Burmester Reference Line 970 D/A converter & 969 CD transport Page 2
What I heard in the hall was something else: The balance was less even than from the recording, but the senses of space and of air around the instruments were much more apparent. If I switched over to the 24/96 DVD version (CHDVD194) and set the 970 to MaxRez, I could restore those attributes to the sound from the mix. I tip my cap to David Chesky and his crew: The 24/96 disc, when given the 970 treatment, captures the best of the live event and the two-channel mix. (I went to a Chesky press launch for the recording, at which Mangione and band played several selections. Lots of fun, but they had speaker and hum problems. It didn't sound as good as either disc!)
As one might expect from my experience with the Mangione recordings, the 970 was superb at delineating the differences between 16/44 and 24/96 discs of the same material. My favorite, still, is John Lee Hooker's Mr. Lucky (Charisma/Pointblank 91724-2 vs Classic Records DAD 1007); the high-resolution disc simply had more space and more slam, while Hooker's stubbly growl dominated front and center.
As for the other options, I found negligible difference between the Standard and the MaxRez filters, but more useful differences between the filters' Sharp and Soft settings. As a result, I rarely used the Standard filter, choosing between the PMD-100 for most CDs, especially HDCDs, and the MaxRez for 24/96 sources. The PMD-100 sounded much like the MaxRez in Soft mode. These settings made for a relaxed presentation, with a slightly less etched upper end than the MaxRez filter did in Sharp mode. Some of the Classic Records DADs derived from older analog sources benefited from the MaxRez in Soft mode. Nonetheless, with the best 24/96 discs, I definitely preferred the MaxRez filter's Sharp setting. In fact, the MaxRez/Sharp setting became my default, unless there were infelicities in the recording (or HDCD discs to play) that required reconsideration.
The 970 was so revealing with all source material that switching from another DAC to the 970 provided a change similar to, if much less pronounced than, switching to a higher-resolution source. Can't wait to hear Diana Krall's All For You (Impulse! IMPD-182) in 24/96? Well, the Burmester 970 gave me an idea of what to expect: tighter and fuller bass, more stable imaging (not the strongest feature of this disc), a sense of instruments in a real space, and, best of all, a more liquid, more palpable presentation of Krall's smoky tones. The most consistent observation, however, was of less noise, despite the fact that none of the DACs used for comparison were noisy. It seems to me that, as one goes to playback devices of progressively higher resolution with better source material, one's ears are able to discern finer details that, with lesser equipment, are lost in the noise. Thus, although there isn't less noise, there is more information that the listener's brain can use to separate the musical and ambient sounds from the noise due to the recording and playback process. The result is what I would describe as a more relaxed presentation, since it is easier to comprehend the music, quite apart from issues of tonal balance, dynamics, or the explicit depiction of detail. So far, the Burmester 970 is better at this than anything else I've heard.
But I was less than impressed by the Burmester 970's volume-controlled outputs. Burmester says that they provide the volume control as a convenience, and I agree with their preference for the use of a line preamp with the 970's fixed outputs. Convenient and remote-controllable as the variable output may be, it is out of character with the 970's otherwise outstanding performance. Feeding the power amps from it, the 970's small but delicious advantages recede to inconsequence. Feeding the Sonic Frontiers Line 3, the volume-controlled 970 was simply one of the guys in this elite group. I cannot say if a simple pot would have been better, but the 970 deserves better.
Against the mbl & Mark Levinson
I toiled in vain to find a good analogy to help me characterize the distinctions among the Burmester Reference Line 970, Mark Levinson No.360, and mbl 1611HR digital processors. First, I thought of The Three Tenors. It seemed appropriate to equate the mbl 1611HR with Luciano Pavarotti's glorious ring and imposing presence, and to associate the Burmester 970 with Plácido Domingo's vocal accuracy and dramatic tone. However, it would be wrong to equate the excellent and future-proof Mark Levinson No.360 with José Carreras, a great tenor but one who once was even better.
Next, I thought that the three bears of Goldilocks' acquaintance might suit. Describing the Mark Levinson No.360 as Baby Bear works, since each is the smallest in its category and seems to have found an ideal balance of taste and performance. In addition, the No.360 (due to its reprogrammable DSP) and Baby Bear (due to its youth) probably have their best years still to come. The problem is in choosing which parent bear to represent the mbl, which the Burmester. Papa Bear's oversize chair, hard bed, and hot porridge might match the mbl's large chassis, taut dynamics, and vivid performance. However, Mama Bear's small chair, soft bed, and cold porridge are ill-matched to the Burmester's substantial construction, solid bass, and neutral balance. And I won't even attempt to grapple with the gender issues this analogy might imply!
So, falling back on a more conventional approach, I thought each of the three DACs to be a successful but idiosyncratic statement of how, late in the Sony/Philips "Red Book" era, a DAC should be made. Each is different in the design aspects that are given emphasis, in its physical execution, and in its analog depiction of the digital bitstream. After all, you can drive a Ferrari, a Mercedes, or a Porsche from point A to point B, but the three experiences will not be the same. And a Honda will get you there just as reliably.
The Burmester 970 is 20dB$ more expensive (footnote 1) than the MSB LinkDAC and 8.7dB$ more than the Levinson; now we're way out on the return-on-investment curve, where high prices can buy only subtle improvements. However, if I had enough pennies, I would regard the Burmester 970 as worth every one of them. Regardless of the configuration, the 970 seemed to have less character than all the other DACs on hand. Less immediate but more subtly detailed than the mbl, slightly more spacious than the Levinson, and possessing as much slam and detail in the bass as the dCS Elgar, the Burmester is simply the finest-sounding (and, I suspect, least-sounding) DAC I have used to date. Its flexibility lets it be tuned just enough for optimum performance with almost any source, and one can step quickly through one's programmed configurations to find that optimum arrangement.
But the Burmester 970 is a tough sell. While I preferred it to the others, the increment of its superiority is quite small—real, but small. Having just purchased a new car, I simply cannot justify buying a DAC for a similar price. If you can, I envy you.
Aside from the volume control, is there anything wrong with the Burmester 970? Not really. I haven't used another DAC that was as consistently neutral and enjoyable with all source materials. In addition, you certainly get a heap of technology—the multiple filters, slopes, and memories offer the flexibility to "tune" the 970 for various sources. The only caveat is that, when compared to other top-grade DACs, the 970's sonic superiority will not bowl you over, however significant that superiority may be to the uncompromising. Thus, despite the 970's small but consistent advantages in detail, ambience, and tonal integrity, it's hard for me to justify its cost over the dCS Elgar, or even the Mark Levinson No.360.
But that's my problem.
Footnote 1: Since we are dealing with exponential differences in price, I conjured a new unit of measure, the dB$, to help my deliberations. The dB$ is equal to 10log(price1/price2). I used a multiplier of 10, not 20, because money is power, not merely potential. So, if we make the MSB LinkDAC our reference, the Levinson No.360 is +11dB$, the mbl 1611HR is +16dB$, and the Burmester 970 is +20dB$. Note that the Levinson is merely 1.8dB$ more than the Camelot Uther, and the Burmester is less than 4dB$ more than the mbl! These dB$ values seem to offer a better perspective on the nonlinear relationship between performance and price.—Kalman Rubinson