Interviews

Harley: Theoretically, the digital recording medium has no effect on the sound quality provided there are no uncorrected errors. Do you believe different digital tape transports have different sonic characteristics?

Sax: They do. I don't know why. I don't use Sony decks. I can tell you that any Sony video deck will kill the sound. I've discussed it with Sony and they were about as interested in hearing it as you are in buying New York City for three trillion dollars. They said, "We'll do something about it—go away." I did and they didn't. But it's a major, major factor in degradation. Major factor. It's interesting to take A/D and D/A without tape storage and then put the video recorder in the signal path. It doesn't get better. Then when you put a Sony deck in the loop, you don't recognize it coming out. I've demonstrated it—to a non-caring world. I was hoping somebody would find it and fix it. You know from your CD mastering work what video tapes are like. Why don't we have disc storage for master tapes?

Harley: There is a sigh of relief sometimes after a master tape has been cut and the tape is no longer needed. Many tapes are right on the edge of working and not working.

Sax: If computer storage were on U-Matic, the country would be out of business. Somebody's bank account would suddenly be short $2 million.

Harley: Do you think DAT is an improvement?

Sax: Not as it stands now, no. But it has the potential to do whatever you want. The weakest part of DAT is the storage. I can make a DAT tape that sounds excellent—not using their converters, but my converters. But if you play it 10 times it doesn't sound the same. Look at the density of packing. The 1630 video storage is one thing I talked about in Japan because they know it is not good. I can talk about that without apology and not get into anyone's bailiwick, like why they use six op-amps when two will work. You don't step on any toes talking about the storage. They just don't like to talk about it. Sony sewed up the world. You can send a 1630 tape to any plant in the world. Why rock the boat?

Most record companies will not use "master" tapes for CDs. They run it through their digital faders, clean up the fades, put digital black in, and they're thrilled. But the fact remains that it doesn't sound like the original tape that was made. The consumer has no idea, though. A couple of cuts on our new Clair Marlo CD [Sheffield Lab CD-29] sound about as least CD as anything I know—with resolution and echo. I'm very proud of that. The eighth and ninth tracks on that CD don't sound like other CDs. But that tape was made by me, not on a Sony deck. It was transferred directly from the original tape by Disctronics. It wasn't a copy. Nobody got into it. And it sounds quite excellent. It has resolution. Though it doesn't sound as good as the master tape, which is analog, it sounds right to me.

Harley: You have the opportunity to hear the real musical event and immediately afterward, hear both analog and digital playback of that music. What differences do you hear between the two?

Sax: The digital loses nuance. It doesn't flow as well. I don't know how you would quantify that. It has less echo. Transients aren't as good in the high end. If I had a terrific, linear, A/D it would sound closer. I've gone as far as I can with what's available in one sense. But we're a long way from needing to change the format. I think the oversampling technique is meaningful to get rid of the phase shift and some other problems. The Apogee filters (footnote 6) don't have time delay but they lose resolution. They go through a lot to correct the phase, but they don't have the resolution. To me, it's a bad tradeoff. I'd rather have the resolution and a little bit of artifact than have the artifact removed and a little bit of opaqueness to the sound. That is preferable to something more clear that you can see through. It's in its infancy.

The LP is probably dead except for maybe the tweakos. Harmonia Mundi, in their American productions, record everything analog and the engineer, with whatever converters he can get, transfers to digital. I don't know if they are losing money or not. But the analog tape comes to me and I make an LP of it and the LP is generally believed to sound superior to the CD.

Harley: How do you approach a project differently when making a CD master tape and cutting an LP master?

Sax: Many of the albums done today are made by three different producers, four different engineers, some tracks were done in LA, some in Nashville, and some in London. I make changes that put them in the same ballpark. One is low level, one is high level, one is dark, one is thin. I'll find a medium ground and tweak them into this area. That we do for the music, not for the storage medium. It doesn't matter if it's for cassette, CD, or LP. This intro has got to come up, that song is lumpy and bottom-heavy and has to be trimmed. There are certain things you do to record the disc itself. Maybe a sibilant "s" can pass as it is onto the CD, but would be a problem on the LP—it might be de-essed on the LP. That's about the main thing. But if it's a Harmonia Mundi recording of a Mozart horn concerto, you do nothing to the sound except to get it as cleanly as possible from the tape to the disc.

Harley: Is everything in the signal path tube in your mastering system?

Sax: My tape machines are tube, my cutters are tube, my limiters are tube. My EQ, which is passive, has one class-A solid-state recovery amp. But when I do transfers for Harmonia Mundi, that's not in the circuit at all: I'm not EQ'ing. It comes off the tape and directly into the cutters. I may bring up one selection that they thought was recorded too low, but that's it. It couldn't be done more simply. It's surprising because these recordings, which are quite good (they're done by Peter McGrath), are just better on the LP—with all of its faults. It's much closer to what's on that tape than what has been able to be done digitally. The American head of Harmonia Mundi, Robina Young, still wants to hear the music the best way, and that's not digital. The parent company might disagree with her, but she has autonomy over what she does.

Harley: What tape format do most of your clients send you for disc mastering and CD master-tape preparation?

Sax: These are not budget acts. They can mix to whatever format sounds the best. Cost is not a consideration. 30ips half-inch analog tape with no noise reduction is the preferred format today. There are sonic disadvantages to half-inch over quarter-inch when you get down to the nitty-gritty. But the advantage is quite simple: for the same recorded level, 1/2" is 5dB quieter than 1/4". On an already quiet format, 5dB is quite meaningful. You won't pick up a CD made from half-inch analog and complain about noise. The thing that interests me is that the half-inch, 30ips format that really started to get going about six years ago has only returned the analog tape format to what it started out as in 1948. In '48 the tape speed was 30ips, mono on quarter-inch. There are people who say, "Look, I know digital is not quite there, but look—it's only 10 years old. Look at how much time analog has had to get itself together." And I say if I could get my hands on master tapes made in 1950 I could shock your pants off. The analog format started exactly where it is today.

Harley: And they used tubes back then.

Sax: They used tubes. They started off with 30ips tape speed and quarter-inch track width and a quieter curve than they use today. That's how it came out of the box. When we go back to the Lab, I'll show you the 23rd machine Ampex ever made, delivered to Radio Recorders in October of 1948. So the truth of the matter is that the analog tape recorder was running at its prime when it was introduced in this country. It comes from the Magnetophon that the Germans developed in the late '30s and was brought over here in '46-47. Bing Crosby and a few people put up a bunch of money and it became Ampex. Ampex doesn't make any audio equipment anymore, but they were the world standard for many years. A classic example of an American company checking out of a whole industry.

There are those who feel they can get a better CD coming from an analog tape than they can by putting it live on the digital. I know some people who record the rhythm tracks digitally and record everything else on the analog machine. What is interesting is that the major classical recording companies—DG, Philips—decided en masse that they were going to record only in digital. It is only the American independent companies who, with the artist and engineers, really make the decision on their own how they're going to record. Some record digitally and some record analog. The only thing that can be said accurately is that there is nothing in a professional half-inch analog tape that keeps it from being the best, without hurting the product or making these artists feel they're not getting a great-sounding product. Some of them hate digital. But they are in a position, regardless of what they come out with, to make their own decisions about which format they use.

Mitsubishi now has a 96kHz–sampling-rate digital 2-track that I hear is terrific. Almost every new digital machine that has been raved about has been reviled a year later. It's like you hear this amp is great, then two years later you see it's not on the "Recommended Components" list anymore. And you write a letter and someone says, "Well, we found the coloration of the midrange was not recognized instantly." So you get the amp of the moment and the speaker wire of the moment. It is these type of ventures that lose credibility for the audio magazines.

A funny thing is if the amp is really good, it's good. There are not that many good amps around. It has to be that some of the amps they said were good really weren't. They were new, they liked the bells and whistles. But they have to survive the test of time. In 1958 the very first solid-state amps were offered for sale. I heard one of the first ones. Prior to that I had never heard a solid-state device. I walked into where I went to listen to my audio. Got a call to "Come on down, you won't believe it." Went down with my recordings, which I treasured and knew sounded good. There was one telling recording of a solo violin recorded outdoors, no room acoustics. A very good recording of unaccompanied Bach. I still have the record. I hooked up a big KLH 9, a legendary electrostatic speaker, to this brand new solid-state 100W amp. I put on the record and couldn't believe my ears. Jack, the proprietor of the store, said "What do you think? Have you ever heard anything like it?" And I said "No, not live, not recorded, I've never heard anything like it. It's horrible. It's the worst thing I've ever heard in my life."

---

Footnote 6: Replacement filters that reduce phase shift in the Sony PCM-1630 are supplied by a company called Apogee (not the loudspeaker manufacturer).