Digital Processor Reviews

The analog filter is a passive, fourth-order type designed for high out-of-band attenuation, low phase shift, and good transient response.

The analog output stage is based on the circuits Keith Johnson has been refining for more than 20 years. The SDR-2000 Pro's output stage is very similar to the electronics used throughout the signal path in Keith's recording electronics, which are used to make all the Reference Recordings releases. The circuit features a differential JFET input stage and a MOSFET cascode, followed by a differential cascode voltage-gain stage. High-current power MOSFETs form the analog stage's output section. The bandwidth of the SDR-2000 Pro's analog stage is 1.2MHz, in keeping with Spectral's belief that a very-wide-bandwidth circuit improves the audioband performance.

How the analog stage handles the balanced DAC outputs is unusual. The + and – phases are combined so that the common-mode rejection occurs in the processor, not in the preamplifier (or power amplifier, if the preamp is a fully balanced design), yet the circuit is fully balanced. Moreover, this topology ensures that the SDR-2000 Pro's single-ended outputs benefit from the differential DACs. Some fully balanced processors merely take the + phase of the balanced signal and connect it to the output RCA jack, throwing away the benefit of balanced DACs for single-ended users.

The HDCD license calls for 6dB of attenuation to non-HDCD discs (or 6dB of gain for HDCD discs) to maintain a comparable apparent level between standard and HDCD-encoded discs. As I mentioned in my Follow-Up last month on the Mark Levinson No.30.5, the attenuation can be done in the digital domain by the PMD100 HDCD decoder/filter, or in the analog domain. In the SDR-2000 Pro, the attenuation is analog—a better alternative than digital attenuation. De-emphasis, however, is performed in the digital domain by the HDCD decoder.

To say I was impressed by the SDR-2000 Pro's design, execution, and overall build quality would be an understatement. This is a tour de force in digital-processor design—the best processor, in fact, that Keith Johnson knows how to design; absolutely no corners were cut to make the product less expensive to build, I was told. The SDR-2000 Pro's relatively reasonable price reflects Spectral's philosophy of making the best sound possible in an elegant—yet not extravagant—package.

Setup
I connected the Spectral and MIT equipment to get it working and warmed up, but didn't tweak the system until Spectral's Rick Fryer and Keith Johnson visited my listening room two days later. I removed my usual accessories (including the Tice Power Block) so that we could start from scratch in optimizing the system for the Spectral electronics.

It turned out that the system setup, room treatment, loudspeaker placement, and needed accessories were significantly different for the Spectral than for my reference system. Where I had achieved wonderful space and depth easily with the Audio Research LS5 Mk.II preamplifier and Audio Research VT150 monoblocks, it took most of a day of tweaking to get good soundstaging from the Spectral system.

Moreover, the Spectral electronics appeared to be much more setup-sensitive than I've experienced with any other components. For example, re-routing the long interconnects between the preamp and power amplifier away from other cables made a dramatic improvement in the purity of instrumental textures. A slight layer of grain was removed just by running the interconnects across the middle of the listening room and draping them over the front of the power amplifier. This experience was repeated with different tweaks as Keith, Rick, and I spent the day listening and adjusting.

It was an amazing experience to watch Keith listen to and critique a system. He has a remarkable ability to instantly hear when something isn't right; more importantly, he knows what causes a specific problem and how to fix it. This sensitivity to setup may be unique to Spectral components, or it could be that the Spectral electronics unmasked very-low-level artifacts obscured by other components. Spectral was correct, however, in insisting that I hear the SDR-2000 Pro with their entire system, as some of the qualities I heard in the SDR-2000 Pro were more apparent through the Spectral electronics.

I should note that I didn't really hear a full-blown Spectral/MIT system: the interconnects were the entry-level variety, and the loudspeaker cables were more than ten years old. Further, the AC-power conditioning wasn't optimal; the system needed an MIT Z-2 isolation system to keep noise from the digital electronics from getting back into the AC line, where it could affect the preamplifier. Putting AudioQuest's RF Stoppers on the AC-power cords helped somewhat, but not to the degree I've come to expect from a transformer-isolated conditioner.

Listening
I went three rounds with the SDR-2000 Pro. First I auditioned it with my usual reference system, then in the Spectral system, then, toward the end of the auditioning, back in the Audio Research system. Comparisons with the No.30.5 and SFD-2 Mk.II in the Spectral system were made with the MIT-made single-ended interconnects, and with AudioQuest Diamond x3 balanced when listening through the Audio Research electronics.

Interestingly, my impressions of the SDR-2000 Pro gained with my reference system were consistent with what I heard in the all-Spectral system, although the two presentations were very different.

The SDR-2000 Pro very quickly established itself as a topnotch processor. Even a brief listen before the unit was fully warmed up (I couldn't wait) suggested that the SDR-2000 Pro was a contender for the best digital playback I'd heard.

With two days of warmup before the serious listening, the SDR-2000 Pro really opened up and showed what it could do. The first thing that struck me about the unit's presentation was the astonishing resolution of fine detail. Very-low-level musical details suddenly became alive and vivid, with an ability to clearly and easily reveal the most subtle of nuances. Importantly, this detail wasn't forward, etched, or analytical, but soft, subtle, and delicate. A perfect example of the SDR-2000 Pro's extraordinary resolving power was the fine cymbal work during the bass solo on Mike Garson's "A Song for You," from The Oxnard Sessions, Vol.2 (Reference RR-53CD), and the brushes earlier in the same track. I could hear so much more of the fine detail of the stick hitting the cymbal and the cymbal's subsequent ringing and decay that I felt a much greater impression of hearing live music. Similarly, the sound of the brushes on the same track took on a specific, unmistakable character.

Through lesser processors, these low-level details tended to become blurred, making the sound less identifiable and more synthetic. Although I could identify that the drummer was using brushes on the snare, my brain had to work to fill in what was missing when listening through processors with less resolution. Through the SDR-2000 Pro, the low-level cues were right there, with no need to strain or work to hear the detail. I must reiterate that this detail wasn't hyped or forced, but subtle, understated, and refined.

The SDR-2000 Pro's high resolving power also produced an amazingly true rendering of instrumental timbre. The fine detail that gives an instrument its unique character was beautifully portrayed, infusing textures with a palpability, presence, and life that were unmatched in my experience reviewing digital products. There was a realness to the sound that was captivating. For example, the voice and harmonica on Doug MacLeod's great Come to Find disc (AudioQuest AQ CD1027) had a lifelike, believable presence that raised goosebumps more than once in my pitch-black listening room.

Compared to the Mark Levinson No.30.5, the SDR-2000 Pro provided more of the inner detail and fine resolution I've been talking about. Although the No.30.5 was excellent in this regard, the nod went to the Spectral. Similarly, the SDR-2000 Pro bested the Sonic Frontiers SFD-2 Mk.II in resolution. Note that the SDR-2000 Pro had a livelier, more immediate midband compared to the No.30.5. The latter sounded more laid-back and gentle, with a less incisive and immediate perspective. The SDR-2000 Pro was, however, less forward than the SFD-2 Mk.II, with a more refined presentation.

Another aspect of the SDR-2000 Pro's portrayal of timbre I found musically important was its ability to keep instrumental textures separate from one another. Instead of fusing the textures into one big, synthetic sound, the SDR-2000 Pro maintained the distinctness of instruments—even during loud passages. Listen to the horns during the ensemble playing on the HDCD-encoded disc From the Age of Swing (Reference RR-59CD). I could clearly hear each instrument's timbre—baritone sax, alto sax, trombone, trumpet—as separate entities within the soundstage. This total lack of congealing was apparent on HDCD-encoded discs as well as standard CDs (although to a lesser degree).

A very tough test for a digital processor is Frank Zappa's orchestral masterpiece The Yellow Shark (Barking Pumpkin R2 71600)—particularly the final track, "G-Spot Tornado." This piece has so much going on that it turns into a congealed mess unless the system can keep the individual elements separate. When reproduced well, this music has an intricacy and depth that reveal the full measure of Zappa's genius. Hearing this disc through the SDR-2000 Pro was a revelation, with a whole new level of nuance and musical information revealed. Moreover, the SDR-2000 Pro didn't congeal or thicken—even during the full-scale, very complex passages.

The soundstaging was similarly impressive—particularly when the SDR-2000 Pro was at the front of the all-Spectral system—sounding very tight and focused, with compact and sharply defined images. The example that jumps to mind here is the tambourine at the beginning of "Festival Day in Seville," from the HDCD sampler disc. Through the SDR-2000 Pro, the image had a startling pinpoint precision. These razor-sharp image outlines were surrounded by a wonderful sense of air, further giving the impression that the sound was composed of separate, discrete instruments rather than appearing as variations on a continuous fabric. This is one quality that was most apparent through the entire Spectral system.

I also found the soundstage very wide, with an almost wraparound quality. A sense of air extending to the soundstage edges blurred the boundaries between the illusion and reality. In other words, the soundstage didn't end abruptly, instead seeming to have space and air well past the outside edges of what I had previously regarded as the soundstage limits. I noticed this particularly on the sax on the previously mentioned Mike Garson disc; the way it lights up the acoustic on the loud passages was particularly revealing of the differences between the processors under audition. Through the Spectral, I got a more expansive view of the hall, with a huge sense of air and bloom around the instrument.

Similarly, the SDR-2000's depth was stunning, with an ultrafine resolution of depth gradations. The soundstage didn't just sound big and deep, but had a remarkable spatial precision from front to back. Moreover, the Spectral had an astonishing transparency that rendered images at the back of the hall lifelike and immediate. The palpability I described earlier in relation to the SDR-2000 Pro's rendering of timbre was partially a result of the crystal-clear transparency.

Another area in which the SDR-2000 Pro excelled was in the lack of glare and edge. Piano didn't have a glassy sheen or a breaking sound on transients, instead sounding pure and liquid. I also noticed this quality on vocal sibilants—Robben Ford's vocals on Robben Ford and the Blue Line (Stretch STD 1102) was a good example. Through the Spectral, I was less aware of sibilance, and what sibilance there was didn't have quite the degree of hashy sizzle. The SDR-2000 Pro's very clean midrange and treble were also manifested on violins, horns, and flute. Cymbals lacked the hash and grain riding on the waveform. The sound had a purity and rightness of texture that was addicting. This lack of glare and grain was partially a function of the digital interconnect. With the MIT coaxial digital cable, I heard some hardness on transients; through the AudioQuest Diamond x3 balanced digital cable, the hash went away, but at the expense of some of the resolution of fine detail described earlier.

The SDR-2000 Pro's bass tended toward lean, tight, and articulate, rather than to big and full. The Spectral had less sense of weight and power in the bass and midbass than the No.30.5 did, but that bass was "faster" and had better pitch definition. Through the SDR-2000 Pro, I was better able to hear the pitch in the very simple bass line in the opening track of the Doug MacLeod disc. Similarly, John Pattitucci's amazing bass playing on "Got a Match?," from Chick Corea and the Elektric Band's eponymous release (GRP GRD-9535), was easier to follow. The high-speed dual leads between the bass and keyboard were superbly delineated by the Spectral, largely because of the processor's excellent resolution of pitch and complete lack of bloat. The Spectral's bass had less "bloom," however, with a drier rendering.

The combination of these extraordinary qualities worked together synergistically to draw me into the music in a way I haven't before experienced. Conventionally coded CDs had a level of detail and smoothness of texture I didn't think possible. And when the SDR-2000 Pro reproduced HDCD-encoded discs, particularly through the all-Spectral system, the result was a musicality beyond that of any reproduced music I've previously heard.

Conclusion
As much as I've been impressed by the two other serious HDCD-based processors I've reviewed recently, Spectral's SDR-2000 Pro is the best-sounding digital processor I've heard to date. The Spectral offers a smaller increment in performance over the $15,950 Mark Levinson No.30.5 but sounds significantly better than the $5295 Sonic Frontiers SFD-2 Mk.II. While the SDR-2000 Pro and No.30.5 shared many exceptional qualities, I thought the Spectral excelled in low-level resolution, soundstaging, and harmonic purity. The SDR-2000 Pro redefines what we can expect from the compact disc format in terms of transparency, palpability, resolution of fine detail, and rightness of timbre—particularly with HDCD-encoded discs.

I can't stress enough how important system setup is with the SDR-2000 Pro. The processor must have many conditions in its favor for it to achieve the level of performance I've described. Proper routing of digital cables in relation to the AC-power cord and DC ribbon cable, for example, can mean the difference between stunning musical performance and merely good sound. Further, the SDR-2000 sounded its best when used with the Spectral electronics and MIT cables.

Don't let the SDR-2000 Pro's $8195 price fool you: this is the state of the art in digital playback, and my first choice in digital processors—regardless of cost. What more can one say?