Box 978: Letters to the Editor (Issue No. 14 Summer through Winter 1989-90)

We are fully aware that some people read this column as a kind of "Theater of Cruelty"-even if they have never heard of Antonin Artaud-but that is not its purpose. We greatly prefer correspondence from which we and our readers can learn something about audio; however, once we have brought up the name of a person or a company in our pages, we feel obligated to publish their letters, friendly or not, to which we then respond editorially. In the process, a few jackasses may acquire a higher profile than the serious, sincere majority of our correspondents, but c'est la vie. The letters we print here may or may not be excerpted at the discretion of the Editor. Ellipsis (...) indicates omission. Address all editorial correspondence to The Editor, The Audio Critic, P.O. Box 978, Quakertown, PA 18951.

The Audio Critic:

...I just want you to know how de lighted I am that The Audio Critic is back and that it is in top form, too. Bravo! Yours is the only publication that wraps me in "can't put it down reading," like a good novel, as soon as each issue arrives.

The Audio Critic contains consistently the most brilliant writing about audio that I have seen in over 30 years as an audiophile. This has to do, I believe, both with what you say and how you say it. Special favorites of mine are your reviews of recordings. These are clearly from the heart and show you to be as much music lover as you are lover of the science of reproducing music. On the other hand, I am not so much in awe of your opinions that I must run out and buy your recommended components. I think I retain a healthy skepticism.

Whether I agree with you or not, I always enjoy the reading experience itself.

Every word, every sentence seems polished to a high luster and fits logically, rather like the notes in a score by a great composer. The overall effect is the perfect ed effort of a single writer who cares deeply about the excellence of his work. Your publication is obviously not the result of a committee process, which is what the other audio magazines have become.

This brings me to...your statements regarding regularity of publication of The Audio Critic. You seem to be putting pres sure on yourself to achieve your own notion of a regular publishing schedule. But, please, please do not feel any guilt or pres sure to do so.

I realize that pressure can stimulate creativity. It can also cause quality to deteriorate. I have witnessed the transformation of other "underground" audio journals into fat, glossy, regularly published periodicals. In the process, 1 feel they have become watered down, verbose, and too driven by the need to make a "statement" with each issue.

I, for one, sorely miss the old irregular Stereophile in the days when J. G. Holt was "editor, chief tester, and drudge." Then, unlike now, it was written with wit and creativity, focus and consistency of point of view that today can only be found in The Audio Critic. JGH's views and yours differ, but the honest and readable quality of both the old Stereophile and The Audio Critic of today are of the highest caliber.

If pressure to publish regularly causes you to follow in Stereophile's footsteps to what it has become today, then I say to you, don't do it. I am comfortable seeing you publish only when you have some thing to say that lives up to your standards of excellence. Besides, I like surprises...

Best regards, Carl J. Weber Philadelphia, PA Thank you, kind sir. A letter like this relieves us of the rigors of false modesty; you praise us as expertly as we would praise ourselves if self-praise were permissible. Kind as you are, however, we must disagree with you on a number of points:

(1) We could name certain scores by Bach, Mozart and Beethoven that are more perfectly wrought than a typical issue of The Audio Critic. (2) On the other hand, The Audio Critic is, and always was, better written than the old, not only the cur rent, Stereophile. (3) The most significant difference between the old and the current Stereophile is the difference between the minds of J. Gordon Holt and Larry Archibald. (4) The pressure to publish regularly

is almost purely an economic one; income from subscriptions and ads is derived on a per-issue basis-the more issues, the more income. (Our plan is to progress gradually from quarterly regularity in 1990-yes!- to a bimonthly and eventually monthly schedule in the years to follow, and we have no fear of a lapse in quality as long as we can expand our staff without bringing in mediocrities.) Your comments also bring back memories of a particularly poignant letter we once received from a man who had quite obviously never read anything in his life except electronics and audio magazines, owner's manuals, and parts catalogs. He was astonished and apparently deeply moved by the quality of English writing he had discovered in our pages; we had somehow become his surrogate Dickens and Joyce. We found that responsibility a little more than we could handle.

Anyway, you definitely have the right attitude, Carl; what we need now is 50,000 more subscribers like you.

-Ed.

The Audio Critic:

It has come to my attention that considerable confusion and/or questions have arisen regarding bit stream, or 1-bit, D/A converters. I wish to share with you a simple explanation that may provide some answers.

The PCM digital audio signal, as it comes off the disc, is nothing like an analog audio signal. Both channels are mixed, the bits are out of sequence (because of the error-correction encoding), and miscellaneous bits are added for track number, time code and copy protection, to name a few.

Conventional D/A converters recon struct analog signals by creating the "stair case approximations" we have all seen in many ads and articles. The "run" or distance between each "step" is based on the sampling frequency, while the "rise" or voltage amplitude of each step is based on currents running through resistances inside the D/A converter IC. The step voltages are subject to the accuracy of the currents and resistances, and in every sense are analog values.

There are two bit-stream systems now commercially available, and while they are different, they have enough in common for the purpose of this explanation. Bit stream D/A converters digitally, and at very high speed, convert the PCM signal into two digital signals that have much more similarity to analog signals. That is, a digital to-digital conversion separates the two channels, decodes the error correction, re moves the miscellaneous bits, and puts the left and right digital signals into a format that easily lends itself to further simple conversion. The digital-to-digital conversion is accurate, and removes the mechanism for errors and nonlinearity present in conventional D/A converters. The only further conversion required is low-pass filtering, which is inherently a simple linear process.

I firmly believe that bit stream converters will soon be in wide use. The cost of bit stream D/A conversion is reason able; it requires no production alignments for linearity, and has excellent sound quality and measurement data.

With best regards, Marty Zanfino Vice President Engineering and Technical Services Harman Kardon, Inc.

Woodbury, NY An in-depth overview of the state of CD playback technology, introducing an important new contributor to The Audio Critic, will be published in our next issue (No. 15). The article will include, among many other things, a detailed critique of the new 1-bit DAC architectures. So far we have heard both endorsements and caveats regarding this new approach-and from equally authoritative sources. Meanwhile we must state that your explanation of it for the layman is by far the simplest and most lucid we have encountered.

-Ed.

The Audio Critic:

I own an expensive audio system. I have never told anyone the price. Quite frankly I am embarrassed by its price. For years I laughed at Harry Pearson et al. for recommending expensive equipment. I even purchased the budget system you suggested (Hafler amp, PS Audio preamp).

The fact is that I never liked the sound. It gave me a splitting headache. I replaced my electronics with a Moscode 300 and a C-J PV5. Flawed? Yes! But at least I liked the sound. (Quite frankly it was gorgeous.) This brings me to the ABX comparator. I don't believe that all amps sound the same. I believe that it (ABX) is a signal processor. But if I'm wrong, I'm willing to put my money where my mouth is. I will buy for you the Radio Shack amp and equalizer of your choice and accept your Boulder 500 as an even swap. If that is not acceptable to you, I will donate to The Audio Critic the Radio Shack amp and equalizer of your choice. The only condition is that you conduct all your personal listening and professional testing with this amp and equalizer.

While I am sure you will find a way to slip out of this offer, I think your credibility on this issue would be firmly established if you accepted this offer.

In fact, while I am not rich, I extend this offer to Mr. Lipshitz, Mr. Carver and Mr. Clark.

Sincerely, Reginald G. Addison Attorney at Law Washington, DC CC: The Absolute Sound Stereophile To begin with, counselor, "splitting headache" as against "frankly gorgeous" does not quite make the grade as expert witness testimony, does it now? But that was not your reason for writing, so we shall let it pass.

As for the ABX comparator, if it is a

"signal processor," then so is the source selector switch on the costliest preamps;

the basic operating principle is exactly the same. Besides, ABX switching without the box, by means of hand-plugged cables, has been repeatedly proven to yield the same results. Talk about credibility! Today there exists no halfway respectable opponent of ABX testing who attacks the box itself. For better, though ultimately invalid, anti-ABX arguments see our article on the subject in Issue No. 12.

Your prophecy regarding our refusal of your offer is of course self-fulfilling. As you probably know, or must at least suspect, no Radio Shack stereo amplifier can be bridged for mono operation to swing 70 volts into a variety of loads at virtually zero distortion, like the Boulder. Regard less of listening quality, we need that kind of measurable performance for reference in our laboratory. Equalizer? You have lost us there; we hardly ever use one.

Your most fundamental misconception, however, is that we believe-or ever said-that a Radio Shack amplifier will un der all conditions sound indistinguishable from a Boulder (or Krell or Mark Levin son). What we believe and are willing to reiterate is this: Any two amplifiers of conventional architecture-very high input impedance, near-zero output impedance, typically low distortion, normal stability- and operated well within their voltage and current capabilities (very important!) will sound startlingly similar when matched in output level within 0.1 dB. You may end up identifying the expensive amplifier vs. the Radio Shack blind, on the basis of some subtle clue (say, the noise floor or maybe a rise or dip of 0.2 dB somewhere), but your audiophile heart will sink in the process- this is not the night-and-day difference the tweako reviewers led you to expect! On the other hand, the Radio Shack amplifier is much more likely to be driven beyond its voltage/current capability under various field conditions than the Boulder (or Krell or Mark Levinson), and then all bets will be off. Have we made ourselves clear? It should be added that tube amplifiers with their generally higher output impedances drive the speakers a little differently and may therefore sound different to the extent of the resulting spectral shifts and Q changes.

Anent your pointedly listed CC's, we would like you to ponder this: If you wrote a law journal a letter disagreeing with their viewpoint on obscenity legislation, would you then send CC's to Screw magazine and Hustler?

-Ed.

The Audio Critic:

Thank you for your "Seminar 1989:

Exploring the Current Best Thinking on Audio." It's already the most fascinating piece I have read in any of the audio magazines, and I haven't seen Part 2 yet. (And now it will be Part III keeping you in suspense because of our last-minute editorial decision necessitated by the monstrous length of the transcript.-Ed.) | know most of the participants from the AES, and Dave Clark is a good friend. You certainly had the right group assembled for this, and the reason it is so interesting is that it shows what a diversity of opinion there is on how we should "do" audio. In the presence of such an august gathering I hesitate (ever so briefly) to say that I believe I have some thing to contribute to the discussion.

...Your group...got into the real meat of the matter, which is the whole area of how to do sound reproduction in an ideal way-how many channels do we really need, is Ambisonics the answer, binaural vs. stereophonic, room acoustics, etc.

What I have to contribute to this is a new theory for stereophonic sound, a sort of way of thinking about the process, a model against which we can test the various ideas for validity. I call it the image model theory. It appears at first to be quite a bit different from the usual thinking in audio, and it gets pretty fundamental, so I think the best way to approach this would be to lay it out first, and then go back to your discussion and show how it ties in.

We must first clearly distinguish be tween stereophonic and binaural. These are two totally separate systems, every bit as different from each other as, say, sculpture and painting. Binaural is the simpler of the two systems, being nothing more than an attempt to replicate the ear signals that you would have heard if you had been in the same location as the dummy head during recording. You get exactly the same direct-to-reflected ratios, acoustics, every thing as if you were transported to the hall or studio. The stereophonic system is a completely different approach. Rather than attempt to replicate the ear signals that a listener at the original event would have experienced, it starts from the other end and reproduces the orchestra itself, and the early reflected sound near the orchestra, in all its glory, right in front of you in your listening room. It is a field-type system in which we are physically reconstructing all of the salient characteristics of the original sound fields in the playback space. We tend to prefer the stereophonic system because the sound is real-not an illusion.

We can move around, turn our heads, feel the chest-thumping bass, even move closer to or farther from the orchestra. In other words, even though stereo is not an exact facsimile reproduction of the original acoustics and everything, it still wins out in realism because the sound is really right there in front of you, and you can use your natural hearing and move around. Binaural is like a 3-D photograph of Jack the Ripper on the streets of London. Stereo is like be ing eyeball-to-eyeball with a sculpture of him in the Wax Museum. Two completely separate systems, OK? The reason I dwell on this is that the confusion between the two systems arises when audio theorists believe, or start with the assumption that, the object of "accurate" stereophonic re production is to reproduce the two channels as signals at the ears. This approach manifests itself in attempts to kill all room reflections or to put all of the drivers of a loudspeaker on one surface and aim them at your face. This is just the opposite of what needs to be done to get good sound, which brings me to my image model theory.

What we are going to do is to study the pertinent audible characteristics of a live sound field in order to see how to make the reproduction more like it. There are three such characteristics which can be distinguished: spectral, temporal, and spatial. We can study these characteristics by dividing up the arriving sound into its three main temporal stages: the direct field, the early reflected, and the reverberant sound. The direct sound arrives straight from the instruments and has a flat spectral balance. Spatially, the direct sound establishes the lateral localization of the sources. Next to arrive at the listener is the early reflected sound, from the front and side walls of the concert hall. The easiest way to understand its spatial nature is to make a drawing of the image model of the instruments and their first and second reflections. The reflections appear as additional sources behind the front and side walls (see drawing). The early reflected sound has two jobs: it provides us with a complete "view" of the radiation pattern of the musical instrument (giving us more in formation about what the instrument sounds like in an enclosed space than out doors, for example), and it gives us the spatial impression that is so important to the enjoyment of music. This early reflected sound is stereophonic in nature, and the pertinent audible characteristic is a spatial broadening of the orchestra, lending a richness and depth to the sound. The full reverberant field, caused by all the remaining reflections in the concert hall, is virtually monophonic or nondirectional. It is fed by a buildup of the more steady-state tones, which tends to roll off the high frequencies above about 1 kHz and contributes to the "musicality" or sweetening of the sound.

So the spatial "shape" of live sound is a dash of direct sound, a very important and very stereophonic "splash" of early reflected sound from front and side walls, and a virtually nondirectional decay of reverberant sound. Now, if we look at the reproduction, we can draw an image model of loudspeakers in a playback room just as easily as musical instruments in a concert hall. Image model theory says that our goal should be to make the spatial shape of the playback, or the playback image model, as close to that of the live sound as possible.

The theory sees the reproduction as a mod el of the real thing, with speakers pulled out well away from the walls, first-arrival sound coming directly from them, followed by an early reflected field of similar intensity and shape to live sound, and with the room itself (aided by surround speakers) building a reverberant field for play- back just as the concert hall did live. To make a long story short, what I have found is that if this is done right, it will actually decode the spatial information contained in the recording much better than, say, a pair of highly directional speakers in a dead room, which tends to take all of the record ed spatial information and compress it so that it can come from only a limited set of incident angles, as defined by the separation of the loudspeakers.

In any case, if you at least understand what the image model of live sound looks like, you can more easily see the advantages/disadvantages of various theories of reproduction or proposed systems...

On the question of whether two channels are sufficient or whether we need umpteen, I would point out that two are sufficient for the enjoyment of music. The stereophonic system as such has no limitations on the number of channels, and you could use one for each instrument if you thought it worthwhile, but such precise localization of the sources has nothing much to do with the appreciation of the musical message. Yet two channels contain all the information needed to generate lateral localization of the direct sound and all of the spatial patterns that are so important to the feeling of spaciousness and being immersed in a live sound field. Since the full reverberant field is monophonic, the only need in the reproduction is to fill out the back half of the sound field with something that is delayed with respect to the frontal sound. The difference (L - R) signal is just terrific for doing this without causing an echo of the center front soloist, and discrete surround systems are superfluous. I believe that a standard, passive Dolby surround decoder is all that is needed. You can use as many speakers as you need to even out the surround field, but stereo phonic capability is not required in the surround channel-the stereo effects are completely taken care of by the front speakers and the early reflected sound.

On pp. 24-25, Stanley says that to begin to standardize some sort of ideal sys tem, we should begin with how we hear- talk about the ear signals. Again, this would be a binaural system, and Stanley knows that, but then he goes on to imply, as have Duane Cooper, David Griesinger and others, that we could record binaurally and then process it over to stereo or any thing we want. I definitely do not agree with this-the two systems are totally separate and incompatible. Binaural and stereophonic recordings contain completely different kinds of information, from different perspectives, and for different goals. A third possible system that he may be thinking of is one where you record binaurally and then play back on speakers, using a crosstalk cancellation circuit (Sonic Holography or Polk SDA) in a relatively dead room. That's a terrific idea, but it's still bin aural (loudspeaker binaural), and you have to sit still in just the right spot, etc. But this is not a stereophonic system, nor will it trans late to one. And while I'm on this subject, it is just as incorrect to play stereo recordings on loudspeaker binaural, which means that Bob's Sonic Holography is a fine thing but is only correct for recordings that were made binaurally. Interaural crosstalk is not a problem in the stereophonic system.

...[Regarding] Dave Clark's concept of standardizing the reflection signature of the playback situation...[let me] add that when it comes time to get serious about this stuff and sit down and completely specify the temporal, spectral, and spatial characteristics of the reverberation signature that Dave is talking about, we cannot do any better than to look at an image model of live sound. The image model completely specifies the intensities, direction of arrival, spectral content, time signatures, axial vs. power response, everything about all of the sound patterns that exist in the concert hall.

...If you agree with me that stereo phonic sound is a real, physical model of live sound, then perhaps you would also agree that the four basic parameters of fidelity that we can manipulate are (1) physical size, (2) power, (3) accuracy of the storage and transmission in the electronic domain, and (4) spatial characteristics. As John Eargle said, we no longer have a problem with accuracy in the electronics. Paying more money for a CD player will yield no material improvement in your sound, and the only characteristic of amplifiers we should be interested in is how much power they have. Given a choice between a $6000 dilettante audiophile 100 watt amplifier and a Carver M-1.5, definitely opt for the Carver. It would take thousands and thousands of watts of power to come close to the power of live sound, so the more the better. The physical size of the playback room is very important, larger rooms sounding closer to the real thing because the scale of the model is closer to life size. I'm not sure yet what the upper and lower limits are, but most reviewers are using rooms that are too small for high-

Comparison of Reproduction Image Model to Live Sound

The spatial "shape" of the sound that we actually are hearing is a function of the positioning and intensities of all acoustic images with respect to each other. In the reproduction, speak er positions and direct-to-reflected ratios are adjusted to better approximate the live model.

fidelity use. Finally, the spatial characteristic refers to the ratios of direct to early reflected sound in the playback image model and the building of a realistic surround field in your room. It is manipulated by means of the radiation pattern of the speakers, the positioning of the speakers in the room, and the acoustics of the room.

This is where the intense research is beginning to go, and it's about time.

Gary C. Eickmeier

Independent Industrial Designer Lakeland, FL

We found it necessary to trim down your excessively long letter but trust that we have managed to retain the essence of your argument. We have also read the pre print of your October 1989 AES paper,

"An Image Model Theory for Stereophonic Sound," which you were kind enough to enclose and which reveals that your theory is the basis for a new loudspeaker design you are in the process of developing. That puts things in a somewhat different perspective, making your argument a design rationale rather than a disinterested audio philosophy.

Even so, we find little if anything to contradict in your statements; everything you say is basically true, much of it widely recognized as such, but only your finished speaker is likely to convince us that yours is a higher truth in a discipline characterized mainly by trade-offs.

We have some minor quibbles; for one thing, your views on binaural vs. stereo seem to us a bit too black-and-white.

There is an occasional overlap or in between gray area. In 1959, when commercial stereo was very new, we attended a couple of the then-celebrated André Charlin's recording sessions for Erato in Paris. He used a single pair of Schoeps omni capsules on each side of a football like (oblate spheroidal) baffle. Nothing could have been more like a dummy head, yet it did not occur to him call these recordings binaural; they were considered the very finest in stereo. In Part II of the seminar transcript, in this very issue, you will find some comments by Stanley Lipshitz and Peter McGrath about the use of such baffles between closely spaced omnis in stereo recording. Nor is Bob Carver's Sonic Holography strictly "loudspeaker binaural" anymore; he has over the years added a few little tricks and wrinkles that widen the sweet spot for listening and make the system do its thing with a larger variety of recordings. We agree with you, however, that the makers of a stereophonic recording presumably anticipated the four different arrivals (L speaker to L ear, R speaker to R ear, L speaker to R ear, R speaker to L ear), of which Bob tries to wash out the last two, so that his processed wave launch is really an alteration of what we were intended to hear. We have been giving Bob a hard time about this for years, but like most highly creative individuals he has limited use for purists. On the other hand, you should never, never mention in the same breath the simple-minded copycat system inextricably embedded in the Polk SDA speakers.

Overall, we feel you are headed in the right, or at least a right, direction and are eagerly looking forward to the debut of your new loudspeaker.

-Ed.

The Audio Critic:

...Your magazine articles are very good, but the "Seminar 1989" is the very best I ever read in any hi-fi magazine to this date. Thank you very much! Very truly yours, Fermin Avilés Brooklyn, NY We are flattered by your enthusiasm, but think about this: The mainstream hi-fi slicks have no room, and certainly no editorial niche, for anything as lengthy and leisurely as our seminar. The tweako/dilettante high-end journals, on the other hand, lack the intellectual credibility to attract panelists of the caliber of ours. That leaves the field to us, pretty much by default.

-Ed.

The Audio Critic:

Upon reading the article on speaker placement for best bass response in Issue No. 13 of The Audio Critic, 1 am struck by the fact that all too few audiophiles and industry people realize how to point a loudspeaker, and where to sit and position their transducers for optimum sound reproduction.

The key word we should pay attention to is reproduction. A loudspeaker is a very simple electromechanical device for pushing air. As such, there are different axes on which a listener may fire the drive elements into the room.

Home hi-fi is now mature enough to hope that audiophiles might understand that a speaker's smoothest delivery in typical rooms is on axis. In other words, the acoustic center of the speaker should fire at a centrally located "sweet seat." In reality, however, most dynamic loudspeakers comprising multiple drive elements actually suffer a bit of crossover interference on axis. The best designs we are associated with sound smoothest 15° off axis or crossing directly behind the listener's head, for central imaging and a clean, open presentation of the soundstage.

As my seven-year old daughter would say, "It sounds more better, Pop." Oftentimes in travels to stores and fellow audiophile homes, I see speakers firing straight down the room. In other words, the listener is making value judgments based on 30° to 60° off-axis speaker response, which is any speaker's worst air movement. Off axis there can exist large peaks, broadband dips, and highly colored resonant ringing cycles. It is largely due to this off-axis listening that I believe many of the digital-haters have formed their opinions, in order to obtain a soundstage from their most presence-dominant vinyl discs. Perhaps making judgments using highly touted transducers with abysmally nonflat frequency responses has aided in coloring their viewpoint, too! All speaker manufacturers should explicitly state in the owner's manuals the optimum listening axis and height for nor mal distances in ordinary living rooms (14' by 22' approx.). But then that would be agreeing to minimum standards and applying some measure of science. Yet isn't acoustics just that-an applied science? Sincerely, John Otvés, President Waveform Research Otves Industries Brighton, Ont., Canada The Fourier 8e loudspeaker system, a 3-way design in which your Editor was heavily involved, used a mathematical compromise between optimum on-axis and 30° off-axis response to determine cross over network parameters and driver level matching. So, as you can see, some speaker designers do anticipate the problems you are talking about. (A fat lot of good it did in the case of Fourier.) The user should never assume automatically that the speakers must be toed in. It all depends on the design, and-we agree-the manufacturer should tell us about it.

-Ed.

The Audio Critic:

Mr. Rasnake's article on speaker placement in Issue No. 13 contains implications which I feel deserve some further comment.

I believe that Mr. Rasnake's article insufficiently emphasizes that the listening room, including the positioning of loud speakers and listeners, is almost always the weakest link by far in the chain of recording and reproduction, and thus any improvements that can be achieved, however much a compromise they may be, are probably more significant than other possible improvements in equipment or recording quality. Unfortunately, the room is likely to be the most expensive "component" and also the costliest to change. This is not to criticize "mere positioning" as a "cheap fix." In the opening paragraph he states that his analysis would be "especially useful if the room has not been built yet." However, that statement contains the assumption that people would deliberately build listening rooms with rectangular shapes and pairs of parallel surfaces. Apart from building a cubical room, that is the least desirable way to build for good acoustics using flat surfaces (assuming that curved surfaces are more costly to build and thus not considered). Even though per haps fewer than one room in a million is deliberately built with no parallel flat surfaces, this analysis is really merely an at tempt to make the best of a bad situation, caused by long-standing architectural and aesthetic prejudices, and the tyranny of T squares and triangles. It is not necessary to have walls that are out of plumb or sloping floors to achieve non-parallelism. According to Acoustical Designing in Architecture by Vern O. Knudsen and Cyril M.

Harris (of the firm that did the final fix on Avery Fisher Hall at Lincoln Center), as little as 5% offset is sufficient to prevent flutter echo, a related problem (1950 edition, pp. 171, 187). This would mean, in a 20' long (or wide) room, as little as 1' narrower at one end than the other, or 6" wider or narrower at the middle compared to the ends. Similarly, a 10' ceiling would be dropped 6" at one end or side, or 3" at both ends (or sides) or across a center line. This may be enough to be noticeable, but not enough to make anyone dizzy or disorient ed, or cause problems with the use of conventional furniture, or cause the loss of any significant amount of space.

Clearly then, creating a good listening room does not require the duplication of Berlin's Philharmonic Hall in miniature, or anything even remotely so exotic architecturally. The fact that fairly simple mathematical analyses can be applied to fairly simple room shapes is also no argument for using such shapes just to be sure that "at least we know what we've got." More complex shapes for the most part have less need of the necessarily more complex analysis until one reaches the size and complexity of large auditoriums, with the likelihood of some curved surfaces, overhanging bal conies and, of course, high costs of construction or renovation.

Rasnake says very little about positioning the listener, except that listener distance to room length should be within a range related to speaker to room length distance. But just as speakers should not be placed near low-bass resonant areas, neither should the listeners, if the smoothest response is sought. This should be in terms of length, width, height, and diagonals, and for stereo equidistant to and with an angle of from about 40° to 60° between the speakers. But the mirror image positioning he suggests for the speakers would place the listener midway from left to right, rather than the more desirable splits of from 1/5, 4/5 to 4/9, 5/9 with 1/3, 2/3 the best.

To achieve this the speakers cannot be positioned as mirror images, and either ¥ and Z dimensions must be interchanged or two different solutions used. The benefit achieved will hopefully be greater than the differences introduced in left and right channel response.

Thus the best arrangement for loud speakers and listening position(s) in a sym metrical room will be asymmetrical. One must choose between visual aesthetics and aural smoothness or modify one's aesthetic standards.

Perhaps it should be noted that the proper placement of loudspeakers and listening position(s) is not going to eliminate room resonances, much less turn your room into an anechoic chamber. But it will allow you to avoid undue excitation of resonances and undue exposure to them, particularly in the low bass, which is where they cause the greatest response irregularities. Recordists should also use Rasnake's method to position microphones and musicians, and conductors should use it to place instruments with deep bass capabilities.

In the usual size of listening room, say 16' wide, the difference in distance between the listener's ears, say 5%", is enough to place the two ears at different nodes, such as 1/3 and 4/11, with 5/16 and 3/8 close by on either side. Although Rasnake's Figure 3 only divides the room into eighths [must be meaning Table 3, but those divisions are not eighths-Ed.], this represents only the fundamental and first two harmonics, whereas his Figures 1 and 2 deal with harmonics through the fifth

[actually sixth-Ed.]. However, as 16' rep resents about 35 Hz, no more than eight times this frequency is likely to be critical, perhaps as little as four times it. If only resonant areas two feet apart need be considered, then probably 1/7, 1/9, and 4/11 should be eliminated from consideration

[Rasnake never mentions 4/11 -Ed.] in at least the shorter two dimensions (of the average listening room), as nodes and resonant areas are so close together as to create areas of steep pressure differentials and thus too critical of exact placement for practical purposes.

Node and Decimal and Inches decimal resonance difference in 16 feet 19 111 125 18 2.67 7 143 125 1/8 3.43 4/11 364 375 3/8 218 but the following are somewhat better 1/11 091 125 1/8 6.55 2/9 222 250 1/4 533 2/7 286 1250 1/4 6.86 2/5 400 375 3/8 4.80 Apparently the sideways placement of the head in an ordinary room is very critical, and one might also wonder about the use of loudspeakers with woofers larger than 8" in diameter (effective cone diameter 6"), or considering a 7 1/2 to 8-foot ceiling, larger than 4" (effective cone diameter 3"), as at least one part of the cone would be at an undesirable location in any case.

Anyone who considers how close in actual distance the above points are in an average room may well despair of any ideal or easy "fix" for the problems caused by parallel surfaces.

It is not that one avoids exciting or hearing resonances by locating at thirds or fifths, or avoids power-sapping nodes by avoiding halves, fourths and eighths, but rather that by using other ratios these occur at higher frequencies where the more frequent occurrence of resonances tends to smooth out the overall effect. There are no

"dead" spots, only many which are "artificially enlivened" and some which are fairly normal at low frequencies. Any position will be for some frequencies a resonant area and for others nodes of little or no pressure change. But some positions have far more resonant frequencies and/or nodes than do others. One might say that different points have widely varying frequency response, and none are flat.

Sincerely, John F. Sprague

We passed your extremely interesting letter on to Bill Rasnake and asked him to write a reply. He said he would but he never did, even though we gave him plenty of time. It is true that he is a very busy guy, professionally as well as avocationally.

Maybe we can publish some comments by him in the next issue. He did say something to us on the telephone to the effect that once the speaker loads into the room with out exciting significant resonances, you will not then hear such resonances by sit ting in a certain spot or holding your head in a certain position. He also agreed that the differences you analyze are indeed problematically small in small rooms.

-Fd.

The Audio Critic:

After 22 years in the audio industry, I have decided to place my first advertisement in your audio publication. I have never advertised prior to this point, in the belief that when one is involved in fundamental research on the level that I am, the product is so innovative and of such high caliber that it should sell itself.

While other audio manufacturers gloss over engineering deficiencies in their products by spending large amounts of funds on advertising, in the hope that such deficiencies will not become apparent until they have realized a short-term profit, I have put all that money into research and development. I cannot put all the blame on most of the audio companies either, because of the "no ad, no review" policies of some audio magazines. However, your publication from the very start has repeatedly given me excellent reviews on all my products because of creativity and their merits.

I now feel that the maturation of my audio line is at a point where I can increase my product exposure through advertising in your respected audio journal, The Audio Critic.

Sincerely, Sao Zaw Win CEO Win Research Group, Inc.

We agree that we are quicker to understand creativity such as yours than the dilettante cult journals. Also, at our full page ad rate of $335, the temptation to be venal is pitifully small, even if we had no Journalistic ethics whatsoever. But the best reason to advertise in our pages is that some of the finest and most influential minds in audio will be exposed to your ad.

-Ed.

The Audio Critic:

"Dear" Peter, Attacking the person instead of his or her ideas is not a new technique. Its [sic] been used by intellectually and emotionally (in your case add financially) bankrupt individuals throughout history. Before you engage in such ugly tactics again in regards [sic] to me, please spare your reader(s) your petty delusions and send for a resume.

You'll find I have a degree from Cornell University and an Academy Award nominated motion picture soundtrack to my credit, among other accomplishments.

But even if was [sic] a high school drop out, my credibility as a reporter and as a knowledgeable, keen-eared music and equipment reviewer in the high end audio industry is by now well established. I have no doubt that as your bratty, self indulgent, undisciplined, indefensible slur circulates (oozes up from the muck is more like it), you'll be hearing about it from other members of the high end community.

That out of the way, perhaps we can discuss digital recording, CDs and my Music Connection/Goldmine commentary, which by the way, was never submitted to The Absolute Sound. Another delusion of yours. The piece was meant to be provocative. How else do you fight four years of "perfect sound forever" out and out lies in the general press? I wanted to prevent Goldmine readers from making the mis take of their lives: trading in original pressings for CD reissues.

When I wrote the piece for Music Connection, "laser rot" was indeed a very controversial topic. Billboard was covering it weekly and an executive for Nimbus admitted that the inks they were using had indeed eaten through discs. How many other companies were using such inks? No one knew then.

There is now a glut of CD production facilities. Some plants have gone under.

Competition is severe and wholesale prices are dropping. So is quality. Substandard production is yielding discs with thin and sometimes incomplete coatings. I've seen them and I question their longevity. Time will tell.

Stale canard? I think not. But if I was writing the piece today, Id concentrate on the artificial sound. I've yet to hear a digitally remastered analogue recording on CD that sounds as good as a well mastered original analogue record. The problem is not with the analog source. Its [sic] with the digital mastering equipment.

You delude yourself again if you think there's unanimity in the recording industry regarding digital recording and CDs. I've interviewed many engineers and CD masterers who admit both on and off the record that digital recording has many problems, particularly at the frequency extremes, and I've quote them in TAS.

Keith Johnson's CDs? I've heard them. Pretty good for CD. But I've also asked Keith Johnson about them. You obviously haven’t [sic]. He'll tell you he thinks his LPs sound better. What's more, he'll detail for you-complete with oscilloscope tracings-why there are serious problems with today's CD technology that cause the sonic deficiencies I, and other discerning listeners hear, and find thoroughly objectionable-more so [sic] than the problems we hear with analogue.

As for Tom Jung's CDs, to me they sound thick, dull, and airless-nothing at all like real music. The drum sound he goes for is sterile and gimmicky. His preferred mix-panning the drums across the soundstage, is to my ears, just plain silly.

DMP recordings have as much to do with musical realism as did Enoch Light and The Light Brigade's Command recordings from the fifties and sixties. But unlike you, while I criticize Jung's taste, I don't question his credentials or his intelligence, or his "right" to record music as he sees fit- or his preference for CD.

I stand by every other characterization of the current state of digital recording and compact disc sound that you quoted and I really don't care whether I'm in the minority or the majority. I know what I hear, and I have too good a track record at this point to care about majority rule. Reagan was president for eight years.

Which brings me to the part of your bilious rant that does disturb me: your implication that any publication that chooses to run a piece that contradicts your opinion is somehow suspect. Tantrum? Credibility? You had better look in the mirror.

Sincerely, Michael Fremer, Senior Music

EDITOR: Popular The Absolute Sound Sea Cliff, NY

P.S.: You owe me a half a year subscription to your magazine from 1979.

The above is printed exactly as you wrote it, with all errors unedited; we have italicized the names of publications for the sake of clarity but made no other effort to make a hostile letter appear more literate than it actually is.

You remind us of a series of old jokes that took the form of "I'll never forget what's-her-name" and "Support mental health or I'll kill you." Your own protestations on the subject, in your own letter, are the proof that you are bratty, self-indulgent, undisciplined, etc. Any brat with an IQ of 90 can learn the words "intellectually and emotionally bankrupt" and then throw them like stones at anyone he gets mad at, regardless of the context. (Were you a spitter and biter in your fourth-grade fights?) As for your imputation of financial bankruptcy, it is more than bratty; one has to be mindlessly irresponsible to put some thing like that in writing without having seen a financial statement specifying assets and liabilities. Where do you get your facts? One would expect a Cornell graduate (1) to be aware that "provocative" statements tend to provoke people and (2) not to react to the inevitably ensuing criticism as if it were an outrage crying to heaven. If you call digital recording "a sick joke" and its endorsement by the press "out-and out lies," then you have be prepared to take a little heat yourself. You overstate your case so grotesquely that serious people will not take you seriously. You come off as an attention-seeking lightweight.

Note that we are not entering the analog vs. digital debate here all over again. It occupies a good many pages of the seminar transcript in this issue; we have little to add at this juncture. We do want to point out, however, that we never said that The Absolute Sound had turned down your article, only that they would not have published it even ifyou had submitted it. Also, the missing part of your old subscription has meanwhile been fulfilled. You just had to ask, like several thousand others; we had advertised the fulfillment offer for many months.

-Ed.

-------

[adapted from TAC 14]

---------

Also see:

Various audio and high-fidelity magazines