Jump to content

Allison on Soundfields


Howard Ferstler

Recommended Posts

I made mention of this in another part of this thread, but I do want to restate a point or two here in another form. My take on this prejudice against super-wide broad-bandwidth dispersion (well into the treble range, and well beyond what is possible with typical on-inch tweeters) by Zilch, Toole, and Geddes involves them never possibly hearing proper wide-dispersion speakers in proper listening rooms.

I know that Consumer Reports Magazine used AR-LST models as their reference speakers for years (they gave them 95% accuracy ratings, which is the highest they ever got with speakers that I know of) and I know that Toole did at least hear them when he visited their facility some years back. However, he did not carefully analyze them (this is what he wrote to me), and so we do not know how they would have fared in those many listener-panel sessions he pioneered when at the Canadian NRC or at Harman. Nearly ALL of the speakers he auditioned were either very directional (like the classic Quad models) or conventionally directional, like a number of Harman-built speakers that he obviously helped to design. This is one of the criticisms I have leveled at his book, because nearly every analysis he did involved basically conventional speakers located in away-from-front-wall placement situations. Toole may dismiss super-wide dispersion speaker types (everything from the LST, through the Allison models, to what Mark Davis did with his original Soundfield design), but in his book he never even documents his reasons for not liking such an approach. He even showed an illustration of Roy Allison's prototype Model One, noted that it might be an interesting design, and then went on to not bother to hunt up a sample to audition and evaluate.

I am not sure about Zilch's at home listening experiences, since he continues to refuse to give us any input about what specific models he has auditioned (not just in stores, which are usually lousy auditioning environments, but at his own place), and I am not sure about Geddes, either. For all we know, all he has listened to were different horn designs as he worked to come up with his waveguide approach. We do not know about his listening tastes at all, and for all any of us know he listens in a padded room at home, with a single chair locked solidly into the sweet spot. Maybe Zilch listens this way, too.

The AR-3a is not a super-wide dispersing speaker. It has drivers that can approach that ability, but the forward-facing single-panel design, not to mention cabinet edges and the way the drivers are arranged on the cabinet, panel does result in wide but not super-wide dispersion. AR only got to that level with Roy Allison's LST model. But the AR-3a is certainly wide-dispersing enough to deliver a sense of spaciousness with many recordings that allows it to hold its own with many of the best conventional speakers available today. That it appears to sound "dull" to some listeners (like the earlier AR-3 did to some of the people Toole mentions in his book) is mainly the result of the dialed-in upper-midrange and treble rolloff. That rolloff is there to let the speakers deal sanely with many of the classical releases within the era the speaker occupied, and ironically the approach works very well with many similar recordings showing up in brand-new CD form right now.

As I have mentioned on one or two other posts here, the Allison IC-20 model was the only speaker I have encountered that allowed one to vary the dispersion angle (the radiation pattern) of the speaker on the fly, as the music played. With that feature it was easy to see the advantages of super-wide dispersion with typical good recordings of acoustic instruments spread out on a concert-hall stage. As for your comment about me "looking elsewhere," I also have had many opportunities to do level-matched A/B comparisons between broad-bandwidth, ultra-wide dispersing speakers like those IC-20 models and narrower-dispersing models like the Dunlavy Cantata and SC-II systems, which, by the way, made use of Dunlavy's own waveguide designs for the upper midrange and treble output. While the latter would always have a slight (and I mean slight) edge when it came to detail and imaging (and I certainly can see the attraction with many different kinds of recordings and recording techniques, as well as certain kind of ensemble and soloist presentations), the wide-dispersing Allison models always had an edge when it came to simulating a real-world, live-instrument soundstage and sense of hall space. The Allison approach also allowed for a better sense of soundstage space and reverberant-field smoothness from listening positions away from the sweet spot that focussed-design speakers nearly always must have to deliver in-spec performance.

Howard Ferstler

There's been an awful lot of talk here about direct and reverberant sound fields in home listening rooms. At one extreme is a speaker playing at one end of a hollow tube in an anechoic chamber. Every sound you hear from it is in its directly propagated field. At the other extreme is a room made of the most acoustically reflective material available, say glass on cement where the speaker faces away from you and there is a cement wall between you and the speaker. Every sound you hear from it is reflected. All of us have rooms and speakers somewhere between these two extremes. Where that point lies depends on the particulars of your room, your speakers, where they are located, and where you are located. I'd say generalizing is impossible and unless someone has a lot of measurements to back it up, much of what is being written is more speculation than fact.

At a live concert in most concert venues where there is no electronic amplification system, most of the sound the audience hears is reflected. The same variables of location of the source, the nature of the source, its relative orientation, and where the listener is located affects the precise relationship. Other than that, all one can say is that the parameters of these relationships in a live venue have yet to come close to being duplicated in a home listening room no matter how live or dead the acoustics of the listener's room, no matter how directional or non directional the speakers, no matter how close up or far away from the speakers you listen to them at.

Link to comment
Share on other sites

  • Replies 182
  • Created
  • Last Reply
I can see your point. However, I am curious about what non-directional speakers you have listened to. The AR-LST perhaps? The Bose 901 is not really non directional, since it beams in three directions. For me, a truly non-directional speaker, at least if it is aiming away from the front wall in the front hemisphere, is a lot wider radiating than your typical forward-facing speaker. Usually, the latter is beaming aplenty in the top two octaves, and it may have varying dispersion as the drivers involved deal with frequencies having varying wavelengths over the driver operating ranges.

In any case, even super-wide dispersers like my IC-20 units work better if supported by additional channels. They also work better if a center channel is handling the center imaging.

Howard Ferstler

"I am curious about what non-directional speakers you have listened to"

All sound sources including loudspeakers have a directional propagation signature that is a function of frequency. That's where my analysis begins. If you are talking about speakers that propagate sound mulitdirectionally in a frontal hemisphere, I heard Allison Ones briefly at a trade show many years ago, LST briefly on several occasions probably at AR demo rooms, and within the last ten years a couple of Soliloquy models. They do seem superior to conventional monopole box speakers in some respects. But whether the attempts to increase what some call spatial perception, others call imaging is tried through increased dispersion, dipole radiation, rear firing drivers, or other means, the intended effects they produce seem to me to be the result of intuition rather than analytical reasoning and their effectiveness is both highly variable depending on room acoustics and program material and very limited compared to actual music, especially at large venues.

I've been reading Helmut Wittick's PHD thesis on Perceptual differences between wavefield synthesis and stereophony. It's interesting but he seems as obsessed with this imaging problem as audiophiles are. It seems to be the point of his thesis. I get very nervous when people who start out with a semingly analytical presentation and then start throwing around words like gestalt. At least Wittick isn't hitting his head against the same old brick wall others who have trod similar paths have. He and those intrigued by this idea seem to have found their own brick wall. So far his wavefield theory strikes me as less than earthshattering and he seems unsure of it himself; "Wavefield synthesis is a significant step forward from stereophonic sound reproduction. It offers a noticeable enhancement of the sound field’s spatial properties. Nevertheless, there seems to be a broad lack of clarity about the perceptual benefits of WFS. As a consequence, these benefits of WFS may be underestimated, or more likely overestimated." I'm not aware of any commercial products along the lines of this theory yet with the possible exception of a sort of line array speaker system/HT receiver I think Yamaha is selling. I'm not quite sure how it is supposed to work. One thing Wittick seems at least concerned about is timbre. It's one of the few times I've even heard it discussed in audio circles lately.

Speaking about timbre, I've been working on my sensitivity to it over the last 20 or so years ago and have been listening to lots and lots of violins, mostly instruments brought here by students they've borrowed for evaluation from violin makers, dealers, and from commercial sources like Southwest Strings. Most were in the $1500 to $3000 range. We've seen a few more expenisve ones, had some of our own appraised at Sothebys (none turned out to be valuable, wrong pedigree) and we've visited a couple of good violin makers workshops. We may be borrowing a very expensive one, someone's spare which needs playing on. That will probably depend on what our insurance company says about coverage. Anyway the point of this was something a friend suggested that we check out. We visited this website last night.

http://www.luisandclark.com/index.php

There are a couple of videos on how these instruments are made, lots of impressive endorsements, and some recordings of performances on them. See if you can figure out what we liked about them and what the fatal flaw was in their sound we noticed immediately that disqualified them in our eyes. Price in relationship to what was suggested of their capabilities to us was more than modest, we expect the next one we buy will be considerably more expensive.

Link to comment
Share on other sites

I've been reading Helmut Wittick's PHD thesis on Perceptual differences between wavefield synthesis and stereophony. It's interesting but he seems as obsessed with this imaging problem as audiophiles are. It seems to be the point of his thesis. I get very nervous when people who start out with a semingly analytical presentation and then start throwing around words like gestalt. At least Wittick isn't hitting his head against the same old brick wall others who have trod similar paths have. He and those intrigued by this idea seem to have found their own brick wall. So far his wavefield theory strikes me as less than earthshattering and he seems unsure of it himself; "Wavefield synthesis is a significant step forward from stereophonic sound reproduction. It offers a noticeable enhancement of the sound field’s spatial properties. Nevertheless, there seems to be a broad lack of clarity about the perceptual benefits of WFS. As a consequence, these benefits of WFS may be underestimated, or more likely overestimated." I'm not aware of any commercial products along the lines of this theory yet with the possible exception of a sort of line array speaker system/HT receiver I think Yamaha is selling. I'm not quite sure how it is supposed to work. One thing Wittick seems at least concerned about is timbre. It's one of the few times I've even heard it discussed in audio circles lately.

I agree with several of your observations on Wittick's thesis. I suspect that a fair amount of "wavefield" research is funded by organizations interested in "virtual reality" and teleconferencing-type applications. That might explain some of the emphasis on imaging and localization.

-k

Link to comment
Share on other sites

Zilch, have you (or Geddes, too, for that matter) ever listened to a set of really wide-dispersing, flat power responding, primary-channel speakers (wide dispersing well into the top octave, with a strong off-axis response that is smooth) in a really good room? You guys seem to be speculating abpit wode-dispersion sound when the only experience you may have had involves "traditional-design" speakers that are maybe wide dispersing in the bass and lower midrange, but with progressively stronger beaming as the frequency climbs. Most speakers are built that way, because their designers are stuck with OEM midrange and treble drivers that can do no better.

Wide and ultra-wide dispersion is easily accomplished using multiple constant-directivity horns or waveguides, much as LST and Allison did with multiple dome drivers, and it's commonly done in SR installations to achieve wide coverage patterns; the cabinets are trapezoidal to facilitate doing that, even, and clusters in large venues are familiar to most everyone.

The problem with doing this in small rooms is that the sidewalls reflect the source back to the listener, and if it's all perfect, that reflection is identical to the direct sound, but delayed due to the differential path lengths, and interferes with it to varying degrees, depending upon the distances. Read Ken's paper on the AR "Magic" speaker.

To the extent that the reflected field is NOT perfect (and it never is,) it also alters the perceived spectral balance. Geddes terms omni dispersion in small, reflective spaces "disaster;" everything above the transition frequency is out of control. Your presumption that it's all nicely integrated in a diffuse reverberant field is simply erroneous, which is why I call it "imaginary." I believe you can rest assured that Geddes has not only heard this, but also measured it to death.

Allison & Berkovitz clearly illustrate that the "total energy in a reverberant field" thesis does not apply above the transition frequency, and progressively less so with rising frequency, which, by my reading, begins somewhere down in the range of 200 - 500 Hz in the rooms studied. The flat response measured in the reverberant chamber simply does not occur in-room, and upon closer examination, it's apparent that it's the integrated anechoic response which best characterizes what is actually heard by listeners as shown in the averaged in-room power response of 22 locations in 8 rooms....

Link to comment
Share on other sites

Wide and ultra-wide dispersion is easily accomplished using multiple constant-directivity horns or waveguides, much as LST and Allison did with multiple dome drivers, and it's commonly done in SR installations to achieve wide coverage patterns; the cabinets are trapezoidal to facilitate doing that, even, and clusters in large venues are familiar to most everyone.

The problem with doing this in small rooms is that the sidewalls reflect the source back to the listener, and if it's all perfect, that reflection is identical to the direct sound, but delayed due to the differential path lengths, and interferes with it to varying degrees, depending upon the distances. Read Ken's paper on the AR "Magic" speaker.

To the extent that the reflected field is NOT perfect (and it never is,) it also alters the perceived spectral balance. Geddes terms omni dispersion in small, reflective spaces "disaster;" everything above the transition frequency is out of control. Your presumption that it's all nicely integrated in a diffuse reverberant field is simply erroneous, which is why I call it "imaginary." I believe you can rest assured that Geddes has not only heard this, but also measured it to death.

Allison & Berkovitz clearly illustrate that the "total energy in a reverberant field" thesis does not apply above the transition frequency, and progressively less so with rising frequency, which, by my reading, begins somewhere down in the range of 200 - 500 Hz in the rooms studied. The flat response measured in the reverberant chamber simply does not occur in-room, and upon closer examination, it's apparent that it's the integrated anechoic response which best characterizes what is actually heard by listeners as shown in the averaged in-room power response of 22 locations in 8 rooms....

"To the extent that the reflected field is NOT perfect (and it never is,) it also alters the perceived spectral balance. Geddes terms omni dispersion in small, reflective spaces "disaster;" everything above the transition frequency is out of control"

Has it ever occured to you in your mind of limited paradigms that this critical problem can be solved? Why do you think I bought all those tweeters and equalizers, just to deplete my bank account? It's interesting. As you keep adding them and tweaking them, you hear them getting closer and closer. And when you get it right, you just know it because it sounds right to ears familiar with the real thing at its highest level. All that's left to do then is to equalize to compensate for variations in each source's spectral balance which only takes a few days or a couple of weeks of iterative listening and careful tweaking for each one. Of course this only matters if you are interested in accurately reproducing the exquisite sounds of beautiful acoustic instruments, not if you are trying to recreate the mind numbing ear shattering experience of deafening dreck or the falsetto adolescent voice of a boy in puberty and the nasal twang of an Eeeeelektik geetar. (I wonder what Shacky will do if Stereophile Magazine blows up his AR3as. Maybe they'll let him keep the Eico 80$ tube amp kit they loaned him as a consolation.)

By the way, Shamus O'Toole agreed with Leo the Lion Beranek that lateral early reflections with low IACCs were particularly liked by listeners, in fact it was the number one correlation between preference for 59 concert halls and 20 measurements of each one Beranek wrote up in his paper which can be downloaded from his site. In Baranek's case, preference is a legitimate issue, in Uncle Floyd's world, it shouldn't be unless profits are all that matters. There's a difference between painting a landscape from your imagination and taking a photograph of a real one.

Link to comment
Share on other sites

I have provided you each and both with access to economical means for exploring these concepts, but you obviously prefer to blather on interminably about yourselves, instead.

Have you nothing of more substance to offer up for discussion here? :blink:

Link to comment
Share on other sites

I have provided you each and both with access to economical means for exploring these concepts, but you obviously prefer to blather on interminably about yourselves, instead.

Have you nothing of more substance to offer up for discussion here? :blink:

I have deliberately made myself personna non grata at AudioKarma and I am not going back. It happened several years ago. AudoKarma was among the most boring sites for audio on the internet. Someone got all excited over a find of a bunch of old tubed table radios while someone else nearly had a sexual climax over finding some old 19" portable tube television sets in a warehouse. :unsure: One day this guy starts ranting about how he bought a subwoofer on e-bay for ten dollars and when he got it, he plugged it in and tripped a circuit breaker in his house because the power cord was damaged. He ranted on for two pages about he'd gotten cheated. ;) I told him he was an idiot. :blink: Then the moderator got on to defend him and attack me and I told him he was an idiot too. B) Then the owner came on, a guy called old goat or something and got on his high horse and I told him he was an idiot three. :D While the bunch of them were scrambling to figure out how to block me and delete my posts, I had no end of fun making one final splash with one comment about their parentage after another. :lol: That's the most I ever enjoyed blogging on AudoKarma and I am not going back even though I have a different e-mail address and could probably set up a new account under another monker.

If you have something to say Zilch, say it here and now. Otherwise your Karma Indigestia is giving me a case of indigestion. I'm not going on any more of your fishing trips. Bait your own hooks. :P

Link to comment
Share on other sites

There are over 5400 posts in that thread. I am NOT starting it over here. Nobody says you have to post over there to read any of it.

Then again, if your primary desire is to remain ignorant about this work (you're obviously clueless regarding the applicability of constant directivity in small spaces,) that is your prerogative.... :blink:

Link to comment
Share on other sites

There are over 5000 posts in that thread. I am NOT starting it over here. Nobody says you have to post over there to read any of it.

Then again, if your primary desire is to remain ignorant about this work (you're obviously clueless regarding the applicability of constant directivity in small spaces,) that is your prerogative.... :)

"I have provided you each and both with access to economical means for exploring these concepts"

I am not interested in anyone's third rate effort at anything to "dip my toe in" because when it comes up short, they always have an excuse that this was the economy version, you should hear what the real thing can do. I heard all this blather about these high end $125,000 a pair Audio Note speakers and when I finally got to hear them in the VTV show, when I walked in, the owner of this company Peter Qtvropwqaoeqooojgtrop or however he spells it was blasting this god awful rock whatever it was. He told me he went to live rock concerts. Funny how people in what is proported to be a business of making high accuracy loudspeakers so often burn their ears out on deafening rock. I asked him to play a disc I had brought with me. Both of my main sound systems killed his half million dollar pile of junk hands down. Not even a contest taken on its own terms of just making Marian McPartland's Baldwin SD-10 sound like a grand piano. Then he went on to play his reissues of Bing Crosby's 1940s reissues. You have anywhere in the greater New York City Metro Area preferably in NJ where I can here Gedde's TOTL speakers and judge for myself or do I just take your word for it until I go out and spend $168 on a kit or whatever that I'll hear once or twice and then put in my basement with piles of other equipment I'll never use again?

Why don't you send a pair to Stereophile Magazine. Maybe they could do a review as a sort of filler piece. Wonder how they'd stack up against AR3a in their eyes...especially if you took an ad for them out in their magazine. Hey don't I get a pro bono pair just for putting up with your dribble all these weeks?

Link to comment
Share on other sites

I'll give you the enlightenment part for free, but the rest, I'm afraid you'll have to pay for.... :)

Even if I was interested in buying speakers, I don't buy anything unheard. Everybody has a world's best speaker to sell and Stereophile Magazine proclaims a new one with each and every new issue. Subscribers would feel cheated if they didn't. Why should I believe you. I only trust my ears and my brain and so far you have made no sense AFAICT at all. I told you why I won't show you mine, what's your excuse for not showing anyone yours? If they are so great, why doesn't the world beat a path to your door like the proverb says?

Link to comment
Share on other sites

You still don't get it.

I don't sell speakers....

Then what exactly is it that you do do? I'm not so sure I really want to know. I might be better off if you don't tell me. :)

Seems to me you measured a few speakers in your own way and then posted 20,000 times about it. Is there more to it than that? ;) BTW, who did you ship those Karma Indigestias for, Santa Claus? Are you one of his elves? :lol:

Link to comment
Share on other sites

Then what exactly is it that you do do?

If you had even a modicum of common sense, you'd have figured it out by now.

BTW, who did you ship those Karma Indigestias for, Santa Claus? Are you one of his elves? :)

Your relentless dedication to ignorance is quite obviously unbounded. Hell, Shacky knows the answer to that one.... :lol:

Link to comment
Share on other sites

Over fifty four hundred posts? Yep, it's Disney World.

Yeah, and it just rolled over 230,000 views.

There's obviously a bunch of people who know more about this than you, Howard.

Really, you should get out more; you might learn something.

I am certainly pleased to have first brought Geddes to the attention of such an esteemed expert in the field of loudspeaker design as yourself.... :)

Link to comment
Share on other sites

FYI,

The May 2009 issue of Voice Coil contains an interview with Roy Allison covering 5 pages. He discusses his Allison Model 1's uniqure driver designs and reiterates his philosophy that flatness of power response over the widest possible freq. range and uniform dispersion is most essential in good loudspeaker design.

See the issue for more info.

Link to comment
Share on other sites

Ed Villchur put the whole concept on the line with his live-vs-recorded concerts back in the 1960.

Tell us again, please, Howard, how wide dispersion was the keystone of realistic reproduction in the LVR demos, when AR3 barely even makes it to 30°:

post-102716-1242089525.jpg

post-102716-1242089559.jpg

Link to comment
Share on other sites

Thats a very interesting interview of Roy Allison. I'm glad to hear that he is still around and involved.

Did anybody else notice that the stated the importance of achieving a flat power response, and then got a little more specific: that the ear judges balance based on the first 30 msec of sound arrival. In a typical room size that includes the direct sound and typically the first and second bounces. This isn't exactly the reverberent field, nor does it justify a requirement for flat power response.

Interesting.

David

Link to comment
Share on other sites

Thats a very interesting interview of Roy Allison. I'm glad to hear that he is still around and involved.

Did anybody else notice that the stated the importance of achieving a flat power response, and then got a little more specific: that the ear judges balance based on the first 30 msec of sound arrival. In a typical room size that includes the direct sound and typically the first and second bounces. This isn't exactly the reverberent field, nor does it justify a requirement for flat power response.

Interesting.

David

This comes as no surprise to me. In my theory of psychoacoustic modeling, I believe the brain takes the direct field and early reflections from the same direction as the source and builds a model of the source based on remembered comparisons it can draw on. This model is distinct from later reflections which the brain uses to model the space you are in.

Here's an interesting experiment you can try. Play a speaker with a crossover to the tweeter in the 5 to 10 khz range as you would normally with wide range material, especially material rich in high frequencies. Adjust the sound to be on the bright side. Now switch off everything except the tweeter and listen to it for awhile. Then while it is playing and you are at some distance from it, switch the rest of the system back on (bi-amping is one way to facilitate this.) If your own reaction to it is like mine, it will take a second or two for your brain to re-integrate the tweeter and the rest of the sound, they will be heard as two distinct sources for the first briefest of moments. Then it will sound as it usually does (it may even take longer to fully re-integrate.) Now try it the other way with the tweeter off. When you switch the tweeter back on, you will hear it re-integrated instantly. At least that has been my experience. If you try it, let me know what your own reaction is.

Link to comment
Share on other sites

Did anybody else notice that the stated the importance of achieving a flat power response, and then got a little more specific: that the ear judges balance based on the first 30 msec of sound arrival. In a typical room size that includes the direct sound and typically the first and second bounces. This isn't exactly the reverberent field, nor does it justify a requirement for flat power response.

I did, of course, but haven't thought through the implications. If 80 ms is the criterion, we can't get there in small rooms @ 1.1 ft/ms. 30 ms = 34 ft, a minor contributor, I would say, "early," clearly, and not reverberant field by Toole's definition.

Ken used 20 ms in "Magic;" I don't know what any of these sound like in isolation.

He slipped wide dispersion in there, but didn't define the solid angle. I read the rest as constant directivity.

He ranks Toole right up there with Thiele, Small, Villchur, and Kloss.

[Neither Soundminded nor Zilch made the cut, alas.... :rolleyes: ]

Link to comment
Share on other sites

I did, of course, but haven't thought through the implications. If 80 ms is the criterion, we can't get there in small rooms @ 1.1 ft/ms. 30 ms = 34 ft, a minor contributor, I would say, "early," clearly, and not reverberant field by Toole's definition.

Ken used 20 ms in "Magic;" I don't know what any of these sound like in isolation.

He slipped wide dispersion in there, but didn't define the solid angle. I read the rest as constant directivity.

He ranks Toole right up there with Thiele, Small, Villchur, and Kloss.

[Neither Soundminded nor Zilch made the cut, alas.... ;) ]

Hey Zilch,

I'm not sure what you mean with the reference to 80ms. That time is often used to define a border between early sound and late sound for music in concert halls (50ms is typically used for speech). Concert hall reflections prior to 80 ms are desired to give increased clarity, reflections later degrade clarity. Ditto reflections prior to 50ms for speech. Lecture halls, in spite of lowish Rt targets overall, are designed with hard ceilings to increase reflections prior to 50 ms. This increases the received level in the audience without decreasing inteligibility.

Of course with music reproduction, we get fussier about our reflections and don't want them to modify the reproduced timbre at all.

I just looked at Ken's paper on the MGC-1 again and he uses 20ms as an appropriate delay because lesser delays were felt to give reflections that distorted tonal balance. Why greater delays wouldn't be even better, I'm not sure. (Ken are you out there?)

Maybe its just semantics, but I think reverberation refers to everything following the direct sound. You might split that into early reflections and later reverberation, but it seems an artificial distinction with no real definable dividing line.

Regarding neither you or Soundminded making the cut; we love you both just the same. :D

David

Link to comment
Share on other sites

Hey Zilch,

I'm not sure what you mean with the reference to 80ms. That time is often used to define a border between early sound and late sound for music in concert halls (50ms is typically used for speech). Concert hall reflections prior to 80 ms are desired to give increased clarity, reflections later degrade clarity. Ditto reflections prior to 50ms for speech. Lecture halls, in spite of lowish Rt targets overall, are designed with hard ceilings to increase reflections prior to 50 ms. This increases the received level in the audience without decreasing inteligibility.

Of course with music reproduction, we get fussier about our reflections and don't want them to modify the reproduced timbre at all.

I just looked at Ken's paper on the MGC-1 again and he uses 20ms as an appropriate delay because lesser delays were felt to give reflections that distorted tonal balance. Why greater delays wouldn't be even better, I'm not sure. (Ken are you out there?)

Maybe its just semantics, but I think reverberation refers to everything following the direct sound. You might split that into early reflections and later reverberation, but it seems an artificial distinction with no real definable dividing line.

Regarding neither you or Soundminded making the cut; we love you both just the same. ;)

David

"Of course with music reproduction, we get fussier about our reflections and don't want them to modify the reproduced timbre at all."

That is not correct. Whether you are talking about concert halls used in the production of music or home listening rooms used in the electronic reproduction of music, all rooms alter the timbre of musical instruments usually significantly.

All concert halls alter the timbre of sound. RT at 8khz is typically 40% to 60% of RT at 1khz and for lower frequencies often much higher. Reflections at bass frequencies are very important. Beranek says as a reflector of sound, wood stinks. This according to him is a popular misconception among musicians and concertgoers alike. The wooden wainscot on the stage of Boston Symphony Hall was a concession to this misconception, it really didn't belong there but it wasn't enough to ruin the acoustics. For wood to be acceptable, it must be at least 2 1/2 inches thick. Beranek cites one concert hall built in Scandanavia by a lumber company that sounds good because it meets this criteria. Any thinner and it absorbs bass. Typical steady state transfer functions of concert halls aren't even close to flat. Their falling high end is another way of expressing the more rapid absorption and therefore lower RTs of high frequencies. This is what Villchur and Alison tried to emulate. But their fatal mistake is that their steady state filtering does not give audibly comparable results to the filtering of each note individually over time. The high end rolloff of the speaker results in reduced clarity, reduced sharpness of initial transient attack of each note that defines each instrument distinctively. This is preserved at the live performance, the relative difference in RT as a function of FR making the sound mellower. Mellower and clearer at the same time. A seeming contradiction in concept in audiophile thinking. A reality of experience for concertgoers.

Link to comment
Share on other sites

Maybe its just semantics, but I think reverberation refers to everything following the direct sound. You might split that into early reflections and later reverberation, but it seems an artificial distinction with no real definable dividing line.

80 ms seems to be the IACC dividing line for ASW enhancement; this must somehow all fit together.

I don't think it's semantics, as we discussed earlier. I can see how late reflections might sum into generating a diffuse, integrating, reverberant field such as what is dominant in large spaces like concert halls beyond the first few rows, which Howard assumes also prevails in small rooms. I argue that Allison & Berkovitz clearly demonstrated that it does NOT exist in typical home listening spaces, and thus, it's the combination of the direct source and early reflections that best describes the spectral balance we hear above a relatively low transition frequency, and those are far more accurately characterized by the on- and off-axis anechoic measurements than by the "total energy" response as measured in a reverberant chamber. We can then argue whether wide dispersion is advantageous or even necessary to achieving the desired objective, and that's not so clear as Howard presumes, either.

See Toole p. 342, bottom. I think he makes a credible case for a combination of derivatives of anechoic response measurements as definitive, as shown in Fig. 18.5. We don't need in-room measurements to predict how a speaker will sound in typical listening spaces, once their own characteristics are known and understood in a normative sense, and Howard's total energy/reverberant field theory provides a decidedly erroneous answer.

Allison tells us the East Coast design objective: "to simulate live acoustic concert sound as closely as possible." To the extent that was achieved, it was more as an artifact than a direct result of the principles employed to accomplish it....

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...