A Matter of Dimensions Page 3

The sound has a "where," which already covers three dimensions. It has a "tone," which includes pitch and timbre, themselves independent variables. Its intensity, a "how much?" that varies with time, represents its dynamics. It has a "when" aspect. The listener's instantaneous perception of musical values depends very much on what has gone before. And which of these aspects is the most important when assessing quality will be different for each listener.

And is "objectivity" even the right tool to assess worth in an area which uses technology in the service of art? You can't measure the difference between a good and an inadequate performance of, say, a Mozart piano concerto. The reductionism inherent in Scientific Method will inform you of "how" the two performances have been constructed, but it will not tell you which is "better." "Quality" can only be inferred via a holistic approach, which is of necessity subjective.

When we leave the conventional world of electronics and loudspeakers and look at components for which a set of measurements to assess their worth has still to be developed, the test-bench engineer is in big trouble, as can be seen from Peter Mitchell's discussion of the recent Consumer Reports review of CD SoundRings in this issue's "Industry Update." Does the fact that their effect cannot yet be measured mean that they have no worth? No, of course not. As the letter from Stew Glick on p.27 points out, if the SoundRings do make a sonic difference, that is all that is needed to justify their use. (Providing, Mr. Glick, that they're not actually making things worse, something that Sam Tellig has been wrestling with in his "Audio Anarchist" column.)

The anguish and the inflamed rhetoric in the responses to Peter's comments from Noel Lee, Bill Low, and Mike Goldfield in this issue's "Manufacturers' Comments" to me reinforce the fact that all the Rings have to justify their use is the fact that they change the sound. If absolute evidence for a SoundRing's ability to improve sound quality did exist, we would have received three much shorter letters.

Hold on, I hear you saying. Steven Sims's response does outline a measurement that shows the Rings to have a beneficial effect. They appear to reduce the amount of jitter in the raw FM signal recovered from the CD. But, with respect to Mr. Sims, this may be true but his letter is disingenuous in that it entirely fails to address the question of whether ringing a disc will make it sound better. He attempts to jump from a proof that A=B to a sonic result, C, without showing any primary relationship between either A or B and C.

To examine his points in more detail, time jitter in the FM signal will increase the likelihood that a "1" code will be mistaken for a "0," and vice versa. The error rate in the raw signal off-disc will therefore be reduced by using a ring. But this doesn't matter, as such errors will always be 100% corrected so that they cannot—repeat, cannot—lead to a degradation in sound quality! Only when the intrinsic error rate on the disc is already so high that the additional jitter error pushes the player's error correction over the edge, forcing it to interpolate or mute, will such errors affect the sound (footnote 5).

This was the gist of Peter's own experiments; if Mr. Sims really feels that this particular measurement is the one to reveal the sonic benefit of ringing a CD, then I feel it incumbent on him to show why. Until then, we have just anecdotal evidence, such as that provided in a letter also on p.27 from a Stereophile reader, Bernard Engholm, who found that use of a CD SoundRing enabled both his CD players to cope better with the catastrophic gaps in the recorded data on the Pierre Verany test disc.

But remember (to paraphrase Heyser): Because no one can convincingly find a measured reason for the Rings to bring about sonic improvements does not mean that they don't. I think they generally give an improvement; Sam Tellig thinks they generally bring about a degradation. What do you think?



Footnote 5: With the benefit of 13 years' worth of hindsight, I was taking too simplistic a view of the situation. The time-base error in the raw FM signal recovered from the disc propagates through the CD player as word-clock jitter; while this jitter can be low-pass-filtered, it cannot be eliminated entirely, and will therefore affect the reconstructed analog waveform.
ARTICLE CONTENTS

X