler stops moving. Changes in spectral balance are a different matter. As the reflective surface may selectively reflect some frequencies to the preference of others due to the nature of the material, the brighness or timbre of what is no longer truly pink noise can subjectively change. However, when the reflection is a relatively large surface compared to the dimensions of the wavelengths at issue such as a foot or two of reflective surface where the wavelength is a few inches, the number and distribution of cancellations and reinforcements at any point in space will be so numerous as to make them in all likelihood inaudible.

Thank you for explaining pink noise to me.

Comb filtering places a series of peaks and dips into the spectrum. They would be equally spaced on a linear scale but always get closer together at high frequencies when portrayed on a log scale. At very high frequencies they fall within a critical bandwidth and can generally be ignored.

Perceptually I always hear comb filtering as a single distinct pitch. I don't know if this corresponds to the band edge of the first hole or the first return or peak at the frequency just above, but it is an effect I have heard times to numerous to count. It is a stationary effect. If the ruler stops the pitch of the effect is held because the comb filtering is set by the relative timing of the multiple paths in effect. Same with a floor bounce and deep knee bends: the pitch you hear is a function of the distance your ears are above the floor. If you move slowly or stop moving, the effect doesn't go away as it would with a doppler effect. Same with the Alllison rear wall dip. If you moved a system slowly back and forth relative to the wall, you would hear and be able to measure a varying notch frequency. It doesn't dissapear if the system stops moving.

Regarding the edge reflection and diffraction effects of an AR3, I would refer you to the measurements already provided in the Allison/Berkowitz paper. They are exactly that and show well more than "1 dB" effect.

David

[genek]

In fact audiophile notions generally fly directly in the face of real scientific knowledge and the experience of engineers even if the audiophiles have gone into the manufacturing business and have products to sell.

I'm thinking this should probably read "especially if..." rather than "even if..."

[speaker dave]

I did the diffraction evaluation procedure Speaker Dave advocated and came up with results that showed that diffraction is no big deal, at least not for me. Maybe you should also give it a try. Remember, you need to find somebody to move the ruler or yardstick while you sit off at a regular listening distance. No arm-length stuff, at least without doing it at the longer range, too.

Cannot find anybody to drop by and do the holding job for you? Call up "Dial A Ride" or "Meals on Wheels" and see if they will send somebody over.

Howard Ferstler

Sad, when a man has to hold his own stick.

David

[genek]

Sad, when a man has to hold his own stick.

I was considering closing this thread down until you said that.

[speaker dave]

On the other hand, regarding the actual Allison boundary effect in the direct and reverberant fields, the suckout notch will exist NO MATTER WHERE the listener sits, because the cancelling is happening at the speaker position and NOT the listening position. If the distances to those multiple large boundaries are staggered adequately the suckout impact is minimized, and this is why some people have no problems with it - even accidental problems. On the other hand, if the distances of two or three boundaries to the woofer centers are similar then the bounce-back to the speaker cancellation dip can get pretty emphatic, and this will have a singular impact on the sound of instruments like the cello. Some people laud the supposed tightness they hear in the midbass when this problems exists, but what they are actually hearing is a weakness in midbass reproduction. More than one reviewer has commented about how the bigger Allison speakers tend to have a mid-bass fullness that some like and some do not. That fullness, is, believe it or not, the speaker inputting flat mid-bass power to the room. Most systems with woofers mounted away from large room boundaries cannot do this unless electrical equalization is applied.


Howard Ferstler

Interesting summary, Howard, especially regarding the 2-4-2pi effect.

I'd quibble with the "Happens at the speaker" distinction you make. Clearly, if you moved from in front of the speaker around to the side, the distance or time difference between the real speaker and the behind-the-boundary virtual speaker that together make the Allison dip, would reduce. At 90 degrees, as you approached the wall, the dip would disappear. This is much like taking a boundary plain measurement: when you place the mic on the ground the cancelation notch disappears that would bee seen at all other angles.

Allison shows a power curve in one of his papers with the notch clearly in the power curve, but this is to be expected even if the notch doesn't exist at some angles. The power curve is no different than an average of all response curves in a sphere (or in this case a half or quarter sphere) around the sytem (Strictly speaking it is an average of pressure square multiplied times sample area). For the great majoritiy of forward angles the geometry is relatively constant and the dip is present. This shows in the average, even though the right angle curves have no dip. So it is possible for the radiated power to have the dip without it being present "everywhere".

David

[tysontom]

Well, according to Tom Tyson Villchur gave 75 of those concerts. I suppose he should have gotten different ensembles for even more instrumental variety, but, heck, the project was tedious enough as it is. It would also have been interesting to see how some other-brand speakers handled the presentations (and for all we know he may have done that in private, even with other AR speakers), but I think that Villchur would have been loath to do something like that as a way to show the competing designs as inferior. He did not care about lambasting the competition or putting it on the spot in that way. Given how different most competing speakers sounded from the AR-3 during that era, it is likely that no other speakers could have done the job as well with the LvR presentations.

According to Roy Allison, all Villchur had to normally do to balance the curve (depending on the hall used) was adjust the treble output of the Dynaco preamp.

Well, I as being a tad unfair when I used "popularity." However, the best Toole could do (or anybody else has done, as best I can tell) is use commercial recordings that add in their own variables. It is quite normal for a pair of speakers that sound like winners with one recording when participating in an A/B comparison to come off second best when a somewhat differently recorded recording is then used. I said as much in a Stereo Review artcle years ago.

Howard Ferstler

Howard,

AR conducted at least 75 live-vs.-recorded concerts across the country (Chicago, LA, Washington, NYC, Philadelphia, etc.) with the Fine Arts Quartet alone; additional concerts were performed with guitarist Gustavo Lopéz, and other concerts were conducted with the Seaburg nickelodeon in 1966. Mattes Electronics, builder of the first high-power solid-state amplifier, compared the oscillograph output from a Steinway Grand piano and the output waveform of that same piano being fed into an AR-3 -- showing the almost identical waveform reproduced by the AR-3 (at the time Mattes did not disclose the manufacturer of the speaker), but this wasn't a LvR, per se, and not billed as one.

The purpose of the live-vs.-recorded concerts, however, was not to demonstrate the veracity of the process itself of comparing live music with reproduced music -- that was a "given." There was no attempt to put on a special show to demonstrate various speakers and let the audience determine which was "best." This series was done simply to demonstrate the AR-3's faithful-sound reproduction, especially considering all the critical acclaim and press fanfare. Comparing live music with instantaneously switched recorded music is an ultimate subjective test of faithful-sound reproduction -- few (save Zilch & Co) would argue that fact. Previously, there had been much hoopla by the press and critics alike that AR (nearly alone in doing this) spent too much time printing distortion and frequency-response curves and graphs, and many critics felt that the only way one could determine the true quality of a speaker was through listening tests -- to some the "golden-eared" viewpoint. For this reason, Villchur felt that the true test for subjective accuracy would be the instantaneously switched, live-vs.-recorded comparison of live music with reproduced music through the AR-3. It completely silenced the critics.

The only level control adjustments were to the treble control of the Dynaco preamp, and these were done to adjust the balance in different playback venues. In some cases the treble had to be increased; for some of the guitar playback, the treble control had to be attenuated. There was no FR equalization whatsoever.

By the way, the other "big" so-called live-vs.-recorded concerts were conducted by Gilbert Briggs of Wharfedale in the 1955-1956 time-period. These recordings were not done in anechoic space; moreover, Briggs was merely trying to show the ability of his speakers to play back the full power of a symphony orchestra, but no instantaneous A-B comparison were made as in the case of the AR LvRs.

--Tom Tyson

(Img. 1) Fine Arts Quartet recording at Woodstock (Leonard Sorkin, violin; Abram Loft, violin; Gerald Stanwick, viola; George Sopkin, cello) and AR-3s -- standing: press visitor; Ed Villchur; Roy Allison; Villchur's admin assistant "Sunny."

(Img. 2) Gustavo Lopéz with AR-3s during session at Carnegie Rehersal Hall, NYC.

(Img. 3) 1919 Seaburg Nickelodean -- AR-3s.

[tysontom]

Howard,

AR conducted at least 75 live-vs.-recorded concerts across the country (Chicago, LA, Washington, NYC, Philadelphia, etc.) with the Fine Arts Quartet alone; additional concerts were performed with guitarist Gustavo Lopéz, and other concerts were conducted with the Seaburg nickelodeon in 1966. Mattes Electronics, builder of the first high-power solid-state amplifier, compared the oscillograph output from a Steinway Grand piano and the output waveform of that same piano being fed into an AR-3 -- showing the almost identical waveform reproduced by the AR-3 (at the time Mattes did not disclose the manufacturer of the speaker), but this wasn't a LvR, per se, and not billed as one.

The purpose of the live-vs.-recorded concerts, however, was not to demonstrate the veracity of the process itself of comparing live music with reproduced music -- that was a "given." There was no attempt to put on a special show to demonstrate various speakers and let the audience determine which was "best." This series was done simply to demonstrate the AR-3's faithful-sound reproduction, especially considering all the critical acclaim and press fanfare. Comparing live music with instantaneously switched recorded music is an ultimate subjective test of faithful-sound reproduction -- few (save Zilch & Co) would argue that fact. Previously, there had been much hoopla by the press and critics alike that AR (nearly alone in doing this) spent too much time printing distortion and frequency-response curves and graphs, and many critics felt that the only way one could determine the true quality of a speaker was through listening tests -- to some the "golden-eared" viewpoint. For this reason, Villchur felt that the true test for subjective accuracy would be the instantaneously switched, live-vs.-recorded comparison of live music with reproduced music through the AR-3. It completely silenced the critics.

The only level control adjustments were to the treble control of the Dynaco preamp, and these were done to adjust the balance in different playback venues. In some cases the treble had to be increased; for some of the guitar playback, the treble control had to be attenuated. There was no FR equalization whatsoever.

By the way, the other "big" so-called live-vs.-recorded concerts were conducted by Gilbert Briggs of Wharfedale in the 1955-1956 time-period. These recordings were not done in anechoic space; moreover, Briggs was merely trying to show the ability of his speakers to play back the full power of a symphony orchestra, but no instantaneous A-B comparison were made as in the case of the AR LvRs.

--Tom Tyson

(Img. 1) Fine Arts Quartet recording at Woodstock (Leonard Sorkin, violin; Abram Loft, violin; Gerald Stanwick, viola; George Sopkin, cello) and AR-3s -- standing: press visitor; Ed Villchur; Roy Allison; Villchur's admin assistant "Sunny."

(Img. 2) Gustavo Lopéz with AR-3s during session at Carnegie Rehersal Hall, NYC.

(Img. 3) 1919 Seaburg Nickelodean -- AR-3s.

I failed to upload the 1910 Nickelodeon image; so while I'm at it, I've attached another related LvR image.

--Tom Tyson

[Steve F]

I failed to upload the 1910 Nickelodeon image; so while I'm at it, I've attached another related LvR image.

--Tom Tyson

I've heard it said that the Nickelodeon L v R's were not quite the overwhelming successes that the FAQ L v R's were. Is that true? If so, why do you suppose that is? Is it because the Nickelodeon had far more percussive/cymbal-type sounds than the FAQ that the AR-3 was less well-equipped to replicate?

Also, I wonder if anyone has any comments on AR's class in mentioning other companies (Wharfedale and KLH) in their ads and brochures, as I mentioned in post #191 of this thread?

Steve F.

[Zilch]

I did the diffraction evaluation procedure Speaker Dave advocated and came up with results that showed that diffraction is no big deal, at least not for me.

That is not responsive to the question. You are arguing that the LvR demos, having been performed in larger spaces, would exhibit less of a contribution from reflections than a typical home listening room, emphasizing the direct field, and thus diffraction effects, if they were significant, should have been audible.

My question is, since we see that the front wall is draped, and thus, presumably, damped, would that not also contribute to emphasizing the direct field, as you suggest is the inevitable consequence of a similar condition in the NRC listening room?

[Zilch]

Keep it up, Howard, if you want Gene to close this thread....

[tysontom]

I've heard it said that the Nickelodeon L v R's were not quite the overwhelming successes that the FAQ L v R's were. Is that true? If so, why do you suppose that is? Is it because the Nickelodeon had far more percussive/cymbal-type sounds than the FAQ that the AR-3 was less well-equipped to replicate?

Also, I wonder if anyone has any comments on AR's class in mentioning other companies (Wharfedale and KLH) in their ads and brochures, as I mentioned in post #191 of this thread?

Steve F.

Steve,

The Nickelodeon LvR demonstrations weren't as notable as the Fine Arts LvR concerts, but part of the problem was the recording itself and the mechanical music belt used on the instrument. The 1910 Seaburg Nickelodeon was bought in upstate New York in 1965 for around $1500. Part of the purchase included a complete restoration, but the music belt tended to slip a bit, and this might have affected the ability to synchronize the recording. For this recording, AR used a Magnecord 1028B, a very fine semi-pro broadcast stereo tape recorder, but not quite on the level of the Ampex 350-series used earlier. Nevertheless, the percussion part was handled with relative ease by the AR-3; the AR-4x did not do as well. I also don't believe that the recording for the Nickelodeon demonstration was performed by the famous Jan Syrjala ("sajalla") -- an absolute recording "genius" who managed all the Fine Arts and guitar recordings.

--Tom Tyson

[soundminded]

I've heard it said that the Nickelodeon L v R's were not quite the overwhelming successes that the FAQ L v R's were. Is that true? If so, why do you suppose that is? Is it because the Nickelodeon had far more percussive/cymbal-type sounds than the FAQ that the AR-3 was less well-equipped to replicate?

Also, I wonder if anyone has any comments on AR's class in mentioning other companies (Wharfedale and KLH) in their ads and brochures, as I mentioned in post #191 of this thread?

Steve F.

Since I attended both the guitar and nickelodeon demos, I'm in a position to make some comments but it does tax my memory. It was after all over 40 years ago.

"Is it because the Nickelodeon had far more percussive/cymbal-type sounds than the FAQ that the AR-3 was less well-equipped to replicate?"

No, the AR3s reproduced the percussive sounds very well IMO. (theoretically this is not surprising since transient response is directly related to and interchangable with high frequency response.) Nor was the demo a failure at all. In fact IMO it was quite successful. The Nickelodeon was against a wall. The AR3s were on the floor flanking the Nickelodeon and the AR4xs were on the floor flanking the AR3s. Everything was with its back against the wall.

The AR3s were not at the same elevation as the Nickelodeon which of course mattered. Also, unlike the guitar which was out in the center of a large room where there wasn't much in the way of early reflections there was in the case of the Nickelodeon. So there were differences in the interactions with room acoustics that would make differences in the propagations much more audible. What was surprising though was not that these differences existed but how relatively small and unimportant they seemed. It was the similarity that struck me. Even the AR4xs did surprisingly well although not quite up to the AR3. But the difference would hardly make you guess that there was a fourfold difference in price.

I'm sure we all wish we could hear the demos again, I know I do. While my hearing may have been marginally better then than it is now, my listening ability and attention to details is far greater now than it was in those days. I'm a much more critical and experienced listener. I would be able to glean much more about where the speakers succeded and where they failed to match the live instrument.

As I said in other postings, I heard the guitar and it was also not a exact match, the speakers were very slightly brighter. I was directly on axis in a center aisle seat as I recall. This difference was immediately surprising to me because I usually considered AR speakers to sound muffled with their high end rolloff. That is the way my AR2axs sound even with the tweeter at maximum if the treble isn't also equalized. The demos convinced me that AR3 was the state of the art and it made me wonder if anything different wouldn't necessarily be inferior. Until I heard about the treble control on the Dynaco Preamp being turned up, I had no explanation for why the AR3s always sounded like they had a dull high end when I listened to recordings using them. Many people seemed to have an obsession about keeping those controls flat. I realize later on that this was nuts. With the treble control set flat, in direct comparison between AR3 and KLH Model 17, the KLH speaker easily was brighter and clearer than AR3. AR3 killed KLH Model 17 in bass frequencies though and that difference was also obvious.

[Zilch]

Again, while you and some others feel that speakers should do much the same thing as headphones (and nobody here will dispute the ability of good headphones to deliver superb detail and clarity, beyond even what one will hear at live concerts), my belief is that speakers, sited in good rooms, should work to deliver a simulation of a live performance in a good hall (or club, or auditorium).

You continue to mischaracterize what we advocate, which is NOT narrow dispersion. Just as Toole describes, there is an optimum balance between spaciousness and localization that is most conveniently accomplished using moderate-dispersion constant directivity, the accepted standard being approximately 90° as first pioneered by CSP's own Speaker Dave fully thirty years ago.

If the performance standard is the sonic mush which comprises replication of the acoustic concert hall "realism," then yes, wider dispersion works, to the detriment of soundstage and imaging, and also requires commitment to the requisite listening room design. For the 96% of others who enjoy more diverse musical fare, however, there are more viable options....

[Zilch]

What you and Toole advocate is narrow dispersion by my standards.

As I recall, we previously agreed that wide dispersion began at 120°. It's real easy - 0 -60, 60-120, 120 - 180, 180+.

Geddes agrees with you, actually; he calls his designs "narrow," and when I attempted to correct him on that, he advised, in his case, 90° refers to the flare angle, the actual -6 dB dispersion being 70°. Go figure.

[He's not buying my suggestion that narrow is 60° or less, either, obviously. ]

Actually, my own home-built jobs are certainly narrower in broad-bandwidth dispersion than my IC-20 models, and with some recordings, especially smaller-scale stuff, I think that my units actually sound better.

Who'd have guessed...?

And, sure, in that particular picture the drapes will obviously absorb some sound, although not much treble energy gets behind a speaker in the first place. The Toole room, as shown in his diagrams, also had drapes partially down the side walls, and those drapes WOULD have an impact. The photo of the LvR concert does not show the side walls, so we have no idea if they were draped. Also, as I have noted, the AR-3 is a more directional speaker than, say, the later AR-3a, AR-LST or any Allison model, and so in an auditorium environment it would be able to aim more direct-field energy and not have it dissipated by open space (or drapes) well to the sides.

Again, the Toole room drapes stop at the speakers, and do not damp the lateral reflections.

SO, we now argue that the success of Villchur's LvR presentations was in some significant part attributable to dominance of the direct field.

COOL!

[soundminded]

As I recall, we previously agreed that wide dispersion began at 120°. It's real easy - 0 -60, 60-120, 120 - 180, 180+.

Geddes agrees with you, actually; he calls his designs "narrow," and when I attempted to correct him on that, he advised, in his case, 90° refers to the flare angle, the actual -6 dB dispersion being 70°. Go figure.

[He's not buying my suggestion that narrow is 60° or less, either, obviously. ]



Who'd have guessed...?



Again, the Toole room drapes stop at the speakers, and do not damp the lateral reflections.

SO, we now argue that the success of Villchur's LvR presentations was in some significant part attributable to dominance of the direct field.

COOL!

"SO, we now argue that the success of Villchur's LvR presentations was in some significant part attributable to dominance of the direct field.

COOL! "

What a strange lot audiophiles are. Among the advocates of narrow dispersion tweeters, one poster tells me that the reflection of a perimeter wood molding on the baffle only a couple of inches wide will change the output 10 db and create horrible distortion while another such advocate says that a speaker with a tweeter whose highest octave starts falling off appreciably just a few degrees off axis can successfully emulate the sound of a guitar whose radiation of all of its sound including higher harmonics is so wide, you practically have to stand behind the musician before enough harmonics are blocked to change the timbre. And you expect me to believe that an audience over a fairly wide angle will not hear this difference. Yes a strange lot.

[speaker dave]

What a strange lot audiophiles are. Among the advocates of narrow dispersion tweeters, one poster tells me that the reflection of a perimeter wood molding on the baffle only a couple of inches wide will change the output 10 db and create horrible distortion .

So we are making up facts now to support our argument? Find me the quote please.

[speaker dave]

I was looking at Olson's study on enclosure diffraction/reflection and he showed a worst case peak to trough of 10dB (I'm guessing that would be audible) for drivers mounted dead center in a square or circle. Typically 2 to 3 dB effect for other cabinets.

David

And then:

You need to read what I said again. I was refering to Olson's study and his measurements of a small driver on the face of a variety of cabinets. Worst combinations, with the unit in the center of a highly symmetical cabinet, gave 10dB peak to trough. More typical cases gave 2-3dB. These are diffraction related edge reflections alone. Perimeter dams would be a seperate additional issue. I'm not sure I would lump these in with cables and vacuum tubes. Olson didn't strike me as a tweak..

I twice explained that the reference was to Olson's experiments with cabinet diffraction which showed as much as 10dB peak to trough ripple.

I guess reading comprehension is a problem for some.

David

[soundminded]

And then:

I twice explained that the reference was to Olson's experiments with cabinet diffraction which showed as much as 10dB peak to trough ripple.

I guess reading comprehension is a problem for some.

David

"I guess reading comprehension is a problem for some."

I'd guess extreme exaggeration, unwarranted generalizations, and focusing minutely on the irrelevant are problems for others.