Hoisted on your own petard?

After completing a PhD in electrical engineering at Imperial College London, Floyd E. Toole joined Canada’s National Research Council (NRC), where he would stay for more than 26 years doing audio-related research. He continued his research at Harman International after leaving the NRC in 1991. When Toole left Harman in 2007 (footnote 1), Harman kept the work up under NRC alum Sean Olive (footnote 2)—which fact surely has much to do with the excellence of their current loudspeaker lineup.


The importance of Toole’s project is hard to overstate. His goal was to provide a scientifically rigorous foundation that could inform choices made by audio designers, especially designers of loudspeakers. He succeeded. Drawing on his own research and the research of others, he established a template for what I call the “classical” loudspeaker: flat frequency response; excellent, well-controlled off-axis behavior; nonresonant cabinet, etc.


Toole’s main technique was to carry out blind listening tests over many years with many subjects and analyze their preferences statistically. He learned that when it comes to loudspeakers, people mostly like the same things. As they get more training as listeners, they still like the same things; they just become more certain about their preferences.


The fact that there tends to be agreement between trained and untrained listeners adds depth to the research—it’s much more than a mere survey of broad consumer preference—yet the work does not attempt to capture individual variation. It averages over preference. That’s what needed to be done—really the only way forward, especially at the time the work was done and maybe still. And yet, while a loudspeaker designed to the classical Toole template will sound the best to the largest number of people, some people will prefer something different. That’s not a defect of the research—it is what it is—but it is important to note what it isn’t.


Some of the most passionate, deeply committed designers in the audio world refuse to buy in to the classical model. Their creations may lack the broad appeal of a classical loudspeaker, but many of them are very good at certain things and appeal to the significant subset of audiophiles who value those virtues over others. As the late Art Dudley wrote in one of his last columns, “From its acoustical beginnings, when two incompatible forms of physical media—Edison’s cylinders and Berliner’s flat discs—slugged it out for primacy, domestic audio has attracted an almost incalculable number of iconoclasts, heretics, mavericks, nonconformists, lone wolves, enfants terrible, and hidebound kooks. Because the above are among my favorite people, I don’t have much of a problem with that state of affairs.” (footnote 3)


Nor do I—I like the fact that the world is rich and varied. There is no single path forward but, rather, many paths leading in many directions and ending at many vistas (footnote 4). You may not prefer the view there over another view, or the sonic perspective, but someone does, and if your mind and ears are open, you can enjoy it. Do we want to live in a world where everything sounds the same? I don’t.


It’s disheartening, then, when speakers (and other components) that so obviously do not aspire to classical behavior continue to be judged by classical standards. “That speaker doesn’t have a flat frequency response!” shouts an all-caps critic on some online forum, about a speaker whose designer never aspired to a flat response. “This speaker has a resonant cabinet!” exclaims another, about a speaker with a cabinet that’s tuned to vibrate in a particular way. “That designer is inept!” writes a third, about an engineer who has sold tens of thousands of speakers—perhaps more—and won awards.


I admire Toole’s work, but I do not admire conformists who insist, often with insufficient self-examination, that everything be judged by the same narrow criteria. There may be a single best way to roast a chicken, but I’m glad different chefs use different recipes. We at Stereophile encounter this problem ourselves sometimes, especially in measuring but also in listening. A loudspeaker (for example) that is intended to have a certain sound should not, I feel, be panned because it sounds different from what the reviewer expects or prefers. The reviewer’s job is to characterize, not to condemn. As Art often advised reviewers, “Tell us what it sounds like!” If a loudspeaker is a sonic outlier, we must tell our readers that, but we have no responsibility to condemn it unless its sin is grave indeed.


And when a component is measured, what should it be compared to? Here, again, the reviewer’s job is inform readers when a component deviates from what is classically thought of as excellent behavior, especially when that deviation is likely to be audible (although it is often difficult to know). We must of course tell them when that flaw is especially egregious and without obvious justification.


But there’s a difference between ineptitude and nonconformity. Confusing the latter with the former makes the world a more arid, less-rich place. One can pursue excellence without excluding passionate outliers.


It’s especially disheartening when narrow-minded online critics use one aspect of our coverage—our measurements—to attack the other side: our subjective judgments. We’re providing a complete picture; the two halves make a whole. You don’t get that from our competition.


Broaden your mind. Seek perspective. Look at the big picture.—Jim Austin

Footnote 1: Today, Toole lives in L.A. and heads his own acoustics and psychoacoustics consulting firm.


Footnote 2: A report by Kalman Rubinson on taking part in listening tests at Harman’s facility can be found here.—John Atkinson


Footnote 3: Art addressed this topic in Listening #207—his fourth-to-last column.


Footnote 4: Which assuredly does not mean that forward progress isn’t possible.

Click Here: custom plastic molding