Digital Processor Reviews

First, the Theorem had easily the best soundstage of the four processors. The feelings of space, depth, and air was excellent for any digital processor, never mind one costing a hair under $800. There was a distinct impression of individual instruments, each surrounded by air, hanging in three-dimensional space. The presentation was the antithesis of flat sterility, in which images fuse into a congested mess. Instead, I heard a remarkable openness, front-to-back layering of images, and a wonderful sense of bloom surrounding image outlines. I had this impression within seconds of hearing the first music through the Theorem (Robert Lucas's Usin' Man Blues, AudioQuest AQ-CD1001), an impression repeatedly confirmed over extended auditioning. The sense of the guitar and vocal appearing in the listening room, surrounded by the recorded acoustic, was palpable. Moreover, the Theorem had a remarkable transparency, which, combined with the excellent resolution of spatial cues, created an engaging "you are there" quality. In this respect, it had the most analog-like presentation of the group.

The Theorem's overall tonal balance was a little tilted toward the treble for my taste. Although not bright or hashy, the treble tended to be a bit forward and overly analytical. Cymbals, percussion, and other transient instruments with lots of high-frequency energy bordered on becoming etched. Perhaps contributing to this impression, the Theorem seemed to lack warmth in the mids. Robert Lucas's guitar lacked the sense of body and fullness heard through the PS Audio Digital Link, and particularly the CAL Sigma. This could tend to make the treble a little threadbare. There was, however, a good sense of liveliness and detail; the Theorem had a feeling of vitality and life missing from the CAL Sigma.

The other aspect of the Theorem's presentation that grabbed my attention was its very strong sense of rhythm, pace, and drive. The unit had an "upbeat" quality that made me tap my foot. Music was infused with a greater rhythmic propulsion and energy in comparison with the Sigma and Digital Link II. This quality is usually related to a component's bass presentation, but in the Theorem's case, I attribute it to the processor's excellent dynamics. The Theorem's ability to portray music's fine dynamic structure was superb, easily the best of the group. The processor was particularly adept at conveying the speed and zip of transients. Drums had a steep leading edge and good impact, although the Theorem was better at conveying microdynamics than sheer slam and weight.

Bass presentation was tight and well-defined, but slightly lean and dry. The Digital Link II had more warmth and roundness, but at the expense of "speed" and agility. Bass guitar lines had less "purr" and weight through the Theorem than any of the other processors, but pitch definition was excellent. The Theorem was tighter and faster than the Sigma, but lacked the Forté DAC 50's weight, greater pitch definition, and deeper extension. Nevertheless, the Theorem's bass was more than acceptable in a processor of this price.

Another favorable aspect of the Theorem's presentation was the lack of grain in the mids and treble. The upper octaves were surprisingly clean and free from coarseness. Female vocal—Sara K.'s Closer Than They Appear (Chesky JD67)—had a purity and silkiness not heard through the other processors in the survey. Similarly, massed violins had a smoothness reminiscent of much-higher-priced processors.

It's interesting to listen for the differences in reproduction of analog tape hiss between digital processors. The differences can be large, often reflecting the processors' musical characteristics. The Theorem seemed to "whiten" tape hiss more than the other three processors, but its texture was very smooth—just like undigitized hiss. Some processors coarsen the hiss's texture, making it sound granular and rough instead of smooth and even. The Theorem's reproduction of tape hiss correlated with the musical impressions: slightly up-tilted treble, lack of midrange warmth, and very little grain.

Resolution of detail was superb; the Theorem clearly presented the most information to the listener. In this respect, the Theorem had more in common with the high-priced processors. The music's fine inner detail and subtlety were resolved, greatly adding to the listening experience. The nuances in Robert Lucas's vocals revealed a greater degree of expression through the Theorem. Although the treble detail was slightly etched for my taste, it was lively, open, and airy.

Overall, I was very impressed by the Theorem. In the area where it clearly excels—soundstage size, focus, and resolution—the Theorem was superb. It had the most "analog-like" presentation, devoid of the flat, sterile "cardboard-cutout" images endemic to low-cost digital. The Theorem was my favorite of this group of four processors and an excellent value at $799. Some listeners, however, may be willing to sacrifice the Theorem's higher resolution and superior soundstaging for the Sigma's more relaxed presentation. A personal audition is mandatory, therefore.

Highly recommended.

Putting it all in perspective
After getting to know each of the processors, I auditioned them in relation to the $1495 Bitwise Musik System Zero (favorably reviewed last month) and $399 Audio Alchemy Digital Decoding Engine. Where does the Sumo Theorem fall compared with a $1500 processor and the inexpensive DDE?

Starting with the DDE, it was no contest; the Theorem was superior to the DDE, particularly in the treble. Although the DDE had good clarity and resolution—greater resolution than the Digital Link II—its hashy treble was a significant liability.

The Theorem was closest in sound to the excellent Musik System Zero. Both threw a spacious, three-dimensional, precisely focused soundstage, but the Zero was clearly the better processor. The $1500 unit had a warmer balance in the mids, weightier bass, a more delicate treble, and greater soundstage transparency. The Theorem, however, acquitted itself well against a processor nearly twice its price.

Finally, I couldn't help comparing the Theorem to a processor that is, in my opinion, the processor to buy for under $1000—the Meridian 203. Despite improvements in low-priced digital, the 203 has maintained its competitiveness. I should add that the 203 I compared to the Theorem is the older version, not the new model with the Philips DAC7 chipset.

The Theorem was clearly better at resolving space than the 203, with a more palpable soundstage and greater transparency. Bass presentation—pitch definition, pace, tautness—were superior to the Theorem's. The Sumo also had a more laid-back quality in the mids, sounding almost recessed compared to the 203. One aspect of the Meridian I did prefer was its overall tonal balance. It was warmer, richer, and more full-bodied than the Theorem, with a fully fleshed-out midrange. Nevertheless, I felt that overall, the Theorem was the better-sounding processor.

Conclusion
I was generally impressed by the Theorem's price/performance ratio. The Sumo Theorem was in many ways the best-sounding converter in this survey. Its soundstaging abilities and resolution of spatial information were superb. The Theorem also had an excellent sense of pace and liveliness, qualities that made it musically compelling. Its slightly lean and up-tilted tonal balance, however, may not suit all listeners. Some may prefer the CAL Sigma's warmer, gentler presentation over the Theorem's greater resolution of detail and superior soundstaging. The Theorem is the processor I'd buy out of this group. In fact, it gets my vote for the best-sounding converter under $1000.

The bottom line? There are musical digital processors that don't require a second mortgage. Moreover, this current crop provides better performance than yesterday's bargains—the Rotel RCD-855 and Audio Alchemy DDE—in my opinion. If you own a CD player and are thinking of upgrading to a digital processor, one of the three processors recommended may breathe new life into your CD collection.