Ed Meitner: Audio Maverick Page 2
Meitner: If we want to advance the state of the art, we might be stuck initially with something that looks like it is more expensive than it should be. But if electronics are handled properly and the design is elegant, the product is just going to get cheaper and cheaper.
But the design has to be simple. I don't like band-aid designs with massive amounts of chips and all sorts of stuff hanging around. To me that means it doesn't work. That's my basic philosophy: The simple circuit is the one that works, and the elaborate circuit is the one that should be rethought.
Harley: How much do we know about digital audio and how much do we have to learn?
Meitner: I wouldn't use the term "digital audio." We have a lot to learn about audio in general. Digital is a nice tool because it allows us to do things that we could never do in the analog domain, and this will aid very much in moving up the learning curve. As we go along and identify problems, it's probably not going to be very difficult to fix the problems using digital technology.
Harley: You've designed both multi-bit and 1-bit digital processors. What are the tradeoffs between the two approaches?
Meitner: A single-bit converter is still a little like cheating compared to multi-bit converters. Undoubtedly, single-bit converters will be the future of the audio business, [but] for the kind of resolution I'm looking for in a digital system, I haven't seen a 1-bit converter that does it. From a price point, it's obvious the single-bit converter has an advantage. Now that we have our DSP system [the IDAT] doing what it does, we'll have to look at some new converters and see if it is the converters or what's preceding the converters that causes the differences. But I think both types of converters will be around for some time.
Harley: The processor in front of us [the IDAT prototype laid out on the bench] has a custom input receiver, a unique digital filter, class-A DAC scheme, and other novel topologies. How do you come up with these circuit innovations?
Meitner: I have a wish list, and I work from that wish list. I identify a problem, then try to solve that problem. After you've been in the audio business for 20-some years, you know where a lot of the big problems are; those get solved first. A lot of the wish list is wishing I could make this circuit or that circuit better; you work from that. It's really not magic.
Harley: What else is on your wish list?
Meitner: Oh, I'd still like to look at output filtering [in digital processors], because I don't think filtering the analog signal is the right way to go.
One thing on my wish list is to have as much user-friendliness as possible within a digital audio system. As the system complexity grows, I can envision us ending up with a PC- or Macintosh-based audio system where the screen becomes the front panel of the system and is controlled by a trackball or mouse. Audio purchases would be made by software rather than hardware. Your system becomes an equalizer for the room, tuning different DAB [Digital Audio Broadcasting] stations, changing CD tracks, having a jukebox with a hundred CDs.
Doing all that ties neatly into a microprocessor-based package. You have the screen anywhere in the house where you want it and you access your system. To modify the frequency response, you draw a different curve on the screen. To me this would make a nice system in the near future. All of it is perfectly doable now.
After having found a way to clean up some of the fundamental problems of digital audio, like jitter, it doesn't matter what complexity we have because we can analyze what we're putting out. To me this is the promise of digital audio. It will get to the point where we can have an equalizer. People who have spent thousands of dollars on an audio system shouldn't be told they can't have an equalizer. You could have an equalizer that remembers the settings for each CD and would automatically modify the frequency response to make the CD sound the way we want. That's going to bring audio forward a step and bring back the enthusiasm that's been lost in the transition from analog to digital.
Harley: So you envision much more use of DSPsuch as in loudspeaker crossovers and room equalization?
Meitner: Absolutely. I'd like to build a negative distance generator in DSP, where I treat the time delay differently vs frequency. If I want to bring the low frequencies forward, for instance, I can bring them forward or put in a contour without penalty. Since audio is a personal thing, it should be as adjustable as possible. We played with a tiny delay on the low frequencies and what it did was wonderful. Every millisecond of delay was clearly audible.
Harley: You are unusual in that you design a wide range of productspower amplifiers, preamps, digital processors, loudspeakers...
Meitner: Don't forget FM tuners...
Harley: That's right. What's behind this diversity of design knowledge?
Meitner: It's all just basic physics. The way I look at it, you bring it all down to a common denominator, which is physics. That's what I studied and grew up with. Electronics is merely a tool.
Harley: What's your educational background?
Meitner: I have an engineering degree, and I've dabbled in physics. My relatives are involved in physics in one way or another. When I was a kid in Austria, I made all my own toys. If you were a kid growing up in Europe after the war, you couldn't go to a shop and buy things. There was a guy who ran a radio shop who was taking old radios in trade, and he didn't know what to do with them. So I took them home and tried to make something with them.
Industry was very different then. You had craftsmen and companies who were involved in the community. So for me it was no problem, when I was 12 years old, to go to work for Siemens during the summer holidays. They wouldn't treat you like someone who didn't count; they would actually put you in the interesting jobs and expect you to learn. It was a totally different way of growing up from what I see now. It is so impersonal and businesslike now. I think the education suffers as a result.
When you have to make things to enjoy things, it's a totally different story. I remember when I was a ham [radio] operator; all of a sudden Heathkit came out with ready-made ham gear, and it wasn't the same anymore. You're not a ham operator anymore if you open a panel and throw a switch. You haven't done anything.
Hands-on is essential. Theoretical physics is interesting, but the empirical part is fascinating. Then you find out electricity and mechanics are not that different, and it's easy to jump around from one to another.
Harley: You were talking last night about an experiment you did in which you recorded an electric guitar through a 10" speaker, then tried to play that recording back through another 10" speaker, and it didn't sound remotely close to the original event. What did this say to you?