BATmen: Victor Khomenko & Steve Bednarski Page 4
I'm hesitant to suggest particular measurements, because there are so many techniques to measure each parameter. For example, you can measure output impedance under different conditions with different techniques. Sometimes those techniques produce different results.
Bednarski: I think I understand what Stereophile readers look for from the measurements: a sense of any technical flaws inherent in the design. For example, you see some products that have a very difficult time reproducing a semblance of a squarewave. You would tend to think that an amplifier should do that credibly. If it does not, then you begin to question the fundamental design.
Deutsch: I'll ask the other perennial audiophile question: what's your take on the digital vs analog issue? Do you still listen to LPs?
Khomenko: I have a turntable in my system, and I have a very good record collection, but most of my listening lately has been to CD, because I need a repeatable reference. However, I have strong opinions about the general issue: I think analog is the way of the future. Not analog in the form that we have today, which is a continuation of decades-old technology. But provision of ultimate performance is not going to be digital.
My vision is new media for analog recording; they may be optical or magnetic. There are some things that are fundamentally wrong with the digital approach to sound recording. Increasing the number of bits and the sampling rate is not going to get us very far. People are saying that digital with 20 or 24 bits will finally equal the quality of analog. Analog LP as we have it today, maybe. But I think ultimate performance is only achievable in analog.
Digital is an artificial form of signal processing. I like the analogy where you take a live snake and chop it into 10 pieces, and you have a dead snake that's been chopped into 10 pieces. You can now chop it into a thousand pieces, and you still end up with a dead snake, but it's now more finely chopped. The life is gone. In the same way, once you convert the signal from analog into digital, you destroy the signal to some degree, and you can never recover it fully.
Digital came along not as a good way of handling the signal, but as a convenient way of handling the signal. I've done a lot of work with digital signal processing. I know why it's done; there are many reasons, but performance is not one of them.
Bednarski: I'm also in the analog camp, but I find myself—partly because of the convenience factor—increasingly listening to digital, and digital has improved substantially in the last few years. I do think we're at the point of diminishing returns in current digital. What the new DVD format should do, with the right implementation, is to rejuvenate the marketplace. It raises the stakes.
Deutsch: So the future could be good?
Khomenko: The future could be tremendously good. We're not even close to approaching the end of what's possible in audio. There is so much that can still be done. One way of improving the performance of audio systems today is to change from a voltage to a current interface. High-voltage/low-current circuits are not the right way to go. In instrumentation we routinely achieve 108 dynamic range, using a current rather than a voltage interface between components. The possibilities are endless. Unfortunately, we have to break the audio industry practice of using a voltage interface.
Deutsch: What would a current interface be like?
Khomenko: In simple terms, you would have an input receiver with zero impedance, as opposed to the high impedance you have now, so you invert your circuits. What we've tried to do with the Khomenko-5 and the Khomenko-60 is to go toward having low-voltage/high-current circuits, but striking a middle ground, to ensure compatibility with other components.
Deutsch: How is a current interface better than a voltage interface?
Khomenko: Much of it has to do with the limitations of a voltage interface, because of components like resistors. For example, in a normal environment, you can't exceed about 10V. It would be impractical to design circuits with 1000V. At the bottom of the dynamic range, performance is limited by noise levels in resistors and semiconductors, so your dynamic range is restricted. In the current realm you can easily go from the femtoamp (footnote 3) to the amp region. This gives you several orders of magnitude improvement in dynamic range. The current interface is also much more robust, less sensitive to interference from RF and magnetic fields.
Deutsch: What products are in the future for Balanced Audio Technology?
Khomenko: A lot of people think of us as a tube company, but I'm certainly not a "tube designer." I'm an electronics designer. I feel equally comfortable with semiconductors. We don't want to lock ourselves into a narrow niche. We have a solid-state preamp and a power amp now in prototype form.
Bednarski: Our next product will be the Khomenko-3 line-stage preamp, offering a large measure of the performance of the $3995 Khomenko-5 for $2495.
Deutsch: Any plans for digital?
Khomenko: We have wonderful ideas for digital, and someday we may have a very good product based on those ideas. It's an important part of the market, so we're giving it serious consideration. We're also thinking of bringing out a phono stage, based on the phono stage of the original prototype. We have more plans than we have time!
Footnote 3: Femtoamp= 10-15 amps or 0.000000000000001 amp.—Robert Deutsch