Deeper Meanings Letters part 3
Editor: Not surprisingly, Robert Harley misses the point [in "Deeper meanings"] of that 1987 Stereo Review amplifier comparison. Golden Ears like his were asked to set up the conditions and their recommendations were followed! Is he clever enough to understand the implications?
No, he clearly is not a charlatan. He's far worse—utterly sincere!—Mike Silverton, Brooklyn, NY
Translational errors & fudge boxes
Editor: As an AES member and regular reader of Stereophile, I took interest in Robert Harley's commentary regarding the Golden Ear/Meter Reader debate. I regret that the conversation has become so polarized that it encourages both sides to take positions more extreme than either should be comfortable with.
While I consider myself one of those fiendish "Meter Readers," I would like to offer a slightly different twist on the definition, namely, "If a sonic difference can be reliably heard (absent of perceptual distortions), there will exist a physical manifestation of that difference that can be measured, and therefore controlled." Perceptual biases caused by visual or other stimuli should be researched and used as understood.
By this I don't suggest that we understand everything we hear, only that if it's "real" and "repeatable," we can meter it. Whenever you deal with a phenomenon that occurs in more dimensions than the storage medium, translational errors are bound to occur.
Some very successful "systems" vary from linearity while sounding more "truthful" than their more accurate brethren. While it's tempting for every component manufacturer to "fudge" his part of the chain to make his reference system sound better, it's too easy for us to end up with more "fudge" than we need when mixing components from different manufacturers. A more rational approach is to make the individual components linear and put the "fudge" in separate black boxes. When we get the "fudge" well enough figured out, we can encode it onto the recording.—John H. Roberts, Hickory, MS
Experiments & placebos
Editor: One part of my curriculum as a first-year medical student was a course in Biostats. In that class, I learned the importance of a well-conducted experimental test. For example, my teacher told us of a surgical procedure that was performed on patients to prevent heart attacks. A test was done to see if it was effective and it seemed that many patients did feel better and had less occurrence of heart attacks. However, when a more thorough and complete test under double-blind conditions was conducted, the result was statistically significant that there was no improvement with the surgical procedure. Apparently, the placebo effect was the effective agent that caused the patients to believe that it was effective.
I am telling you this because I have recently gained interest in the world of audio and I too want what is good. However, after reading Robert Harley's article in July, I had to add certain remarks.
First of all, I do believe that how good [a piece of] equipment sounds cannot be determined by any objective tests or measurements. Bob is right in that what makes something good is how well that component can involve the listener with the music. However, I do believe that a double-blind test is necessary or else the reviewer will not know if the equipment in question is better or it's simply a placebo effect. Even Bob says "Who really believes that a pair of Futterman OTL tube amplifiers, a Mark Levinson, and a Japanese receiver are sonically identical?" This is exactly what the placebo effect is!
I am not saying that an expensive tube amplifier sounds exactly like that of a cheap Japanese import, but one cannot have the mentality that just because something looks better, costs more, and has a better reputation as hi-fi, that it is actually better-sounding than a Sanyo receiver, for example. The better-built tube amp probably does sound better than a Sanyo receiver, but we have to prove it, and prove it properly with double-blind [tests]. This is even more significant when comparing two [pieces of] equipment that are in the Class A category, for example.
I believe that, beyond a certain point, most equipment sounds alike, if not identical. That is, a $4000 amp is going to sound very much like another $4000 amp (unless the amplifiers are obviously different), and these minute differences are probably not detectable with a double-blind test. Without the double-blind, the placebo effect can carry the reviewer's imagination and exaggeration to the limit of the sky. This is one reason why I believe there are so many letters that disagree with the reviewer articles. Take Armor All and green paint, for example. Many letters say that there is no difference, even though Sam Tellig has said there is a tremendous difference.
I also disagree with Bob in his view that double-blind testing leads us away from reality. Judging equipment is a subjective, emotional experience, but one can still have this with double-blind testing. Instead of having the equipment switched back and forth, one can listen to amp A for as long as he or she likes, then switch to amp B and, again, listen to this for whatever length of time. The setup and the switching can be done by another person so that the reviewer would not know the true identity of the amps. Now this would be a good double-blind, while the reviewer could audition the equipment as he or she normally would; ie, by listening.
Remember, without a double-blind, you and I will probably go out and have surgery done even though it really doesn't work, just as so many people are made to believe that amp A sounds much, much better than amp B, even though there might not be a true difference.—William W. Jou, Walnut, CA