Recording of May 1991:Dick Hyman Plays Fats Waller

DICK HYMAN: Dick Hyman Plays Fats Waller
Dick Hyman, piano
Reference Recordings RR-33DCD (CD*), RR-33CD (CD**), RR-33LP (LP). Keith O. Johnson, recording eng.; Robert Harley, mastering eng. (RR-33DCD); J. Tamblyn Henderson, Jr., prod. D/DDD/AAA. TTs: 59:28,* 59:22,** 59:22

The original release of Dick Hyman Plays Fats Waller was the world's first—and, to my knowledge, still only—direct-to-digital CD (footnote 1). Available at a premium price in a deluxe numbered edition limited to 25,000 copies, it has become one of Reference Recordings' most popular releases, and is expected to sell out by the summer of 1991. The reproduced piano sound is just about the most realistic I've heard in my system. The sonic perspective is intimate rather than distant, so that the effect is more piano-in-the-listening-room than you-are-in-the-concert-hall, with a most pleasing ambience surrounding the instrument. The sonorities of the Bösendorfer reproducing piano ring out with tremendous power and clarity, from thundering bass notes to crystalline treble.

In my review of the original release (Vol.13 No.5), I suggested that it would be interesting to compare the direct-to-digital release with the conventionally mastered CD and all-analog LP when they became available. Well, they're here now, and the comparisons are indeed interesting, although not as straightforward to interpret as one might think.

As usual in these sorts of comparisons, I wanted to keep all potentially confounding variables constant; thus, the new CD, like the original, was painted (outer edge and inner "trough") with AudioPrism's CD Stoplight, and had a Sims Reference Band placed around it. Even before I played the two discs, I noticed an obvious physical difference: the original "D" release has gold reflective material, whereas the new "DDD" has silver. I don't know if there is any convincing evidence that gold backing "sounds" better than silver, but some record companies certainly claim that it does. Furthermore, the discs come from different mastering plants (Disctronics vs Disc Mfg., Inc.), which can itself produce sonic differences. What I'm trying to say is that this is not just a comparison of direct-to-CD vs digital-tape-to-CD, but also of CD manufacturing methods. In fact, there are additional differences unrelated to the utilization of an intermediate storage medium: the new CD does not derive from the same real-time performance but an earlier one (repeated "performances" on the B;dosendorfer Reproducing Piano are, in theory, identical, but producer Tam Henderson notes that the piano was in somewhat better tune during the earlier take), and there were some changes in balance (0.4dB to the left), level (0.8dB higher), and equalization (1.0dB up at 16kHz) for the new CD (footnote 2).

Putting aside all these caveats about the interpretation of sonic differences, how do these recordings sound? First, I think it's fair to say that, by any reasonable standard, "stunning" still describes the sonic quality of all three releases. If you can't have the Bösendorfer Reproducing Piano and its associated digital hardware/software (or perhaps Dick Hyman himself) in your listening room, here's the next best thing. The D release sounds a touch warmer in tonal balance, with the bass notes having a little more apparent weight. The piano on the DDD sounds as if it's a bit closer and the upper registers ring out with more clarity (probably a function of the equalization and level changes; I'm told there was no change in miking). Overall, I prefer the D, but by a very small margin.

Listening to the LP after having familiarized myself with these state-of-the-art CDs proved to be an fascinating experience. As one might expect, noise level is higher, but not annoyingly so. Tonal balance is more like the DDD than the D, at least in my system (footnote 3). Does the LP sound any better? In a word, yes. The sound of individual hammers striking the strings, the decay of resonances, the ambience surrounding the instrument, the pace and drive: they're all subtly but significantly superior to either CD version. To quote Sam Tellig (or is it Gertrude Stein?), there is more there there. I would be quite happy listening to any of the three versions, but my feeling is that if I could persuade the Bösendorfer people to lend me the Reproducing Piano (and assuming that I could get it up the stairs and squeeze it into the listening room), it would sound most like the LP.

Oh, yes, the music. It's wonderful! Dick Hyman Plays Fats Waller is quite unlike "audiophile" records of yore, where superlative sound quality was lavished on inferior music/performances. Thomas "Fats" Waller was arguably the most prolific of the composers/performers of his era, and the selections here include the cream of his abundant crop. In some of the numbers (eg, "African Ripples"), Dick Hyman's performance is very close to Waller's own; others (eg, "Honeysuckle Rose") have parts that sound more "modern," but still idiomatic. In terms of piano technique, Hyman—aided by, as he acknowledges, the editing capabilities of the reproducing piano—is actually superior to Waller, and the Bösendorfer's powerful bass register provides an impressive highlighting of the "striding" left hand.

At times, Hyman's technique seems almost too precise, so that it does not quite have the feeling of spontaneity and playfulness evident in Waller's own recordings. This may be related to the one possibly controversial aspect of this project: use of the reproducing piano, which digitally samples and records keyboard and pedal action, then recreates the original movements with motors. The results certainly don't sound mechanical or player-piano–like, but I wonder if the reproduced performance is as indistinguishable from the original as they claim.—Robert Deutsch

Footnote 1: The digital data resulting from A-to-D conversion of the microphones' output were sent directly to the mastering facility via microwave link, where the CD master would be cut in real time, with no intervening storage stage.

Footnote 2: Confession Time: when I first listened to the two CDs, I was under the impression that they were based on exactly the same microphone feed. The sonic difference was such that I found this hard to believe, so I called Reference Recordings for confirmation. I was told that not only were the recordings not identical, but that all the information about the technical differences was in the liner notes! If all else fails, I suppose one might as well read the manual.

Footnote 2: Within the past few months, I've brought my analog front end considerably closer to Class A status: first, replacing the Talisman S with an AudioQuest AQ 7000 (a Class A cartridge if ever I heard one); then, just in time for the present LP-vs-CD comparison, getting the Lingo power supply for the Linn (expensive, but well worth it, even if you're already using an AC line conditioner like the Tice Power Block). On the digital front, I'm looking forward to the coming upgrade for the Aragon D2A; of course, it's retrofittable to existing units, thus allowing one to avoid the dreaded "Digital Hell."