Interviews

Scull: Now that you're using DSD for your recording, I assume you have to downconvert that master in order to produce a CD?

Jung: Right. We used Sony SBM Direct on a couple of projects, and on the last one I came up with my own way of doing it. There's a 24-bit PCM output in DSD systems, so you can bring that 24-bit signal into any of the current PCM studio tools.

As projects get more complicated—more musicians, more instruments, more microphones—you have to do a little tweaking here and there. You've got to try to make somewhat of a commercial recording, even in terms of maximizing the levels, for example. And there's no reason not to fix it up, because you have so much more to work with in DSD; there's so much more information than the 16-bit CD is capable of delivering. It's like reducing an eight-by-ten negative for a small magazine shot. So a little mastering is certainly in order.

Scull: What CD pressing plants do you work with?

Jung: We have an excellent relationship with Sony Disc Manufacturing in Terre Haute, Indiana and Salzburg, Austria. We also work with Sanyo in Richmond, Indiana, who make all our 24-karat gold discs. I've heard the horror stories people tell about pressing plants. Knock on wood, I haven't had those kinds of problems because I've always worked really, really close with the mastering people...

Scull: Is that one of the secrets to a good-sounding recording?

Jung: Yes, but I think the most important thing, without question—and nothing else even comes close—is to start with good music. From there, it's a real trickle-down thing. Unfortunately, particularly with pop music, the technology has gotten so sophisticated that musicians today are doing their own recording—they're trying to become recording engineers! They wind up not allocating enough of their right brain to sitting down and writing a good song. That's a very sad thing, and I hope we can kind of move through it as a passing phase.

Every musician I know has a home studio, and their noses are in the manuals. They're all trying to figure out how to punch in every other note so they can make their performances perfect. But I don't hear many good songs being written. I mean, with all the demo material that comes to us, I can't remember the last time I heard a decent tune!

Scull: An interesting insight. I understand you're a multichannel fan.

Jung: Yes. It's only recently that I've been forced to go back to two-track! As you know, DSD is a two-track format for now, so that's it. [laughs] The multichannel thing goes back to our Minnesota days. We did a fair amount of [Quadraphonic] recording back then, and we just tore our hair out trying to get multichannel onto LP. CD4 was another format, but none of them really worked, so we took a break from it. Then, about three or four years ago, things changed with the Smith Brothers—that's what I like to call the two Irish guys from DTS, Steven Smythe and Michael Smythe. They came over with a codec they wanted us to try.

Scull: Codec?

Jung: An enCOder/DECoder. [The DTS codec] used a data-compression scheme [that put 20 bits of 5.1-channel data on a conventional CD! They wanted to know what was wrong with it. [laughs] We took our 20-bit masters a track at a time—piano, drums, horn or bass—encoded and decoded them, then compared that with the originals.

Scull: And...?

Jung: Well, I'd never heard any kind of data compression sound as transparent as that. I'll have to say it wasn't totally transparent, but it was amazingly good. It was certainly a lot better than 16-bit, because we were able by then to realize our 20-bit master tapes on a disc. Suddenly, DTS was a valid release format, so we began recording multichannel projects, all of them big-band. We did three: The Glenn Miller Project, Salutes Duke Ellington, and then Big Band Potpourri. But those recordings were all kind of...different. It's not like I'd suddenly found The Answer to recording in surround sound. Like I mentioned, the Duke Ellington project was done with a Decca tree and a couple of ambience mikes. The rhythm section was miked separately, whereas Big Band Potpourri used even more microphones.

But I think that acoustic music really benefits from surround, even in the audiophile sense. I think there's a lot more information that can be put on the recording that puts you more in the venue where the music was performed. And that all goes back to my original goal. I think surround is totally valid. It's also misused a lot—putting instruments behind the listener and stuff like that. What I like to do is put the listener in the room; not just throw ambience in the back, but actually let the stage wrap outside the speakers a little bit and start toward the back.

Scull: To re-create the experience of hearing the music in the original acoustic.

Jung: Right. But I don't think a hall is necessarily the right venue for a lot of jazz music—particularly with a rhythm section, bass and drums. So many halls have a long reverb on the bottom end, giving a room bloom that just kind of lingers in the air. So I think in many cases of a club, not a teeny-weeny club maybe...

[At "teeny-weeny," Paul Jung rolls his eyes so far back in his head I swear I hear them click.—J-10]

Jung: [laughs] Well, a club of a decent size is maybe a more appropriate venue for that music. The studios we work in are more in line with a room that's acoustically balanced for that type of music. We don't have a two-and-a-half-second reverb time. Maybe we have three-quarters of a second, and we might add a little tail to that three-quarters of a second.

Scull: So, again, you're not averse to sweetening the sound?

Jung: No, not at all. I'm trying to make it what I call "right"—trying to make it closer to what's going on in the room—because often what's going on in the room and what comes out of the microphone are totally different things. I feel we should try to use any tools that are available to us that don't have side-effects that make them unworthy.

Scull: Are you planning to return to surround recording?

Jung: I'm not moving away from surround at all. As a matter of fact, we just did a project that we played a few excerpts from at HI-FI '98, in the Sony/Philips room: a choral recording that we did up in Hartford (footnote 4). Philips has an eight-channel DSD machine! There's only one of them, and it comes in 16 road cases—600-some-odd pounds. It's a Windows 95–based environment with a big array of hard drives. You need serious storage for that kind of work. Anyway, the whole idea of taking that level of resolution to surround is very appealing. I mean, you get your cake and eat it too.

Scull: You mean you eat your cake and have it too. [sighs] I just got hoisted by a reader about that. DSD appears to translate well to the 16/44.1 format?

Jung: Yes, it does, and I don't really understand why. One thing that's clear is that we're recording more information on the master tape. But there's still a real limitation with 16 bits at 44.1kHz. One of the exciting things about Sony's Super Bit Mapping Direct is its very elaborate DSP—something like a 32,000-pole noise shaper that does some pretty serious number-crunching. In the process, I think it takes a lot of that wonderful information from the pure-DSD signal and in some way puts it into the 16/44.1 CD package. In a sense, I think we're starting with a whole different animal.

Scull: How so?

Jung: Well, to me the DVD approach to high-quality audio is a little bit back-door. First of all, as you know, DVD was originally designed as a video medium. The way it works, its clocking and so on, is optimized for video, not audio. Super Audio CD started life as an audio format. I think PCM in general, though it's very good and all, still has some inherent limitations that we'll probably never get around. For me, DSD just leapfrogs PCM. And I guess it's logical to me that, if we're going to ask the consumer to buy all new hardware and software, we'd better do something pretty damn significant. I think just going to a longer word length and raising the sampling frequency is just not enough.

One of the things I wish we could get a handle on, and which I consider very significant with DSD, is that there's a kind of a relaxing, involving part of listening that I think went away with CD. Audiophiles, and rightfully so, have held on to their vinyl because there's something special about it that we can't even identify and we sure as hell can't measure. But I feel DSD moves back in that direction. It's more attractive, in my opinion. I want to listen to it.

I wish I could verbalize it better. There's something there that's really intangible. I'd love to see somebody do research in that area, and maybe some serious listening tests. It's possible that whatever our bodies don't like or sense is wrong with today's digital might not even be perceived through our ears. It could be through the skin, you know. It's just...it's very deep.

Scull: Where do you think we'll be in five years?

Jung: I have no idea. Perhaps that's not the answer you wanted to hear...

Scull: No, give it to me straight, Tom—no chaser!

Jung: [laughs] I don't know. I hope it's something beyond the whole 24/96 brouhaha, I really do. Because I just don't think that's quite enough. I want more!

---

Footnote 4: I voted this unaccompanied choral recording, which I heard in the Sony suite at the 1999 International Consumer Electronics Show, as the best sound of that show. The two-channel DSD-encoded version, played on a prototype Sony SACD player, beat the CD version into the ground. And the discrete-surround version takes the musical experience to an even higher level.—John Atkinson