Digital Processor Reviews

In the beginning of the digital age, many reviewers abandoned ship. They listened to the first generation of digital equipment, compared it with their optimized analog systems, and concluded that digital was hopelessly flawed. They carried the analog banner high with seemingly justified righteous indignation, criticizing all things digital at every opportunity. Meanwhile, records rapidly became an endangered species, and audiophiles found themselves in a box: If we wanted anything new, it was going to be digital. If we wouldn't accept digital, we weren't going to be hearing much new music.

I made an unpopular decision at the advent of the digital age (or, as Sam Katz of The Audiophile Voice would say, the "Digital Winter"). I decided to stop comparing my cherished analog system with the parade of disappointing digital equipment that found its way into my system, opting instead for peaceful coexistence. I accepted digital as a separate but unequal medium. Like any analog-loving audiophile, I continue to carefully seek out good vinyl at every opportunity. I've painstakingly maintained my cherished record collection with new plastic covers for most of my albums, new inner sleeves, prudent use of my VPI HW-17 record cleaner, and regular use of LAST record preservative. However, I also elected to buy everything I liked musically on CD, if that was the only way I could get it.

And at least one courageous equipment reviewer bucked the prevailing trends, continuing to review every digital product of merit. He was there to describe each new improvement in CD players. He quickly let the rest of us know about the advantages of separate transports and processors. He told us about the relative performance of coaxial connections versus Toslink. He introduced most of us to the ST/AT&T standard. He bombarded us with technical explanations of different DACs and filters. He used his ears to tell us about such tweaks as CD Stoplight, whose sonic benefits he couldn't verify with his own measurements.

This brave soul is none other than Stereophile's own Robert Harley.

I had read everything RH had written and agreed with virtually every word of Bob's reviews of digital hardware. As a result, I was thunderstruck by his recent assessment of the $13,950 Mark Levinson No.30 processor (Vol.15 No.2). Could it possibly be as good as he said? I couldn't contain myself, so I called him. Yes, he assured me, it really was that good. I reread the review. I called again. Patiently, RH once again assured me that the No.30 was everything he'd described, and more. In his opinion (which I valued more than that of any other reviewer on things digital), the No.30 had indeed set a new standard for digital reproduction. I had no choice. I ordered my own No.30.

Since RH did his usual splendid job of explaining all of the No.30's technical details, I won't rehash any of that here. I will say that connecting and fully exploiting the No.30's capabilities is a significantly more imposing job than plugging in most other processors. I used the No.30 in my standard reference system, fed by a Theta Data transport (reviewed by RH in Vol.14 No.11) through a wide variety of Toslink, coaxial, and ST/AT&T cables.

My initial impressions of the No.30 were mixed. Yes, there was no doubt that it was the best digital processor I'd ever had in my system. It had remarkable tonal neutrality, with a very realistic mix of fundamentals and harmonics. The bass was strong, powerful, and articulate; the midrange was both full and detailed; and the trebles were smooth, delicate, and extended. Soundstaging was also first-rate, with very good width and depth as well as clearly delineated locations in both dimensions. Performers were placed precisely within the stage, had the requisite amount of body, and were appropriately surrounded with air, creating a truly believable sense of space. Dynamics, transients, and inner detail were the best I'd heard from digital.

In spite of all of my very favorable impressions, I was actually disappointed. Yes, the No.30 was superior to every processor I'd heard. But I had expected more. RH had been so enthralled with the No.30 that I thought I would be "blown away." While I was very impressed, I wasn't convinced there was a need to revise all of the "Recommended Component" categories to make room for the No.30 in a class by itself (as was done in Vol.15 No.4).

I continued to listen to the No.30 and was consistently impressed with its performance across the board. No, there were no glaring faults. Yes, it truly was the best I'd heard. But, damn—a new category? Had RH finally failed me? I read his review over and over again. I really had to agree with everything he had to say, but I still wasn't convinced that the No.30 was that much superior to the very excellent competition.

Suddenly, after a dozen rereads, I found the answer. There it was on p.140, staring at me all the time. According to RH, "The electrical interconnect brought the No.30 down a notch."

In my eagerness to maintain a fair comparison, I'd listened to the No.30 only with electrical (coaxial) interconnects. Because of the various reviews I'd been working on, I'd been carefully using a set of four electrical cables with every component that found its way into my system. In addition to helping me understand the relative strengths and weaknesses of these cables (which I'll report on in an upcoming review), I didn't want to give the No.30 an unfair advantage over the other processors.

With callous disregard for my carefully thought-out review procedures, I quickly overruled myself. As rapidly as possible, I inserted a bright orange Altis Glass Fiber ST/AT&T cable between the Theta Data and the Levinson No.30. I selected Take 6's sonically splendid debut CD (Reprise 25670-2) and literally raced for my listening chair. Of course, I didn't have to race. The Theta Data is the sloth of transports in getting started, and the wait seemed endless. I nervously kept pressing Play and muttering to myself. Press—mutter—press—mutter—press...

Bingo! My listening room disappeared and I was transported to a front-row seat. Right there in front of me were the six stunning acapella voices of Take 6. Oh, the splendor of their voices! The artistry of their arrangements was better than ever. Sonorous, spacious, dynamic. Real musicians performing in a real space! But wait, there were no musicians in my room—and it still was my listening room. The illusion was compelling. The system had indeed disappeared.

Every subsequent disc was equally stunning. I was able to put Harry James's big band right behind my speakers with The King James Version from Sheffield Lab (CD-3). I moved effortlessly back east with Midori's Live at Carnegie Hall (Sony SK 46742). Pushing a few buttons dropped me into a more intimate setting with the Indigo Girls' Nomads, Indians, Saints (Epic EK 46820). A radical change in mood left me in a raucous setting with the Red Hot Chili Peppers' Blood Sugar Sex Magik (Warner Bros. 26681-2). Amazingly, the No.30 made every type of music sound better.

Although the Levinson had been the best processor I had heard when I used any number of excellent coaxial cables, it truly moved into a class of its own with the Altis Glass Fiber. There was simply more of everything. More information, more width, more depth, more dynamics—more, more, more. It was as if an entire level of smaller details was revealed, making every sound more discrete. Without exception, every aspect of the musical performance improved. I was staggered by the wondrous sounds I was hearing for the very first time from CDs I thought I knew intimately. As I should have expected, RH was vindicated. The No.30 does deserve a Class A rating all by itself.

Or does it? I can't help but wonder about a couple of other possible explanations. The first is that the ST connection itself is vastly superior to coax/electrical. If this is true (and it is), then my comparison of the No.30 with any other processor is not fair. I haven't auditioned any other processor with the ST/AT&T connection.

My second question concerned the choice of ST cable. Was the vast improvement heard in the No.30 at least partially the result of the choice of ST cable? At least on this point we have more information. RH and I used different ST cables. It does not appear that our similar conclusions are actually attributable to some new breakthrough in ST cable.

Which brings me to my final question: Do all ST cables sound the same? Are you listening, RH?

Summing up: Using a standard coaxial connection and cables I'd used extensively with other processors, the Mark Levinson No.30 is the best processor I have yet heard. While bettering the competition in almost every way, the magnitude of these improvements is not astronomical.

Using the ST connection, the No.30 is everything RH had said. It truly sets a new standard for digital reproduction, besting by a wide margin everything else I've heard. It really does merit its new classification. Without doubt, the Mark Levinson No.30 Reference Digital Processor defines today's state of the art.

RH left us all up in the air at the conclusion of his review; I thought I'd do the same. Based on both of our reviews, it appears that an ST cable is the only way to go. If so, we must wonder together about Madrigal's decision to offer their own cable, the MDC-1 (Madrigal Digital Cable), reported to be a coaxial cable with RCA to XLR connections. Madrigal feels this cable is superior to ST!—Jack English