Rounding Up the Usual Suspects
A third of an audio century later, High Fidelity is long gone and, much to the surprise of those of us who work on it, Stereophile appears to be regarded by some as the new Establishment in audio publishing (footnote 1). With a regular publication schedule, an average issue size of 276 large pages compared with the average of 40 small pages when Larry Archibald purchased the magazine in 1982, and, I believe, consistently excellent standards of writing and presentation, this magazine's circulation has grown from 3000 in 1982 to 80,000 today. [87,600 in 2000, though the average issue size has dropped to 208 pages.—Ed.]
I don't see any virtue in remaining small and exclusive. To the contrary, I believe any magazine's fundamental goal is to spread what it believes in to as wide an audience as possible. While it may have come in from the cold, Stereophile has remained true to the tenets on which it was founded: we put our readers' interests first; and we judge audio components on how they sound.
But if you read the new generation of underground magazines that has been drawn into existence both by our growth and by the diminution and demise of other titles, you would think Stereophile has somehow betrayed the high-end faith. "Masters of the Universe," grumbles one (footnote 2). "Rigorously scientific rationalists in Santa Fe," complains another (footnote 3). Such criticisms may be sincere, but their authors forget that, ultimately, readers don't read your magazine to find out what you think about what other magazines do, they buy it to read about what you are doing. To define your editorial identity in terms of what another magazine does is a losing strategy.
The one item that appears to unite our audio-journalist brethren is their distaste for Stereophile's inclusion of standardized measurements in its equipment reports. Those who subscribe to this horror for "graphs," however, miss the point. The integration of measurements into Stereophile's reviews is not to describe or replace the listening experience—that is, and probably will always be, impossible. Without listening, there is no way, for example, of measuring something as universally perceptible as the quality of a stereo soundstage.
In fact, the performance of any audio component is a multidimensional entity. Just about every subjective performance parameter is affected by more than one technical aspect of a component's design. To be feasible, however, any specific measurement must be reduced to two or, at best, three dimensions—as in Stereophile's loudspeaker waterfall plots, where amplitude is plotted against time and frequency or direction and frequency.
This pragmatic process of reduction makes it very easy, therefore, for a measurement regime to amount to nothing more than a roundup of the usual technical suspects. As Chris Sommovigo, the designer of the excellent Illuminati digital datalinks, wrote some time ago on The Audiophile Network bulletin board, "The fundamental incongruity in using objectively-obtained data to determine the value of a particular component lies in the fact that, although the data may be objectively obtained, the 'determination and evaluation' are most certainly subjective. The assumption that objectively-obtained data can be interpreted objectively is a deadly fallacy."
The concept of "value" or "quality" must always be a subjective phenomenon. But to imply that objectivism has no place in the assessment of equipment intended to reproduce music is as untenable a position as that of the engineer who insists that all relevant parameters of performance are known. A complete review needs to combine both subjective and objective assessments: one to indicate what is happening; the other to give a measure of its relevance and its worth.
The measurements in Stereophile's reviews support the reviewer's observations on the component's sound quality with possible or probable explanations. A secondary role is to provide specific information about a component's needs regarding use and potential matching. Decent measurements also validate a design and, by implication, its designer's competence. As Robert Harley wrote last March (p.63), "the better-measuring products won't always sound better, but a good-sounding, well-engineered component inspires more confidence than one that performs poorly on the bench."
Stereophile is also slowly building up a database of what measurements good-sounding products have in common. In the long term, patterns indicating a causal connection between what is heard and what is measured will emerge from that mountain of data. We may be a long way from that enlightening goal—although I believe it has already started to happen with loudspeakers—but if we don't measure at all, we will remain in the dark.—John Atkinson
Footnote 1: See Tom Davis's "The High End at the Razor's Edge," The Abso!ute Sound, Issue 104, p.52.
Footnote 2: Tom Miiller in The Audio Adventure, May 1995, Vol.2 No.5, p.2.
Footnote 3: Andrew Keen in The Listener, Vol.1 No.4, Autumn 1995, p.15.