The Blind leading the Deaf Letter: Why Amplifiers Sound the Same
Editor: In a recent conversation with my astrologist, I asked her why I have gotten so involved in the controversy over amplifier testing and design. She suggested that in a previous lifetime I was an evil radio repairman (which explains my use of tube circuits) who has come back in this lifetime to repent for my past sins. Judging by the number of phone calls I have received in the last two months from crazed audiophiles, I must have been very, very evil. The only way to pay my dues is to comment on recent Stereo Review and Stereophile amplifier tests.
Why all Stereo Review's Reviews Sound the Same Test: There are so many flaws in this test I could fill this magazine with them, but I should first like to comment on an emotional issue. I believe David Clark has compromised his integrity by being dishonest with me. When he requested NYAL OTL-1 amplifiers for the Stereo Review test (SR, January 1987), I specifically told him not to use the 4-ohm Magnepan speakers for critical listening tests as it was an improper load for this design. I suggested the MartinLogan CLS speaker instead. When I asked him what kind of testing he was conducting, he inferred that they were very casual. He did not describe the test or the procedure to me. Does this behavior taint the results of the test? Does Stereo Review's previous record of critical testing taint this test? Would you speak a little louder because I am hard of hearing?
Ironically, not long ago, I suggested to Bob Carver that he use his amplifier null techniques to demonstrate to Stereo Review's editors that different amplifiers (with different transfer functions) will produce a measurable difference when driving a given speaker. This difference will vary with each speaker. Of course, the real question is: does the listening panel have sufficient listening acuity to detect the electrical differences that we already know exist?
In his test procedure, David failed in four ways. (1) He failed to use the null test technique and thereby a priori establish the electrical differences in the amplifier's performance with the given speaker. (2) He didn't measure or control the panel's listening acuity. (If David won't trust my claims about my amplifier, why should I trust his claims about his panel's listening skills?) (3) He failed to describe the acoustical character of the roombecause a room is one of the best filters known to man. Why wasn't a good old fashioned reliable FFT analysis of the room done to make sure it wouldn't mask important sound characteristics? (4) Last but not least, he failed to use an appropriate speaker for the test.
David says that he is a serious student of audio engineering. Why didn't he mention the research by Martin Colloms and a series of articles that appeared recently in the Audio Engineering Society Journal that refute his position and question his methodology? Dearest Dave: Half a test is no test. Was David in control of the article? Is David Julian's heir apparent?
Stereo Review Shooting Themselves in the Foot Test: Ladies and gentlemen, lend me your earsI bet you a zillion dollars that I can prove that David Clark's test procedure also proves that all amplifiers over time sound the same. Yes Siryour 1959 Fisher sounds exactly the same as your 1987 Fisher. Your 1960 Sony will sound exactly the same as your 1987 Sony. So trust medo not buy any new audio equipment. You are wasting your money. All of those ads that claim that everyone's new products are betterdon't believe them. In fact, all those engineers who are supposed to be working in laboratories in America, Europe, and Japan developing new and better equipment: (1) don't know what they are doing; (2) are telling you lies; (3) don't really exist. The implication of the David Clark test is quite simple: everyone who has been in involved in advancing amplifier design over the last 25 yearsincluding designers, the press, retailers, and sensitive music loversis a bunch of charlatans. At least there are two honest men in the audio business: Julian and David.
David Clark has also assured me that differences in the perceived sound of amplifiers are due not to differences in circuit design but to the different emotional appeal of brands: cosmetics, brandname, construction techniques. You might say that David believes that knobs make the big difference. I am convinced, so, in order to justify the cost of a Futterman OTL-1 amplifier, we have instituted a policy that will once again reestablish our reputation as faithful servants of our customer's needs: Included in the purchase price of an OTL-1 ($12,000) is a custom installation by one our VTNTs (Voluptuous Teenage Nymphomaniacal Technicians), who will live with you for 90 days to make sure you are happy with your music system (or whatever else needs servicing).
We are also prepared to send to David, for double-blind testing, one of our VTNTs and an affectionate German Shepherd to see if he can detect any differences in their performance, under controlled conditions.
"And what you do not know is the only thing that you know."TS Eliot, East Coker (1943)
We do not yet have a mathematical or electrical modeling technique of an amplifier circuit that will give predictably good results, though many have been suggested. Oddly, over the last 25 years, those designs proven to be the most electrically perfect usually sounded very mediocre. Does anyone remember TIM-perfect amps? The Hafler test is another variation on a test which has been used for the last 25 years to demonstrate that those amplifiers that test best under condition "X" are more perfect. Why, then, do amplifiers less perfect under test condition "X" sound better?
I would also like to suggest that, while our intellectual understanding of how we make judgments about "sound and music" quality is still very primitive, our hearing isn't. This implies that we still do not know which combination of distortions is acceptable and which a turnoff. The gestalt of listening to music is far more complex than that of controlled listening tests. Recent research on the physiology of listening to music indicates that, when deeply involved with music, we go into a trance-like state because our brain secretes a "natural narcotic." How does this altered state affect our sensitivity to a given type of distortion, which may have been acceptable before we reached this state?
As Richard Heyser has pointed out, the only way to describe an amplifier is relative to a specific speaker. Question: Why don't we design more speakers with dedicated amplifiers, which is the better way to do things? Answer: Boys get cranky if they don't own lots of toys.
Audiophiles already know that one man's Twinkies is another man's cat food. Girls, let's be honest, while we would like you all to think that we are serious scientists in total command of our art form, designing audio gear is still a hodgepodge of primitive disciplines. Never take our technical claims too seriously because the smarter we get, the less we know. On the other hand, we are doing pretty good for a bunch of dummies because our inventive impulse and passions compensate for the lack of megabuck laboratories. Isn't that what high-end audio is all about? And besides, everyone knows that my Black Hole Amplifier is the only more-than-perfect amp.
Some Questions Dostoevsky Might Ask if He Were an Audiophile: What if you could buy a perfect amplifierthe last one you would ever have to own? Would you be happy? What if the perfect amplifier existed and made all your records, FM, and CDs sound terrible? Would you be happy? Why are audiophiles in such pain? What are they searching for?
I suggest that audiophiles are music lovers who seek to explore an altered state of musical consciousness, a state that can only be entered with a precarious commingling of emotional commitment, artistic courage, and a musical stimulus of a high order. The intensity of feelings, knowledge, and decisions we make in this state are quite different from "normal" listening experiences.
The only perfect musical amplifier is you! For those who understand what I said, no further explanation is necessary. For those who do not understand, no further explanation is possible.
In closing, I once again thank the editors for permitting me the opportunity to respond, and further request that John Atkinson not request a VTNT for an extended review. It interferes with listening to music. Trust me, I know.
I remain, bound to the Karmic wheel of audio.Harvey Rosenberg