Audio McCarthyism Page 2
A few people offered comments, but the audience response was primarily hostile to Dugan and the methodology. Boston high-end retailer Clark Johnsen, in town to deliver a paper on the audibility of absolute polarity, said, "They both sound terrible." Another audience member criticized the fact that the microphones were picking up the signal from the Thiels and returning it through the PA. The coloration and delay caused by the live microphones didn't seem to bother Dugan; he responded "Sure, there's garbage being picked up. Let's consider it part of the reverberation of the room."
This from the man the AES entrusted to conduct these official listening tests? This from a representative of the audio engineering establishment whose opinions may become the basis for criminal prosecution of high-end audio retailers? (You read correctly—keep going.)
The open (sighted) listening was conducted a second time, this time with the microphones turned off. Again, some in the audience were critical of the procedure. Dugan now apologized for the setup ("it is what it is") and the music ("it was chosen at random"), as though these were valid excuses. Remember, this wasn't some haphazard event thrown together on the spur of the moment; this was a listening test officially sanctioned by the AES, performed at an international AES Convention, and scheduled months in advance. Further, its stated purpose was, according to Dugan, to "get listening test data that we can analyze later," implying that the data collected under these conditions could be used as a basis for drawing conclusions.
Despite the absurd conditions and criticism from a faction within the audience, the official double-blind listening tests began. But first, Dugan attempted to defend rapid-switching ABX testing with this analogy: "An immediate comparison gives you more sensitivity to subtle differences. If you were going to find out if a piece of cloth on this side of the room and that side of the room were the same or a very similar shade, you would have a difficult time if you took one home for a week and looked at it, and then took the other one home for a week and looked at it. If you brought them together in front of you, you could tell whether one pink was slightly pinker than the other pink. Putting objects you wish to compare next to each other is the logical and intuitive way to make a comparison between things that differ subtly."
Does Dugan equate music with a flat, monochromatic canvas? Is the musical experience analogous to looking at a single-colored piece of cloth in Dugan's mind? If forced to use a visual analogy, I would liken comparing loudspeaker cables to deciding which of two beautiful paintings to purchase for your home. Both paintings contain wealths of subtleties not immediately apparent at a glance. Both create different and complex reactions within us and affect us differently on an emotional level. None of the paintings' qualities that make them worth having are revealed by quick side-by-side comparisons. His analogy leads me to think that perhaps Dugan should be more pitied than scorned. Those of us for whom music listening is an ineffable, deeply moving experience would never equate the experience with looking at a chunk of solid-colored cloth.
David Clark was then called on to explain the ABX double-blind listening test protocols: switch position "A," as indicated by an LED on the ABX box, was always the MIT cable. Switch position "B" was always the generic 12/2 cable. Position "X" was either the MIT or 12/2. Clark instructed the audience to "make the guess as to whether X is A or B." (emphasis added).
To further acquaint the participants with the ABX procedure, the MIT cable was compared over two trials to 30-gauge wire wrap wire. (Thirty-gauge is not much bigger around than a human hair.) The difference between MIT and 30-gauge was gross and audible—as Dugan expected. Because almost everyone could easily hear the difference (footnote 6), Dugan felt this to validate the ABX testing procedure and affirm the suitability of the listening conditions.
During these first two demonstration trials, someone in the audience argued that the switching was made on transitions from one musical passage to another, different one. Clark said he wouldn't "repeat the same segment [with each cable and "X"] because it will take too much time." Another audience member suggested a different methodology (I don't remember the suggestion, and his voice was inaudible on the tape). Clark agreed that "That would be a very good way of doing it. It's a good idea but we'll have to proceed with what we can do easily. We've gone to a lot of work here." One can only imagine what the listening conditions and test methodology would have been like if they hadn't invested "a lot of work" in the procedure (footnote 7).
Footnote 6: When asked whether they did or did not hear the difference between these two cables (a considerable loudness change due to the series resistance of the wire-wrap cable), several members of the audience indicated that they hadn't heard any difference. Remember that the audience consisted almost entirely of audio professionals.—JA
Footnote 7: David Clark was awarded a Fellowship of the Audio Engineering Society for his work on listening tests and the development of the ABX comparator. Note that Mr. Clark was on the panel and actually conducted the double-blind listening, yet expressed no qualms about the listening conditions or open microphones (footnote 8).
Footnote 8: I have now taken part in four formal listening tests in which Mr. Clark has taken an organizational role: cables and amplifiers at the 1988 Los Angeles AES Convention (see Stereophile Vol.12 No.1, January 1989, pp.63-72); surround-sound processors at the 1990 LA AES Convention (Vol.14 No.1, January 1991, p.73); and now this 1991 cables test. This demonstrated lack of experimental rigor would be both ridiculous and irrelevant if it were not for the fact that he would like his null results to be taken as definitive (see "10 Years of ABX Testing," David Clark, AES Preprint No.3167 K-1, October 1991).—JA