Tom Jung of DMP: Making Musical Sense Part 2
Jung: Dressed to Kill, with Pino Donaggio. Star 80, with Ralph Burns. The Cotton Club, with John Barry, was a fascinating project; it went on for about a year. I was still doing records on a freelance basis, primarily jazz, and that's where I began thinking maybe this is the time to start a little jazz label. I started talking about it to musicians I was working with, people I'd see almost every day on sessions, like Warren Bernhardt, Joe Beck, Bob Mintzer, and some of the other guys we ultimately signed and went on to record. The idea was pretty attractive because we had a rapport.
Lander: When did you begin distinguishing the sonic flaws in the PCM process that have come to bother you?
Jung: I heard flaws in the very beginning, but to me they were small relative to the flaws of analog.
Lander: Did you think they'd soon be overcome?
Jung: I did, but to a large extent they weren't. The 3M machine would still hold up to just about any PCM digital recorder at that sample rate. So would my Mitsubishi machine.
Lander: In the early years of DMP, you used a Mitsubishi X-80 digital tape recorder. Why that particular machine?
Jung: It was unique in that it was designed by a Japanese engineer who happened to be an audiophile, knew about discrete class-A electronics, and used them in the X-80. He was also aware that there were inherent problems in the PCM format, so he compensated for them, correcting phase errors created on the record side in playback. To a large extent, he was successful. We'd record and play back in the studio, and musicians always liked the way the X-80 sounded. But it had the god-awful sample rate of 50.4kHz. By the time I got to the mastering stage and had to convert that 50.4kHz to 44.1kHz in the digital domain, the mathematical formula led to significant losses.
Two SACDs in our catalog, Flim and the BB's' Tricycle and Jay Leonhart's Salamander Pie, were made directly from Mitsubishi masters. When I made the SACD versions and couldn't go from 50.4kHz to 1-bit DSD in the digital domain, I remembered that machine had always sounded good playing tapes that it recorded. So I used that same machine and the original master tapes, the ones that actually had razor-blade splices.
Lander: Razor-blade splices?
Jung: Yeah, the X-80 tapes were razor-blade-spliceable. You had to wear white gloves so the oil from your fingers didn't contaminate the tape and cause dropouts. I fed the balanced XLR line-level output into a Meitner A/D converter and recorded it to DSD. If you take those SACDs, put them in a really good player, and compare them to the original X-80 masters, most people can't hear the difference. That's one of the nice things about SACD. It's really valid for record companies to go back to their original analog master tapes that got so murdered by the CD format and remaster them in SACD.
Lander: You first heard DSD very early on, at a by-invitation-only listening session at a Sony Music studio in New York City with three other recording engineers, Bob Ludwig, Bruce Swedien, and Michael Bishop from Telarc. Tell us about it.
Jung: They had a jazz quartet set up in the studio, and David Smith, who is Sony's technical guru, had built a passive four-way switch that would allow us to listen to and switch between the live output of the mixing desk, conventional 16-bit digital, the latest and greatest 20-bit [PCM] converter, and finally to DSD. The DSD was so much closer to the live signal than either PCM converter that it was obvious this was a direction I really wanted to pursue.
Lander: Do you feel DSD solves the problems inherent in PCM?
Jung: It does. I'm finding after all these years that, when I listen to an SACD, especially things that I record, where before I'd probably only listen from a quality perspective and not really be able to get into the music that much, [now] I'll listen to the whole thing. It draws you in, much like analog did, but without the analog problems. I got so I really, really didn't like the way PCM sounded—to the point that I even considered trying some analog recording again—but I caught myself before I did that. [laughs]
As time went on, I felt there was something to this theory that PCM is stressful. I don't think we have a handle on it. I don't think we can quantify it or even verbalize it, but I do think there's something going on. Back in the pre-digital days, the people I worked with at Sound 80 would have great listening parties. We would bring home big studio monitors and big amplifiers and play music really loud. That kind of went away when the CD came out, because it wasn't that much fun to listen any more.
Lander: You've said that PCM artifacts particularly bother you on big-band music, which is something you've listened to live and relished since you were about 10 years old. Can you be more specific?
Jung: There are sonorities in the brass that are harmonically very complex, and PCM digital can't reproduce them. The instrument that probably points it out best is a muted trumpet, which has a lot of energy beyond 20kHz. With PCM, the brick-wall filter at approximately 20kHz has so much phase shift that it manifests itself down at 2kHz and 3kHz, where the ear is very sensitive. In essence, what's happening is that the corrupted harmonics change the overall sound to the point where it's ugly. It can be almost painful. It's not just the playback that's corrupt. It's the whole process—record and playback.
Lander: You've traditionally tended to substitute consumer equipment for pro gear. Has that practice persisted?
Jung: It has. I've always been a critic of what we use professionally. It can be very frustrating to know that what some audiophiles use for playback has a higher level of accuracy than we ever had on the record side. So that's something I always looked to the high-end community for. The most obvious areas are speakers and amplifiers.
Lander: You've used Duntech and Thiel speakers, as well class-A amplifiers from Krell and Pass. Haven't you worked with the Wadia A/D converter as well?
Jung: That's when I started using more than one converter. I began using eight channels of converter to do a project, converting the output of each microphone to digital right then and there. Quality-wise, that was kind of a sideways step. As I look back on it, there were some advantages, but maybe the advantages didn't justify the means.
Lander: You've said that since the big bang—by that I mean the day the 3M crew showed up at your studio in Minneapolis with their new tape recorder—every improvement in digital prior to DSD was merely incremental.
Jung: DSD was a big improvement over the various stages of PCM. I mean a big improvement. I passionately went after David Kawakami, from Sony, to get me one of those machines so I could do a project on it. He did, and I made two or three records, and as I worked with it, I'd start to think, well, this piano doesn't sound quite right to me. So one day, when I was really frustrated with it, I looked inside, and wouldn't you know? It had NE5534 op-amps in it. That's an amp that measures really well and sounds like shit—that's the best word I can think of to describe it. A lot of pro audio gear has these horrible things. You can buy them in quantity for 17 cents, so the bean counters love them. It blurs things. It just can't track the waveform. A midrange or high-frequency piano chord can hurt your ear.
Lander: How did you deal with that situation?
Jung: I went to Ed Meitner. At the time, he had a company called Museatex, and I was using his Bi-DAT D/A converters. I liked the way he thought and designed. He has a real handle on analog design as well as digital, which is pretty unique. I got permission to send one of the Sony DSD recorders up to him, and he looked at the analog circuitry and said, "Oh, we can do a lot better than that." He built a D/A converter in a Bud box and sent it down to me, and the recordings I had made sounded much better playing back through it. I said, "Well, that's good. Now we need to do the A/D side so we can make the recordings better." He said, "That's no problem." He did it at night, after his day job. I used the A/D converter, and there again, everything opened up and got that much better. It became obvious that these Meitner converters sounded better than anything else, so Sony started buying converters from Ed [who now has a company called EMM Labs, in Calgary, Canada].
Lander: Looking ahead at the foreseeable future of the music business, which clearly includes downloaded files, where in the jigsaw puzzle do you think SACD fits?
Jung: The majority of people out there feel the CD is just fine. They put up with MP3 and think it's okay. So the bar is pretty low for Joe Consumer. But I think the hybrid disc makes a lot of sense. I have to think that, at some point, people are going to realize there is better quality available. I do think some people are going to buy an SACD that they can play in their car and at home. If they have a home theater, they can hear it in surround. There's so much data on that disc—and it's not copyable—that people will spend the money to buy it. If you want to build a great-sounding system to play it on, you'll be able to realize tremendous quality and tremendous entertainment value.