dCS Verdi La Scala CD/SACD transport & Delius D/A converter Page 3

But you must bear in mind one unsmall matter: All that goodness depended on a number of things, some of which may confound your expectations:

• Without the slightest doubt, the combination worked better at playing music when the Delius was in Master mode—overriding, as it were, the word clock in the datastream and that produced by the Verdi La Scala. The former didn't so much sound different (although imaging junkies may not agree, given their heightened sensitivity to such things as stage dimensions) as simply allow me to relax and be fooled into thinking I was hearing strings of notes and not just sounds.

• I didn't care for the characters of all of the filters and rolloff choices, and those preferences were not consistent from disc to disc. Unfortunately or fortunately, depending on your approach to the hobby, you have to work with this system to get the most out of it.

• Although I've never been wild about hi-fis built around passive preamps, I heard some of the best digital-source music of my life when I drove the balanced inputs of the First Watt F1 amp (see "Listening," p.43) directly from the Delius's balanced outputs. The string sound on George Szell's recording of the "Prelude and Love-Death" from Wagner's Tristan und Isolde (SACD, Sony Classical SS 89035) was not only more textured than ever before, but that texture no longer seemed a mere outgrowth of the tape hiss. And on Johnny Cash's indispensable Live at Folsom Prison (SACD, Columbia/Legacy CS 65955), literally everything was better, from the room sound to the swaggering humanity of the music.

• With the preamp back in the system, the sound of upsampled "Red Book" digital, taken through the 1394 connection and sent to one set of line inputs, was always distinguishable from the sound of the recording in its native sampling rate and word length, as heard through the S/PDIF (RCA) connection—but I didn't always prefer the former. Upsampling always sounded smoother and, even more so, always airier. Spatial performance seemed the biggest beneficiary of upsampling, and spatial cues and room sound were consistently more obvious when I got all my music via the 1394. But some of the time—on rock and pop, mostly—I distinctly preferred the non-upsampled feed, which seemed to retain greater rhythmic integrity and a bit more drama and attitude.

An example: On "Temperance Reel," from Tony Rice's first solo album (CD, Tony Rice, Rounder CD 0085), when he's backing Richard Greene's first fiddle solo, mandolin player extraordinaire David Grisman adheres to an interesting pattern of stressing the odd-numbered measures—really digging into the strings—and backing off on the even-numbered measures. The effect was smoothed over in the upsampled feed—it was much more pronounced and exciting at 44.1kHz through the S/PDIF inputs of the Delius.

I know: Go figure.

An unfair comparison
Perverse though it may sound given all the stunning software at hand, I couldn't resist comparing the dCS Delius DAC with the most different product I could find: a 47 Laboratory 4715 (Shigaraki) DAC (footnote 2). The Delius contains several hundred parts and provides every audiophile convenience imaginable, and then some; the 0x-oversampling Shigaraki contains literally 20 parts and doesn't even incorporate a digital or analog filter. Hey, sounds fair to me.

The difference, as you might have guessed, was pretty obvious. The dCS Delius was so timbrally neutral and (apparently) accurate that it made the Shigaraki sound colored and lacking in frequency extension. The dCS made Paul McCartney's voice on "She's Leaving Home," from the Beatles' Sgt. Pepper's Lonely Hearts Club Band (CD, Parlophone CDP 7 46442 2), sound more real, more like Paul McCartney. Listening to the same recording through the Shigaraki was like hearing it through an old-fashioned radio: The frequency response was slightly lumpy throughout the upper mids and highs, and it sorely lacked ambience, air, and reverb decay. The dCS sounded like high fidelity; the 47 Lab sounded a bit quaint.

But the Shigaraki was at least as compelling, musically, as the Delius—and I thought the music had more drama and texture straight through the Shigaraki than when upsampled through the Delius's 1394 connection. That distinction was equally pronounced on classical music. The burial mass from Heinrich Schütz's Musikalische Exeqien, performed by La Chapelle Royale under the direction of the great Belgian conductor Philippe Herreweghe (CD, Harmonia Mundi 901261) had more hall sound through the dCS converter, and was arguably smoother. But in upsampling mode, it was almost too smooth. By comparison, the relatively cheap Shigaraki DAC had more texture—including, I admit, more grit in the spaces between the notes—but was also more involving.

Some listeners will be disappointed that the Shigaraki DAC presented the music with audibly less channel separation: In either datastream mode, the dCS Delius spread the music out on a significantly wider stage, while the 47 Lab was closer to mono. There was no question regarding the dCS Delius's wild superiority in terms of spatial reproduction: I've never heard greater stage depth or width—and even psychoacoustically convincing height—from any source component, regardless of the technology it represents. And the sense of presence associated with players and singers within that space was second to none. Through the dCS Delius, even average recordings took on new life in this regard—and spatially superior recordings were well beyond anything I'd previously heard from a digital source. In one of those instances of sonic refinement adding to the music's emotional effectiveness, my preferred version of the Howells Requiem—a mildly tarted-up Sensaura recording (United 88033)—was more effective through the Delius.

Here's a comment probably not germane to the spirit of this comparison that I'll make anyway: The $1400 Shigaraki wasn't the least bit embarrassed by the $9995 Delius, although the latter's sophistication was obvious. After the first weeks of the review period, during which I did the bulk of my component swapping and note taking, many was the time I found myself leaving the Verdi La Scala and Shigaraki combo in place—and enjoying the hell out of it. I don't think the pairing is silly at all.

But then, I didn't have to pay $15,000 in order to enjoy the Verdi La Scala for four months. And this is the third time in two years I've borrowed a Shigaraki DAC, which leads me to wonder if I should let it make an honest man of me, or stop trying to get milk without buying the cow, or whatever that aphorism is supposed to be.

Once again, the high price of a product—or a pair of products—is the 800-lb gorilla in the corner, and it's time to acknowledge his presence. This is not a "maybe" kind of thing: I have absolutely zero doubt that the prices of the dCS products fairly reflect what it took to develop and manufacture them. And I respect the extent to which their US distributor, Audiophile Systems, has resisted profit-taking on the line: A quick comparison of the prices on their website and on the dCS site will confirm that American hobbyists are not being gouged in the least.

$25,000 may not exceed the cost of what went into these marvelous things, but it does exceed what I'm able or even willing to pay for a CD player. Thankfully for dCS—and, arguably, for the notion of progress in our hobby—there remain consumers out there who can and will ante in for the sake of having the best right now. The combination of the Verdi La Scala transport and Delius D/A converter may well be it: It certainly is in the context of my own experience.

The dCS system, the worth of which is so tied to the ingenuity of its software, is like a brilliant new book just out in hardcover: I respect, and to some extent envy, those who line up to buy it, even as I conspire to wait a few years for the paperback.

Footnote 2: AD reviewed the 4715 DAC with the 4716 CD transport in March 2003.—Ed.
Mull House, Great Chesterford Court
Great Chesterford
Saffron Walden, Cambridgeshire, England CB10