Analog Corner #9

(Originally published in Stereophile, April 12th, 1996)

A Consumer Electronics Show is always fun: You see and hear new stuff, greet old friends, and occasionally meet the disenchanted. Since this was my first Winter CES as a Stereophile writer, I was expecting more feedback than I'd previously received at shows, and I wasn't disappointed. Despite an icy shoulder or two, most of it was positive, and some of it was flattering. But one reader I ran into was genuinely pissed at me for "wasting most of an analog column" writing! He was talking about the January issue's "Analog Corner," in which I wrote about the Audio Alchemy DTI•Pro 32 and the major sonic improvements it produced with CDs. He punctuated the reading of his analog riot act with "Hey, I don't even listen to CDs!"

That was also the column where I wrote, "You think I'm a diehard? There are still audiophiles who refuse to listen to CDs, period." I think that slightly derisive comment is what really set him off. So will this column because it's also about digital—sort of—but I hope all you analog devotees will peruse it anyway, along with the digital-only readers who usually cross the street when they reach the "Analog Corner."

Digital versatile disc
Before I get to what I wanted to write about this month, I must tell you about an incident at the '96 WCES. Toshiba sponsored a press dinner to launch DVD. Probably a hundred journalists were there and Toshiba's Craig Eggers was talking up the new DVD format both for video and for audio. I'm paraphrasing now, but Eggers goes "And while the new audio format has yet to be agreed upon, even a committed analog fanatic who thinks vinyl sounds better than CDs, even Michael Fremer, will be impressed with the sound of the new format."

So, much of the audience guffaws, but over at the next table where the "it all sounds the same, let's take the fun out of this stuff" crowd was sitting comes the hiss of a derisive airbrake ("tsssssss"). The sound only served to fan the flames under my already massive ego which had just been greased and grilled to perfection before my peers.

These guys don't give up. How are they going to write about this new and vastly improved digital sound, when they've been spouting "virtually perfect" about 16-bit digital for over ten years? Dissing me with their airbrakes ain't gonna do it, not when, as you will read later, the guys who work in studios and mastering suites every day hear the very significant differences between their mike feeds and the recorders they use.

Hiss and diss all you want, oh digital shills, you'll be eating your words for breakfast, lunch, and dinner before you'll be able to convince your gullible followers that there's something better than perfection coming down the pike on those high-density DVDs.


Actually I know what their cop-out will be: They'll ignore the possibility for improved sound and concentrate on the glories of the surround channels. This will be a good thing if it can be done without crapping up the two or three channels in the front. But if adding extra channels does crap things up (as I've heard at all of the AC-3 demos I've attended), they'll ignore that too! It's the nature of the beast. That's why David Elrich (with whom I shared a table at that dinner) wrote in The New York Times in mid-January that DVD "look[s] and sound[s] as good as the film in your local movie theater."

Yes, as shown at Toshiba's display with a Faroujda line quadrupler, Runco projector, and Stewart screen, all hand-tweaked by Joe Kane literally by the hour, it looked shockingly close to film. But will more than a few Times readers spend the $75,000 or so to get that kind of picture at home? No. So why the hyperbole? Frankly, rather than helping to successfully launch DVD, such talk will only lead to mass disappointment, even though DVD's picture—even on your average motel television—will look incredibly better than anything most of us have seen at home so far, including laserdisc.

A stellar listening session
I've yet to attend one of Stereophile's mass listening sessions. Like you, I've only read about them, imagining the scene: your favorite audio critic dweebs sitting stiffly on hardbacked chairs, eyes closed, grimaced faces frozen in serious sonic concentration as loudspeaker after loudspeaker is paraded behind a dark mystery scrim in a dangerous game of high-stakes audio poker. Wow! The tension! The bad breath! The B.O.! Surely my opportunity to experience such audiophile thrills in the presence of my peers will come to pass. [Keep this up Fremer and it most surely will.—Ed.]

Meanwhile, imagine this scene: crowded around the mixing board at Hollywood's OceanWay, many an audiophile's favorite recording studio, are (starring in alphabetical order) Bob Clearmountain, Bernie Grundman, George Massenburg, Doug Sax, Bill Schnee, and OceanWay owner Allen Sides. Do any of these guys need an introduction?

On the other side of the glass is a jazz combo which Sides has miked and mixed through the console. The board's output is fed to four different recording devices: a 20-bit converter dithered and noise-shaped down to 16 bits to feed a recorder; a true 20-bit optical disc recorder; OceanWay's famous modified 1/2" two-track ATR-100 with no noise reduction, its Flux/Magnetics headstack threaded with Scotch 996 running at 30ips; and a new Sony-designed and -built, blue-laser–driven digital recorder (Footnote 1). The outputs of the recorders feed a selector switch so that each can be compared against the mike feed.

And the envelope please
Contrary to unpopular opinion, none of the recorded versions sounded like the mike feed, according to all in attendance. We can therefore dispense with all that hokum about how those of us who don't like 16-bit digital simply prefer our analog distortions and wouldn't like a "mike feed."

Sides, no digital lover he, thought compared directly with the mike feed, the 16-bit digital was the worst, the 20-bit digital better, the new Sony digital system "significantly better than the other digital recordings", and the two-track analog the best. But...he was "impressed" with the new system to the point where he thought it was the best digital he'd ever heard. "No unpleasant digital artifacts" with a "really open top end" and "very clean, clear" midrange, he told me. It "didn't get small," like most digital.

While Sides preferred the sound of his analog machine to that of the prototype Sony system, he admitted it was a "tough call" in many ways. It was "hard to pick" which sounded better, he told me, though he ultimately decided Sony had "a little bit of work to do" before the system would be ready. Then? Well Sides didn't say it in so many words, but I gathered from what he did say that he would still record in analog—that is, until he runs out of Scotch 996 recording tape (which many engineers think is the best in the world).

According to Sides, 3M/Scotch are quitting the recording tape business and there's only a year's stock of tape left. Isn't that nuts? 3M apparently makes money on the stuff; the problem is that, by today's corporate standards, it's not enough money. And since the consensus seems to be that the new disc-based digital recorders will soon be so good that even diehard analog geeks will be convinced, the five- or seven-year window of continued analog opportunity is not enough to make a buy-out of the tape division attractive to an outside suitor. 3M has no choice but to dump its whole Scotch division on the scrap heap. Remember when JVC Japan did that with its superb vinyl pressing facility? When will they ever learn?

Back to the future
One reason the new Sony digital system sounds so good is that it is supposedly flat out to 60kHz and probably well beyond that. The ATR analog machine is flat to about 40 or 50kHz, so we're talking about a digital machine which finally goes way out. Whether that kind of performance will be translatable to the new DVD audio standard is the big question which has yet to be answered.

According to Allen Sides, all of the other attendees at the OceanWay listening tests—with the exception of Clearmountain who preferred the 20-bit system to everything else—agreed that this new Sony system was the best digital they'd ever heard, and very close to the best analog. So if your attitude is that good sound is what counts, if you're not dogmatically anti-digital, this is good news.

Now the bad news
Don't gloat yet, you digiphiles, because Sides and an increasing chorus of other engineers and mastering mavens with whom I've been speaking have a new complaint about CD sound: Despite their efforts at improving the mastering chain, the final CDs that reach your greasy little fingers are sounding worse and worse as time passes, not better. There is a reason why analog devotees are finding that import LPs cut from PCM-1630 masters frequently sound better than the domestic CDs, and it's not pretty.

Sides related to me the sad history of a new BMG Classics title recorded all-analog at OceanWay and mixed to two-track on the ATR at 30ips. The story is so sordid, so damning of a client and potential future client, I asked Sides if I could relate it to you, and he said, "Yes, what the hell." Love his studio, love his recordings, love his style.

The production is a tribute to Leonard Bernstein, a remake of West Side Story with various pop singers covering the tunes, including—get this—Little Richard singing "I Feel Pretty"! It should be out by the time you read this. Allen Sides feels it's a wonderful production featuring outstanding performances and, of course, superb OceanWay analog sound. But don't expect the CD to sound anything like the original recording. Why?

It's not the fault of the mastering engineer, who was Bernie Grundman (enough said). According to Sides, the analog master tape was sent over to Grundman's (literally next door), and transferred to a 1630-format U-Matic digital master tape using an Audio Alchemy DTI•Pro 32 and Bernie's Apogee UV1000 prototype. (Sides and Grundman think the prototype Apogee, of which only a few exist, sounds even better than the finished product.)

Sides and Grundman shipped the 1630 tape off to BMG, keeping a CD-R for reference. When they got a sample CD back from BMG, Sides says, he "never heard sound so terrible." Compared with the CD-R, the CD sounded tinny, metallic, hard—you know, "digital."

What had happened? Well, leaving out the fact that a "mystery source" who shall remain anonymous—it wasn't Grundman or Sides—told me that Grundman was forced to "compress the shit" out of the dynamic-sounding master tape so that it would sound "loud" on the radio (Footnote 2), Sides found out that on the way to the CD pressing plant, the master tape went through two or three D-to-D copies and one trip through a digital editor. Of course, no one paid any attention to the sound after each of these transfers. As long as all the data are pumped through without dropouts, all must still be "perfect," or so everyone thought.

But wait! There's more! In an effort to increase plant "efficiency" whoever pressed the disc first transferred the 1630 tape to an Exabyte tape. This is an 8mm datastream device which allows the glass master to be cut at double speed. This, according to the mastering engineers with whom I've spoken—many off the record by phone, but in whispered-toned conversations which made me feel as if we were meeting under a bridge in Eastern Europe before the fall of Communism—is an increasingly common practice at pressing facilities. Double-speed mastering, my friends, is the "Dynaflex" of the '90s.

According to an ever-growing list of mastering engineers, this double-speed Exabyte system bites! And it's spreading like head lice. After all, a $1 manufacturing cost for a $17 retail disc is simply not a large enough margin. By cutting the glass master from an Exabyte tape, the plant can halve the mastering time. So what if it halves the sound quality?

And here's the coup de grace: By the time Grundman and Sides had received their sample CD, 240,000 copies had been pressed. You don't think BMG is going to bury them do you? So if you want a copy of this album, and from what Sides told me, you will because this is a wonderful set—Patti LaBelle and Natalie Cole doing "America" is another choice number—you should wait. This disc should sell far in excess of a quarter-million units. When you hear that it has done so, the remastered copy you buy from a high-traffic store will sound, according to Sides, "100% better, but still not nearly as good as the original."

Deja vu all over again
Those of you old enough will remember what happened to the LP between, say, 1968 and the late 1970s: the haphazard, get-'em-in, get-'em-out mastering; the automatic compression and rolled-off highs and lows; the sloppy plating and even sloppier fast-in, fast-out pressings; the use-the-stamper-until-the-grooves-are-worn-out syndrome; and the thin, noisy, warped vinyl. Well, it's all starting to be repeated in the digital domain. While the cost-cutting details are different, the end result is the same: bad sound.

Why is this happening now? Maybe sonic "downsizing" is being done just to increase the bottom line. Or maybe, just maybe, decreasing the sonic quality now will soften the consumer up for the next big thing: the DVD-based Super CD. Maybe I'm just being cynical, but isn't that what happened the last time?

So if, like me, you're sitting there comparing commercial CDs with carefully manufactured LPs and the records sound better, is it because your turntable is adding "pleasant-sounding artifacts and fake L–R spatial information," as some would have you believe? Or is it because ignorant slobs along the digital highway are dumping trash in the bitstream and it's fouling your CD player? That is, if you even listen to CDs!

Footnote 1: This is an experimental "work in progress" using Sony's new "Direct Stream Digital" recording system, according to Sony consultant Marc Finer. Eeek! I'm getting in over my head. Where's Robert Harley when I need him?

Footnote 2: This is a truly dumb idea, since radio "compresses the shit" out of the music anyway, in an attempt to out-loud the competition on the dial.