BATmen: Victor Khomenko & Steve Bednarski Page 3
Khomenko: I don't really have a problem with it. If somebody wants to play around with the Khomenko-5 and the Khomenko-60, they can do it, but I doubt very much that they'll improve the sound, because most of that is determined by the basic topology—maybe 90%. Also, we've selected the components very carefully. Is there room for some improvement? Maybe very minor, but I agree with Steve that in most cases [tweaking] will be a step backward (footnote 2).
Bednarski: I would have to differentiate between a "hard" tweak vs something less invasive. We firmly believe, and have found in show environments, that the use of an isolation platform, cones—even things that people think are unusual, such as the Sakti devices—can have a positive effect on the sound. These tweaks are ones where you can listen to the effect of the device, and it's easily reversible. People should evaluate these, and they should have an open mind, because even though some of them don't seem to have the best fundamental science behind them, they often seem to work.
However, changing, let's say, the output caps, the volume pot, or the shunt resistors to another brand, is not advisable. In the case of the Khomenko-5, a prototype was set up such that we were able to plug in different capacitors. We evaluated 14 different capacitors. We had five volume-pot implementations that were evaluated.
Deutsch: Was the selection made through listening tests?
Bednarski: Absolutely. In the case of the Khomenko-60, we evaluated the difference between toroidal and EI output transformers; the toroidals made a huge difference in the listening evaluation, even though the circuit is the same.
Deutsch: Who does the listening?
Bednarski: We both listen extensively before making a final selection.
Deutsch: Do you do any blind tests?
Bednarski: No. Actually, I shouldn't say "no." We had a friend with very good hearing over when we were doing the capacitor selection, and in a blind test the three of us were able to choose the top two capacitors. However, in the blind testing, under pressure, we were not able to choose between the top two capacitors. That required longer listening evaluation. Philosophically, we're not believers in blind listening panels. When you get very close, the best approach is to listen for a longer period and then make a decision after you've had some reflection.
Deutsch: What's your view of the whole measurement vs listening issue?
Khomenko: The role of measurement in design is tremendous. You can't produce a good circuit without a very long and substantial series of measurements. The designer must be measuring as much as possible, and we do. You must know your circuit completely, and that's accomplished through measurements. Measurements are not as important in terms of published specs.
Bednarski: Our philosophy is that when we have to make a tradeoff between published specs vs sound, we'll go with what our ears tell us. A very good example of that is not having switch-selectable feedback on the Khomenko-60. We simply couldn't find an instance where we preferred the sound with feedback, even though, in bench testing, there's some benefit to feedback.
Khomenko: Also, although measurement is very important for designers, the usual measurements in typical reviews are not as meaningful for the average consumer. I think magazines like Stereophile should be reducing the number of measurements, not increasing them. For the average person, it's very difficult, even impossible, to predict sound quality on the basis of measurements. Except for some very specific measurements, like noise level in a preamp.
Deutsch: What other measurements do you think are useful to the consumer?
Footnote 2: Those inclined to make internal tweaks/modifications should keep in mind that it may void the warranty.—Robert Deutsch