Music Before Numbers?
Editor: How did Kalman Rubinson arrive at such a glowing review of the Blue Circle BC22 amplifier in February, despite its barely adequate measurements?
Why are all the reviewers from other audio magazines in agreement? Because Blue Circle's Gilbert Yeung is right! Music before numbers! That's why I bought one.—Mike Pageau, email@example.com
"As We See It"
It was the question I'd been dreading: "Yo, JA, in what 'Recommended Components' Class are we going to put the Cary CAD-1610-SE?"
I must have looked to my inquisitor, Jonathan Scull, like a president caught in the headlights. For the Cary was the humongous, two-storey, single-ended monoblock amplifier costing $40,000/pair that Jonathan had found to sound wonderful in his system when he reviewed it last December, yet had failed so utterly to impress me on the test bench that I had dismissed it as both "silly-priced" and just plain "silly."
The conflict between sound quality and measured behavior has been raging at least since it caught fire in the Ultralinear Wars of the early 1950s, and was probably smoldering not too long after the introduction of electrical recording in 1927. But no product of which I'm aware has thrown the disparity into such sharp relief as the big Cary. It wasn't that there was only a small degree of overlap between the two areas of performance—there was no overlap at all!
As I've written a number of times now in this space, the "Measurements" sidebar of a Stereophile equipment report leaves nowhere for a product to hide. Inadequate or overoptimistic engineering, poor quality assurance, an idiosyncratic design philosophy—all are laid bare in a few paragraphs' worth of measured data. This is one of the reasons that, almost alone among US audio magazines, we publish those measurements.
But as reader Mike Pageau points out in this issue's "Letters," a component's reproduction of music must take preference over its "numbers." Indeed, Stereophile was founded almost 40 years ago on the idea that it is how a component sounds that matters most, not how it measures. As the old saw has it, "If it measures bad but sounds good, you're measuring the wrong things." In the case of the Cary, however—or even the Lexicon MC-1, which was reviewed last month by Larry Greenhill (see "Manufacturers' Comments," p.229)—the question of whether a product sounds the way it does because of its measured performance or despite it cannot be avoided. Unfortunately, neither can it be answered, at least not to the satisfaction of both "subjectivists" and "objectivists."
As you can see in this issue's "Recommended Components," I gave the benefit of the doubt to the Cary '1610, as well as to some other products that have raised my meter-reader eyebrows. If one or more Stereophile reviewers thinks highly enough of a product's sound to nominate it for recommendation, then I will respect their expertise and their experience as critical listeners. But when the measured performance is at odds with the perceived sound quality, then it becomes even more important that readers follow our guidelines on how to use "Recommended Components":
• Evaluate your room, your source material and front-end(s), your speakers, and your tastes. What will be the right product for your needs will not necessarily be the right one for someone else, which is why we list so many different and different-sounding components.
• Compile a short list of contenders that seem suitable for your needs.
• Read the Stereophile reviews of those contenders. (Reviews available free of charge in the Stereophile.com archives are marked "WWW" in the listing.) Even read reviews in other magazines.
• And, most important, before seriously contemplating a purchase, you must audition those contenders for yourself. As we have said in every "Recommended Components" introduction for the past 20 years, "Recommended Components" will not tell you what to buy any more than Consumer Reports would presume to tell you whom to marry!"—John Atkinson