Roy Allison interview in 1992 issue of The Audio Critic, David Ranada was the interviewer

[speaker dave]

2 pi #5 curve, right? Not #9. Just want to be clear......

Steve F.

Yes, #5 is still 2pi but has the picture frame "lip" added as a reflective edge. Before that #4 has a perfectly flush baffle for all drivers. #9 is a full cabinet with the lip or molding and is therefore the 4pi case as most owners would experience.

Now if you cut off the molding and flush mounted them into your wall.....#4, The ultimate in-wall speaker.!

David

[speaker dave]

Well, if we are showing off our polar curves, then here are the curves from the 4430/35 JBL monitors. Grandaddys to the Zilch econowaves and others. This would be the horn, from 1000 to 16,000. These are 1/3rd octave curves, so less averaging than Roys. The only two that stand out are the 12.5k and 16k.

The most uniform polars I have seen, before or since.

I plotted these out and my wife lettered the captions as part of generating the AES paper on the systems.

David

[speaker dave]

As for the "long-since repudiated" view Allison and others have regarding wide dispersion, well I am not sure at all that everybody in the business or doing research on speaker behavior feels that way, and maybe if classical music ever encounters a renaissance the approach will be reborn.

Howard Ferstler

I would be interested if you could offer any paper written in the last 20 years that emphasized power response over the direct/early soundfield. For the ones that suggest the opposite, search my posts, as I have listed them numerous times.

David

[Zilch]

Well, if we are showing off our polar curves, then here are the curves from the 4430/35 JBL monitors. Grandaddys to the Zilch econowaves and others. This would be the horn, from 1000 to 16,000. These are 1/3rd octave curves, so less averaging than Roys. The only two that stand out are the 12.5k and 16k.

The most uniform polars I have seen, before or since.

Superb!

Does Zilch object to this kind of performance? I certainly do not.

As posted at #45, SpitWad is coming pretty close.

I suspect, when complete, it will be a reasonable match to AR3a.

[it has already bettered it at 0° - 45° in comparison to AR's published polars.... ]

I was referring to the wide-dispersion issue and not to power response.

It's not terribly difficult to put together a directivity Q of 2 with a flat on-axis response and know the outcome....

[speaker dave]

Okay, if you like constant directivity and Zilch likes constant directivity, the only difference is how directional.

The polars I showed have a d.i. of 10dB (meaning the axial response is 10dB louder than the power averaged response). Now if the Allison has perfect semicylindrical radiation (the polars suggest this, but also remember that there will be a little greater directivity in the unseen vertical polars), lets guess that the average d.i. is 5dB. (In both cases we are talking about midrange and treble, the LF directivity will be negligible and determined by system mounting. The Allison will be against the wall for a d.i. of 3, the JBL in a studio would possibly be soffet mounted for the same).

So with a d.i. difference of 5 dB either of you could be happy with the other's speaker just by moving closer or farther. That is the difference due to different d.i.: the distance from the speaker you would need to move to experience a given balance of direct to reflected sound. Howard, with Zilch's speaker you could back up to double the distance and hear the direct to reflected balance that you like. Zilch, with Howard's speaker you could move to about half the distance and hear the direct to reflected balance that you like.

What were we arguing about?

David

[Zilch]

What were we arguing about?

The deleterious impact of early reflections generated by excessively wide dispersion upon both spatial and spectral quality in typical home listening spaces, which Howard has long since conceded....

[soundminded]

The deleterious impact of early reflections generated by excessively wide dispersion upon both spatial and spectral quality in typical home listening spaces, which Howard has long since conceded....

I know exactly what you mean about too much dispersion. It's a terrible problem with pianos and violins. Do you think if I could get them all to each play inside a big closed box with a hole in it they'd sound more like speakers? Should I put a constant directivity horn in front of the hole?

[Pete B]

When we were putting together the AR restoration document, we sent Ken K. a number of tweeters to test in an effort to develop an appropriate crossover modification for the AB Tech replacement tweeter. At that time, Ken compared a number of original, used 3a tweeter specimens to Tom T's NOS tweeters, and there was a measurable difference between them. Two of mine died during testing.

Roy

Hi Roy,

Just wondering how your tweeters failed during that

testing, was it a wire break or something else?

[Zilch]

I know exactly what you mean about too much dispersion. It's a terrible problem with pianos and violins. Do you think if I could get them all to each play inside a big closed box with a hole in it they'd sound more like speakers? Should I put a constant directivity horn in front of the hole?

Alas, STEREO is most commonly practiced using a minimum of TWO laterally displaced sources.

Do we believe for a minute that two pianos or violins similarly deployed would generate an accurate phantom image of a single instrument midway between them in a typical semi-reflective home listening room?

[it's déjà vu we're doing; we've been here a couple of times before.... ]

[RoyC]

Hi Roy,

Just wondering how your tweeters failed during that

testing, was it a wire break or something else?

Hey Pete,

I really don't know, nor do I know what sort of testing was being conducted when they failed. The results for the surviving original tweeters generally showed diminished output, and faster/steeper lower frequency roll off as compared to Tom's tweeters. I'm convinced that decomposed foam in the vc gap, and/or stiffening, deteriorating suspensions are the primary culprits.

Roy

[soundminded]

Alas, STEREO is most commonly practiced using a minimum of TWO laterally displaced sources.

Do we believe for a minute that two pianos or violins similarly deployed would generate an accurate phantom image of a single instrument midway between them in a typical semi-reflective home listening room?

[it's déjà vu we're doing; we've been here a couple of times before.... ]

Do we really believe for a minute that forty violins playing together in a symphony orchestra would sound like forty seperate sources or one large mass of instruments spread out over a wide angle? Hey Zilch, why not go to a live concert, find out, and report back. Then we could compare notes.

[Zilch]

I see no deleterious impact at all, and certainly never have said so.

You just said it again - use a center channel to "fix" it, precisely as recommended by Toole.

My problem with Zilch is that he dismisses my favored approach and almost dictates standards for loudspeaker behavior.

What I dismiss is the contention that wide dispersion is inherently "better," and the theoretical basis posited here for suggesting that to be so. Spraying a typical small home listening space with hemispherical constant directivity is not an optimum approach to replicating concert hall realism. It cannot achieve listener envelopment as the delays are too short in such spaces, but instead, adversely affects localization and spectral quality.

I have had friends over who admitted that the center feed made the soundstage sound more live-performance like, but they still PREFERRED to have the centered performer or instrument be larger than life (as it would appear in a concert hall in front of the orchestra), and thereby preferred the two-channel playback appraoch. So goes listener taste.

There it is, of course, as described in detail by Toole -- we like artificially enhanced spaciousness; that's the attraction of Bose, also derivative of acoustic power theory. The question is whether that equates to realism, independent of whether we like it or not or how much, and the answer is, "No."

[genek]

There it is, of course, also as described by Toole -- we like artificially enhanced spaciousness; it's the attraction of Bose. The question is whether that equates to realism, independent of whether we like it or not or how much, and the answer is, "No."

If we define "realism" as the ability to convince me that I'm in a concert hall or any other environment rather than the space I'm actually in, then I've never experienced an audio system - of any kind - that I would describe as having that quality. At best I can hope for audio to not convince me that some kind of dimensional warp has transplanted a symphony orchestra to the stage of a Black Sabbath concert, or to convince me momentarily that some sound source is happening real-time in my living room. But convincingly reproduce the ambiance of another listening space? It has never happened, not with any system I have ever experienced, in any listening space I've ever been in (and that includes some very expensive professionally-installed movie theaters or theme park rides, which to my ears merely created the unique experience of being in a movie theater or a theme park ride).

If we define "accuracy" as the ability to play every detail and nuance of a recording, whether it sounds good to me or not, I definitely have experienced audio systems that I would describe as having that quality, and I have chosen not to own one, because given a choice between accurately reproducing the reality of what's on most recordings or producing pleasant fantasy versions of them, I choose the fantasies. I already get all the unpleasant audio reality I can handle every time I step outside my front door.

[Zilch]

At which juncture we depart from discussing loudspeakers and are talking instead about ourselves and our preferences. It's an important distinction.

I get similar all the time with respect to West Coast speakers such as JBL L100. Enthusiasts claim they are accurate and realistic. They are neither, but they ARE fun.

[Well, for us knuckle-dragging headbangers lacking refined sensibilities, anyway.... ]

[genek]

At which point we depart from discussing loudspeakers and are talking instead about ourselves and our preferences. It's an important distinction.

Yes and no. Our preferences and the products they lead us to choose are essentially a closed system. You can never completely separate them. And accuracy of some sort is essential at some point in the design-build-buy cycle if you want to have any confidence that you're going to be able to repeat your subjective successes.

[Zilch]

Yes and no. Our preferences and the products they lead us to choose are essentially a closed system. You can never completely separate them. And accuracy of some sort is essential at some point in the design-build-buy cycle if you want to have any confidence that you're going to be able to repeat your subjective successes.

The fantasy is supposed to be in the program, not the speakers....

[genek]

The fantasy is supposed to be in the program, not the speakers....

Now that's what I call living in a fantasy world.

Buying audio equipment that can only reproduce quality program material is like going to the polls on election day expecting to vote for an honest politician.

[Zilch]

Now that's what I call living in a fantasy world.

Buying audio equipment that can only reproduce quality program material is like going to the polls on election day expecting to vote for an honest politician.

Don't forget the quality kit, room, and quality ears, as well....

[genek]

Don't forget the quality kit, room, and quality ears, as well....

The ears set the standard that the audio system and room are required to modify the recording to satisfy. It's the default SOP for those of us who are not judging sound quality by instrument charts.

[Shacky]

The ears set the standard that the audio system and room are required to modify the recording to satisfy. It's the default SOP for those of us who are not judging sound quality by instrument charts.

Hallaluya !!!

That's another factor in the "preference" determination arguement. Many speakers sound great on great recordings and like @$$ with lesser material. Main reason I got into vintage and AR's in the first place.

[kkantor]

Hi Roy,

Just wondering how your tweeters failed during that

testing, was it a wire break or something else?

They were destroyed in my care, which is something I will go to my grave feeling terrible about, frankly. They failed as my technician ran 2.83Vrms of pink noise into them WITHOUT any kind of high pass filter or compressor. I really, really should have warned/instructed him. I remember yelling out of my office, "Hey double check your test levels," before he started, but I never mentioned the test signal bandwidth.

The failure mode seems to have been an electrical open along a section of the coil lead-in which had been somewhat corroded by the suspension dot it ran through. Post mortem also revealed a fair amount of deterioration throughout the unit: hardened dots, crystallized former adhesive, powdered foam button, wire corrosion, etc.

-k

[kkantor]

I see no deleterious impact at all, and certainly never have said so.

Obviously, the only way to subjectively check things out is to locate various speakers in good rooms, get hold of some really good recordings that demand that speakers simulate as best they can a live-music space, and then adjust listening and speaker positions to see what transpires. The is not to say that measuring is not important, but the bottom line is the way the speakers deal with recordings and rooms. The trick is to have good recordings, and of course a good room, or better yet, more than one good room. I have a lot of superb recordings (I reviewed most of them in two books and dozens of magazine review articles) and have two fine listening rooms, and have done such comparison sessions (level matched, of course, at least as close as possible) as part of my speaker reviewing work.

I prefer the Allison approach to speaker dispersion (very wide, broad-bandwidth output, making use of the front and side walls as part of a speaker/room continuum), but I can see the point of Zilch's approach, at least if done correctly. It certainly mimics what Dunlavy did with his approach. However, even Dunlavy once told me that he could see the attraction of monumental room reflections. Heck, he even told me that the Bose 901 approach can be impressive and realistic sounding under some conditions.

My problem with Zilch is that he dismisses my favored approach and almost dictates standards for loudspeaker behavior.

Howard Ferstler

Howard,

Just as one example, the coloration resulting from reflections in the 2ms range is very well known, and has been well documented by Kates at AR, among others. I remember talking (briefly) with RA about this during the early MGC-1 days, and I am certain he would not dispute this. Likewise, earlier reflections have been shown to influence localization.

Of course, I am not suggesting how you should prioritize such signal distortions, or what impact these issues should have on loudspeaker design choices. But, the existance of early reflections in the playback chain is, most definitely a form of distortion, in a way that is more clearcut than the existance of later reflections are.

-k

[Zilch]

I don't review any more, nor do I even write commentary columns. However, I did review two different Dunlavy models (SC-II and Cantata) a number of years back, and Dunlavy, who held several patents in radar horn technology before becoming a speaker guy, managed to build speakers that were paradigms of controlled directivity. I mean, that was his mantra.

Do we know the directivity of Dunlavy designs? What was his target beamwidth?

Ooops, got diverted:

http://blog.stereophile.com/stephenmejias/...nd_only_prayer/

"You are my ev-er-y dream...."

SC-IV/A controlled directivity? Not:

http://www.stereophile.com/floorloudspeakers/166/index6.html

SC-I? Not:

http://www.stereophile.com/standloudspeake...avy/index4.html

SpitWad, presently, before crossover optimization:

[soundminded]

Howard,

Just as one example, the coloration resulting from reflections in the 2ms range is very well known, and has been well documented by Kates at AR, among others. I remember talking (briefly) with RA about this during the early MGC-1 days, and I am certain he would not dispute this. Likewise, earlier reflections have been shown to influence localization.

Of course, I am not suggesting how you should prioritize such signal distortions, or what impact these issues should have on loudspeaker design choices. But, the existance of early reflections in the playback chain is, most definitely a form of distortion, in a way that is more clearcut than the existance of later reflections are.

-k

"the existance of early reflections in the playback chain is, most definitely a form of distortion, in a way that is more clearcut than the existance of later reflections are"

What about the existance of early reflections in live music, is that also a form of distortion? Let's say two violinists are sitting next to each other. The sound of each instrument bounces off the other a few milliseconds after it is launched. Now what about 100 musicians, a symphony orchestra. What a lot of noise and distortion they can make...especially when half of them are playing the wrong notes ...or Charles Ives

[RoyC]

They were destroyed in my care, which is something I will go to my grave feeling terrible about, frankly.

-k

Ken,

There is absolutely NO need to feel that way! That experience was motivation for me to look more closely at the condition of the old tweeters, and to seriously explore alternatives. Further, if you recall, you provided me with a very nice pair of new Vifa tweeters, which were put to excellent use. Your efforts were invaluable, and very appreciated. Thanks again!

Roy