A Matter of Dimensions

I have been reading a lot of late. Whether it is due to the reduced appeal of recorded music owing to the ever-decreasing shelves of LPs in our local specialty record store (the owner explains that he still wants to sell LPs; it's the record companies that make it increasingly harder for him to do so with punitive returns policies and deaf ears to back orders), or the fact that it's Spring, I don't know. But the fact remains that I have recently found myself devouring a shelf-full of titles sometimes only vaguely related—horrors!—to high fidelity. Stuart Chase's The Tyranny of Words, for example, first published in 1938 and a finer examination of what came to be called semantics you wouldn't want to find, should be essential reading for anyone involved in writing articles that are still intended to communicate some meaning.

I also found it hard to put down the second, and presumably final, volume of the late Richard Feynman's non-autobiography—What Do You Care What Other People Think?. This collection a) proves that, of all scientists, theoretical physicists seem to have their feet most firmly on the bedrock of reality, and b) by including the full text of Feynman's appendix to the official report on the Challenger disaster, demonstrates exactly how large institutions become more concerned with preserving their own existences than with carrying out their raisons d'être (a trap that high-end companies have so far nearly all managed to avoid).

But the book that kept creeping back into my life was the comprehensive collection of the published papers of the late Richard Heyser, published by the Audio Engineering Society under the basically accurate but still somewhat misleading title, Time Delay Spectrometry (footnote 1). Each 10 minutes' worth of reading this accessible volume inexorably led to 10 hours' worth of thinking. It should be on every engineer's and every serious audiophile's "must buy" list.

I say "misleading" because, although much of Richard Heyser's audio-related academic work did involve time-delay spectrometry, which he derived as a means for assessing the anechoic response of a loudspeaker in a normal room back in 1967 (footnote 2), the philosophical thrust of his writing involved, as did Richard Feynman's, the wider implications of what at first might be thought trivial. In particular, Heyser, who not coincidentally was also a physicist, was concerned with exploring something that deeply involves this magazine: the social conflict between those who listen and those who measure, and the apparent dichotomy between what is measured and what is heard. To give a taste of his thoughts (and food for your own), I'll offer a not-so-random collection of quotations pulled from Time Delay Spectrometry:

"The evaluation of the acoustics of loudspeakers and the room containing them proved to represent a microcosm of all the difficult problems in wave propagation."

"Nature does not solve equations."

"The thing we call time in audio measurements and the things we call frequency are different coordinates for describing precisely the same signal."

"Pitch is not frequency."

"At the present state of sound reproduction technology, the audio engineer shares the professional goal of a magician."

"The effect that modern sound reproduction strives to achieve is the creation of an acceptable illusion in the mind of the listener."

"If we wish to understand how to 'measure' what we 'hear,' then we must deal with subjective perception and the illusion of sound."

"The actual sound field in a listening environment is not identical to the sound field which we may perceive..."

"The end product is the listening experience."

"One of the worst-kept secrets in audio engineering is that what we hear does not always correlate with what we measure."

"Those whose principal professional involvement is based on the listening experience tend to develop a subjective viewpoint with value judgments seldom related to instrumental measurement."

"One of the most belittling experiences is to deride the 'black art' of a craftsman who gets consistent results by a certain ritual which he cannot explain and then to discover that his actions in fact held a deeper technical significance than we understood at that time from our simplified model."

"If we measure the frequency response of a system, and do it correctly, then we know everything about the response of that system. We have all the technical information needed to describe how that system will 'sound.' But the information we have is not in a system of coordinates that will be recognizable by a subjectively oriented listener...That is the root cause of the continuing fight between subjective and objective audio. It is not that either is more correct than the other...rather it is due to the fact they do not speak the same language."

"The next time you hear an argument between a technologist and golden ear about the audibility of certain types of distortion...is it possible they do not agree because each have [sic] a view through a different window?"

"You out there, Golden Ears, the person who couldn't care less about present technical measurements but thinks of sound as a holistic experience. You're right, you know."

Footnote 1: Time Delay Spectrometry, softbound, 280 pages, 8.25" by 11.25", is available from Audio Engineering Society Inc., 60 East 42nd Street, Room 2520, New York, NY 10165-0075. $27 to AES members, $30 to nonmembers.

Footnote 2: The idea is so simple it took a genius to think of it. The TDS technique, now commercially implemented in analyzers from Brüel & Kjaer and Techron (Crown), uses a swept tone to drive the loudspeaker under test and a synchronized but delayed swept bandpass filter. When the time delay is arranged to exactly match the transit time from the speaker to the measuring microphone, any room reflections of that tone, and the general room reverberation, etc., are rejected by the filter's having moved to a different center frequency by the time they arrive at the microphone. The result is that the effect of the room is completely (within the filter's limits for out-of-band rejection) removed from the measurement.