Absolute Issues Page 2

There is a wider point to be made concerning Mr. Self's conclusions. The danger lies not in his beliefs but in his insistence that those of us who rejoice in all the different shadings of sound quality possible from amplifiers—those of us who not only recognize "brown" but appreciate all the different browns bequeathed us by the Almighty—should cease to do so. Mr. Self's beliefs are a subset of my own since, if differences exist, it is quite possible for those differences to be small enough to be unnoticeable, whereas Mr. Self's world view, like Dr. Shanefield's, allows for no other. But apart from its dictatorial tone, for us to accept without question the complacency inherent in his article is to welcome stagnation. If every area of amplifier performance can be "proved" to be effectively perfect, in that it doesn't affect the resultant sound quality, then the reductio ad absurdum conclusion is that amplifiers must have reached perfection; ie, apart from recording problems and the fact that there is still a pair of loudspeakers in the chain, it should be impossible to distinguish between live and recorded sound.

I am sure that he wouldn't actually put it like that, but the logic of Mr. Self's arguments—with every relevant parameter, present-day performance is at least as adequate as it need be—leads irrevocably to such a conclusion.

Concert-going reveals its absurdity.

A self-serving "technical orthodoxy," as defined by Mr. Self, seems to be more concerned with preserving the status of engineers suffering from a failure of imagination rather than with reproducing music in the home. But, as the BBC's Hilary Lawson said in 1985: "Science is there to be used, not to dictate what is true." Mr. Self's article illustrates physicist Fritjof Capra's examination of the subjective nature of objectivity in his book The Turning Point: "The patterns scientists observe in nature are intimately connected with the patterns of their minds; with their concepts, thoughts and values. Thus the scientific results they obtain and the technological applications they investigate will be conditioned by their frame of mind."

Exactly! The state of the art is not advanced by a rejection of the opinions of those who may not be technical but who actually make use of the end-products. Dr. Johnson once wrote that one doesn't have to be a carpenter to be able to judge the quality of a table: I want engineers to make better tables, not to spew forth tracts, no matter how well-written, telling me that my need for a better table is proscribed by orthodoxy.

A controversial area hinted at by Douglas Self was that if all subjective differences are imaginary, then those who write about those differences will be influenced by arbitrary factors, such as xenophobia. You can see from Mr. Van Den Heuvel's letter in this issue, for example, that he feels that my value judgments in Stereophile owe more to my British nationality than to the intrinsic merit of the products I recommend. The accusation is absurd, but I would like to say that the recommendation of any product in this magazine is based primarily on its ability to play music. That inevitably means that some components I have written positively about will have been made in the country of my birth, though of those that have had the longest tenure in my reference system, only the Linn turntable and the Celestion SL600 have sang-pure UK bloodlines.

My preamplifiers over the last five years—Audio Research SP10, Krell KRS2, Mark Levinson No.26—have all been American, as, with the exception of the VTL 100W monos, have been my preferred power amplifiers, the Krell KSA-50, KSA-100, and Mark Levinson No.20. And when it comes to CD playback, my favorites—the CAL Tempest, Stax, and Accuphase players—are two-thirds oriental. Of my favorite cartridges, the Koetsu Rosewood is Japanese, as is the Linn Troika (by way of Scotland).

But yes, I am biased, if by "bias," you mean that as a result of my experience, I have arrived at certain beliefs. I believe strongly that, with the exception of such models as the Magnepan SMGa, Thiel CS1, Synthesis LM210, Spica TC50 and Angelus, Snell Type Q, and Vandersteen 2C, the UK leads the world in designing affordable, true high-end loudspeakers, though my enthusiasm is tempered by the inevitable Transatlantic price increase which renders them less competitive in the USA. However, putting to one side the B&W 801 Matrix, UK designers seem to suffer from a collective failure of nerve when it comes to cost-no-object design. The laurels here are firmly grasped by such American models as the Infinity IRS Beta and Apogee Diva, which couple the delicacy of imaging and tonal accuracy featured by the best UK designs with realistic dynamic range and the ability to reproduce low frequencies with true weight and extension.

And when it comes to vinyl disc replay, the only two turntables I have heard that would tempt me away from my Linn are the Versa Dynamics and the Goldmund Reference, neither of which I have any likelihood of being able to afford. So the Linn it remains.

With electronics, I feel that the US still leads the way, particularly in preamplifier design, where some models achieve the impossible trick of maximizing neutrality and transparency without losing sight of the ability to play music. Though this latter is something the British have always rated highly, their preamps have tended to fall short on the first two aspects, leading to a rather "ragged" presentation of the music. And I still believe that it takes a very good solid-state design to equal the performance regularly offered by tubes.

If these attitudes are dismissed as bias, then so be it. But rest assured that I will continue to call 'em as I see 'em in my reviews, and when a speaker comes along that stomps all over my beloved SL600 in the areas that it excels, or equals it in those areas and offers realistic bass extension and dynamics, then I will gladly switch allegiances, no matter what that speaker's country of origin—though the Acoustic Energy AE1 and Celestion SL700 I review in this issue are most definitely English, I'm afraid.

Absolute Phase
A topic addressed in HFN/RR's interview with Peter Walker was whether or not reversing a signal's polarity, its so-called "absolute phase," would be audible or not on music. Though JGH is pretty much convinced that some people can detect polarity differences, he admits in this month's review of the Infinity IRS Beta that he himself cannot; in his review of the Mobile Fidelity CD of Dark Side of the Moon, Kevin Conklin finds that having the absolute phase correct has a significant effect on the music. Which brings me to a book by the aforementioned Clark Johnsen that should be on every stereophile's reading list. Entitled The Wood Effect, it is a thorough examination of the history of Absolute Phase audibility, expanding into a more general exposition on some of the wrong turns taken during the development of sound reproduction, one such being the abandonment of the 78 (footnote 5).

Clark (footnote 6) makes absolutely clear his position on the importance of absolute polarity reversal: "Whoever cannot recognize Absolute Polarity, shall be deemed SUPERFLUOUS," which seems a little too absolute, considering that JGH is one the least superfluous audio commentators around. But why should polarity matter? If an acoustic compression at the original event is reproduced as an acoustic rarefaction when the recording is played back, electrically there should be no difference at all. Additionally, the eminent acoustician Helmholtz stated in 1862 (footnote 7) that "the quality of the musical portion of a compound tone depends solely on the number and relative strength of its partial simple tones and in no respect of their differences of phase," and many people feel that Helmholtz said all that needed to be said on the subject.

However, as documented by Johnsen in The Wood Effect (and also in an article I wrote for HFN/RR in 1980, footnote 8), nearly all the academic work performed since Helmholtz suggests that the human ear can detect acoustic polarity differences, although all research indicates that the effect is sometimes subtle. (The book's title comes from work carried out in 1957 by Charles L. Wood of the Defense Research Laboratory, who found that a sinewave, clipped on one side of the time axis only to render it asymmetrical, took on a different timbre when its polarity was reversed.) Work by Stanley Lipshitz in the late '70s (footnote 9), using carefully organized double-blind testing, confirmed that a reversal of absolute signal polarity will be subtly audible on music to a 99% confidence limit! (footnote 10) (Indeed, it is one of the few things that can be reliably detected with double-blind testing.)

There is even a mechanism agreed upon as to why the ear should be able to detect the supposedly undetectable. The nerves attached to the frequency-discriminating hair-cells in the inner ear only fire on the positive-going parts of the waveform, indicating that the ear acts as a half-wave rectifier. It will thus produce a different output to the brain on asymmetric signals if the absolute polarity is reversed. A music signal, unlike a sinewave, is not symmetrical about the time axis, other than that over the long term there are equal amounts of energy on either side, and it would be expected that if an original compression was reproduced as a rarefaction, you would be able to hear it.

What does it sound like? "Wrong polarity is the muffling distortion," writes Johnsen, and it seems generally agreed that to listen to a recording with the wrong polarity is to suffer from a lack of realism, a lack of body to instrumental tone, a lack of integration within the soundfield, and less natural-sounding applause. Unfortunately, as there is no necessity for engineers to preserve absolute polarity during the production of a recording, a recording has a 50% chance of being wrong. In fact, as reported in The Wood Effect, which includes considerable documentation of the subject, Japanese-pressed LPs even have alternate tracks in opposite polarity!

The aspect of Clark's book that I found fascinating, and one which ties in with the general theme running through this essay, is that despite the evidence for its existence, despite there being almost no opposing evidence, the engineering establishment seems to dismiss the matter of the ear's sensitivity to absolute signal polarity. (One would almost think that we were talking about amplifier differences here.) The Wood Effect contains a thorough examination of the apparent philosophy of the Audio Engineering Society, as expressed through the choice of papers to reprint in its Journal. Johnsen implies strongly that only the results of research work which conforms to orthodoxy will benefit from a wide dissemination in the JAES, and that that orthodoxy sticks strictly to Helmholtzian theory in denying the importance of phase effects at all.

There is, of course, the notable exception of the published work by Stanley Lipshitz et al referred to in footnote 9—Stanley's formidable mind brooks very little obstruction in its chosen path—but I don't think it coincidental that Douglas Self behaves as expected in his E&WW article in attempting to cast doubt on the Lipshitz findings on the audibility of phase errors. Once the door is opened a crack, then it could be pushed open even further, and—heavens!—amplifiers might be found to be audibly different after all! Better to dismiss the whole subject altogether, and if that means dismissing something such as absolute signal polarity that has without question a real effect on the quality of reproduced sound, then so be it.

Thus is the thrust of Clark Johnsen's tour de force, and if you feel that the reality of the matter is not so, then let me conclude with a quotation from the BAS Speaker by David Moran, once of dbx and President of the Boston Audio Society, which also concludes The Wood Effect. It neatly illustrates, in my opinion, exactly how "technical orthodoxy" is more concerned with preserving establishment attitudes than with improving the quality of music in the home:

"These are interesting times in audio. The smoke that is blown is now being couched in the language of science, and no longer in the language of marketing. Waveforms of oboes, discussions of room acoustics, etc., with footnotes to the JAES et al., will be given, along with contentions that clarinets will suck instead of blow if speakers are not properly phased—all couched in terms of, 'You've heard hype before: this is science.' "

The Wood Effect, by R.C. Johnsen, cost $7.95 when it was first published and could be obtained at The Modern Audio Association, 23 Stillings Street, Boston, MA 02210. It is out of print in 2006 but copies might still be obtainable from the author.



Footnote 5: Absolute Phase should not be confused with relative phase, when the speakers are out of phase with each other. Johnsen here appears to confuse the two on p.15 of The Wood Effect, where it seems obvious that in the passage quoted from Jens Blauert's Spatial Hearing (MIT Press 1982), Blauert is talking about speakers being out of relative, not absolute, phase.

Footnote 6: The author is a Harvard Physics graduate, with published academic work in the fields of image processing and the measurement of sight objects. He has worked on holography, surveillance satellites, the Mars Lander camera, and the Orbiting Space Telescope. He now runs "The Listening Studio" in Boston and recently founded The Modern Audio Association to advance the state of music reproduction art "without commercial gain or pressure."

Footnote 7: On the sensations of tone, H.L.F. Helmholtz; Dover, 1954.

Footnote 8: "Listening Tests and Absolute Phase," HFN/RR, November 1980.

Footnote 9: "A little understood factor in A/B testing," The BAS Speaker, March 1979, followed by "On the Audibility of Midrange Phase Distortion in Audio Systems" (with John Vanderkooy and Mark Pocock), JAES, Vol.30 No.9, September 1982.

Footnote 10: It subsequently emerged that the 99% confidence limit applied to test results using both music and asymmetrical test tones, not music alone.—John Atkinson

Article Contents
Share | |
Site Map / Direct Links