|Site review by Christine Tham April 9, 2007
|I have been eagerly anticipating this recording ever since I found out about it (almost by accident, whilst surfing the net).
Legendary Canadian pianist Glenn Gould's first recording was him performing Bach's Goldberg Variations. It was recorded in 1955 when Glenn Gould was 22 years old, virtually unknown to the public except his family and friends (and those privileged enough to attend his critically acclaimed New York debut earlier in the year, as well as Columbia Records, who snapped him up in a recording contract purely on the strength of his Manhattan Town Hall performance). When it was released in 1956, the album quickly became a bestseller and forever changed how we play and listen to Bach keyboard music.
Even today, more than 50 years later, the performance sounds fresh, innovative, inspiring, and absolutely breathtaking. Anyone who has ever tried playing the Goldberg Variations will be amazed by the technical virtuosity displayed. Sometimes the notes flowed so fast they made the piano sound like a different instrument. And yet every note was precisely articulated, every polyphonic part so clearly nuanced to reveal new insights and perspectives into the music.
But it wasn't mere technical prowess that made this recording so beloved by Glenn Gould's fans. Never before had any performer so completely fused a musical piece into his personality, turned it upside down and delivered an end result that was as much Glenn Gould as it was Bach. At times stormy, at times pensive, each variation seemed to have a personality of its own. Some were playful, others moody, a few were exquisitely beautiful.
Glenn Gould's "James Dean"-like looks and wild dishevelled hair complemented the audacity of the music and supports the "Rebel with a cause" aura and mystique that developed around him. Soon there were wild stories and rumours about his "peculiarisms" that cemented his public persona and image. Stories about a man who wears mufflers and gloves in the heat of summer, who soaks and heats his hands in scalding hot water prior to playing, who even brings his own folding chair to concerts and recording sessions (and later on, his beloved Steinway "CD 318" piano). Eventually, some of the rumours even turned ugly (there were claims that he was "addicted" to various medications, and some even accused him of "faking" the 1955 performance through tape splicing and editing) and after 1964 he "abdicated" from giving live performances altogether, and became a Howard Hughes like recluse, communicating mainly via lengthy phone conversations with his close friends, and to the public only through his recordings, which he continued to release right up to his death in 1982 of a stroke.
But back to the 1955 recording, the one that started his career. Many treasure it as one of their favourite albums, and its popularity is such that it has never been out of print. Indeed, just on CD alone, there are several versions available (the original CD mastering, the Super Bit Mapped remastering in the early 90s, the version on the Original Jacket Collection, the version on "A State of Wonder", ...). I know - I own all the ones mentioned, plus the LP. There's even a SA-CD of the 1981 recording (Stereo DSD layer only), which is one of the earliest titles released by Sony Classical to promote the format.
Good though the recording is across these different versions, its age and limited fidelity bothers some people. First of all, it's a mono recording, with a limited soundstage. It's also rather bloomy in the midrange, with rolled off highs and lows. And some people are annoyed by the background noises (the creaking of the folding chair, Gould's sharp intakes of breath, foot tapping and occasional moan and groan). Finally, the tape hiss is so pronounced one could almost drink it, or maybe even drown in it.
And this is where this new release comes in, and why it is so potentially exciting. A small company called Zenph Studios claims they have perfected a method of "reverse engineering" a recorded musical performance back into the individual notes (including precise details about not only what notes were played, but their exact moments in time and duration, and dynamics). The output is a computer data file (in a proprietary "high resolution" extension to the popular MIDI format used to program synthesizers) that in theory captures all the nuances of the original performance, but now can be "replayed" or "reperformed" on a suitable musical instrument capable of interpreting this data.
As it so happens, Yamaha produces a line of acoustic pianos (called the "Disklavier") that are effectively like normal pianos except the keys can be mechanically actuated by feeding it enhanced MIDI data. This hybrid multi-channel SA-CD released by Sony BMG contains a "reperformance" of Gould's 1955 recording, lovingly analyzed and captured by Zenph, realized on a modern Disklavier Pro piano, recorded at the Glenn Gould Studio in Toronto directly into DSD using a Sonoma digital audio workstation, nearly 25 years after Glenn died.
What is on disc is actually three different versions of the same performance. Five microphones are used to capture the 5.0 multi-channel version, recorded directly onto DSD with I imagine little or no post-processing. The front left and right channels are then used to create the stereo version (on both DSD and CD layers) with no down-mixing. Finally, an additional set of omnidirectional microphones (built into the "ears" of a dummy's head) were simultaneously used to create a binaural (headphone) version (available as tracks 33-64 on the CD and stereo DSD layers of the disc).
Some of you may be a bit skeptical. Can a computer analysis, no matter how sophisticated, really capture (and reproduce) the essence of a human playing the piano? Even if it could, wouldn't it sound different when played back on a different piano from the original recording?
Others may miss the background noises present on the original recording. Would a version that was totally devoid of extraneous noises sound sterile?
I was interested in exploring possible answers to these questions, so it was with a lot of excitement that I requested and received a pre-release review copy of the SA-CD directly from John Q. Walker (President of Zenph Studios). The title itself has just been released in Japan, but is scheduled to appear in the USA in late May, and at this stage it is unknown whethere it will be available in Australia.
First of all, let's talk about the recording itself. It is technically flawless and perfect, and the direct to DSD technique captures the sound of the Disklavier Pro in amazing, mouth watering detail. The multi-channel version sounded exactly as if there was a real piano in our listening room. Every note came through with perfect clarity, including the harmonic overtones of the piano's timbre, and the reverb tail of each and every note decays wonderfully and naturally into the ambience. It's probably worth buying this recording just to hear how incredibly realistic the piano tone is.
In the multi-channel version, all three front speakers are used effectively, but the surround channels are very restrained and mostly used for ambience. The stereo version sounds nearly as good, with a very wide and deep stereo field, but ultimately doesn't "breathe" as well as the 5.0 version. The binaural version sounds a bit weird when played back using loudspeakers, but sounds amazing on headphones (almost as if I was Glenn Gould himself, hovering over the keys - although the positioning of the dummy's head is slightly higher than Glenn Gould's normal stooped and hunched piano playing posture).
But what about the performance? Is it the "real" McCoy, or a pale imitation of Gould?
Well, the answer is yes and no. It certainly doesn't sound "artificial" or "unnatural": the piano playing is expressive and idiomatic, and any Gould fan would instantly recognise the performance as authentically "Gouldian."
But after listening to both the original recording and the reperformance back to back (the version on "A State of Wonder" compared to the CD layer, on the same player), I feel there are subtle differences between them.
Certainly the versions of the Aria sound remarkably similar between the two recordings, so similar that I was hard pressed to detect any differences.
But in fast passages, the original recording tends to blur the notes so that they kind of run into each other, creating a breathless sense of anticipation as the ear tries to follow the melody. It somehow made the performance more exciting to listen to, not dissimilar to the feeling of being in a rollercoaster. Good examples are of course the last few variations, ie. Variations 26, 28 and 29.
On the re-performance, the individual notes seem to stand out more clearly by themselves. Perhaps this is due to a different piano being used, or perhaps it's due to the higher fidelity of the recording, or perhaps a combination of both. But to me, having the notes stand out more distinctly as individual sounds made the music flow less smoothly, and I don't seem to get as much of a "rush" as I do on the original. The sense of "unreality" that I get on the original version (that the notes almost sound impossibly fast, as if the tape is being played too fast, almost making the piano sound like a completely new instrument) is noticeably absent in the new recording.
I seem to also notice slight differences in the stress of the notes. Perhaps it's my imagination, but occasionally a stray note would "feel" a bit out of place, enough that I would notice it. Sometimes, the character of a phrase would sound a bit different between the two versions.
I don't mean to imply that the original version is superior and that the new version is somehow lacking. Rather, both versions seem to be slightly different interpretations, realized on different pianos. Sometimes I actually get more enjoyment out of the new version (for some reason, the contrapunctal and supporting musical lines seem to come through more clearly in the new version, and made me enjoy them more since they are more distinct). At other times, the original version seemed more subtle. For example, in Variation No. 10 (Fughetta), each new repetition of the fugal theme seemed to have a character of its own in the original version, but in the new version the playing seemed more homogeneous.
Finally, the piano tone was very similar between the two recordings, I wonder if the Yamaha Disklavier was tuned in any way to sound more like the Steinway used in the original recording. However, the piano tone also reminded me of the timbre in Glenn Gould's 1981 re-recording of the Goldberg Variations (where a Yamaha piano was used).
So, in summary, I don't believe that both recordings are identical in terms of the emotional impact of the performances, even though I am willing to concede the notes and playing style may well be identical. The re-performance sounds very similar to the original 1955 version, but curiously inherits some of the characteristics of the 1981 recording.
I wonder what Glenn Gould would have thought of the recording, if he is still alive today. On one hand, we know that he was fascinated with recording technology, and it is certainly tempting to imagine him being very excited by the concept, and working hand in hand with the Zenph engineers to tweak the captured data.
I don't think Glenn would have regarded the process as being "artificial" or "dishonest." Indeed, I would like to quote from one of his articles ("Music and Technology", originally published in the Winter 1974-75 edition of Piano Quarterly) that seems pertinent and relevant to this re-performance:
'Technology, in my view, is not a primarily a conveyor belt for the dissemination of information; it is not primarily an instantaneous relay system; it is not primarily a memory bank in whose vaults are deposited the achievements and the shortcomings, the creative credits and documented deficits, of man. It is, of course, or can be, any of those things, if required, and perhaps you will remind me that "the camera does not lie," to which I can only respond, "Then the camera must be taught to forthwith." For technology should not, in my view, be treated as a noncommital, noncommitted voyeur; its capacity for dissection, for analysis - above all, perhaps, for the idealization of an impression - must be exploited, and no area with which it is currently occupied better demonstrates the philosophical conflicts with which its practitioners and theorists have been too long preoccupied than the aims and techniques of recording.'
There you have it.
On the other hand, Glenn Gould was also known for being somewhat disparaging and dismissive of his 1955 performance. Perhaps he may not be that keen on this re-performance of what in his eyes may be an embarrassing reminder of an "immature and foolish" youth.
However, regardless of Glenn's own value judgement of his own performance, it is true that many people (including myself) prefer this performance over his more measured, and stately 1981 version. And it is to us that this re-performance is aimed at.
My final conclusion is that this is a recording worth owning, and true to the spirit and ideals of Glenn Gould. It may not be an exact replica of the 1955 recording, but it is a valid performance in it's own right, and takes it's place alongside the 1955 and 1981 recordings with pride.
|Review by AmirKessner June 9, 2007 (11 of 15 found this review helpful)
|Don't let anyone fool you this re-performance is anything much like Glenn Gould's. It doesn't even sound like any living pianist attempting to ape Gould – any living person would focus on emulating the ESSENCE of a Gouldian performance, but this is exactly what is missing here. Missing are the special touch, attack, bite, vitality, humor, the unique Gould CHARACTER. You're left with a well recorded, civil, elegant, mellifluous playing, which may match Gould's TIMING to a millisecond, leaving all other aspects out. The very character is lost.
Not that the technical aspects, software and hardware, of Zenph's re-performance are less than astonishing. You can think of it as being able to reproduce, without sampling or recording, a man's speech, with all the pitch and timing aspects, but without capturing the unique expression, insinuations, tone colors, etc., or, in short, character. As the booklet admits, "the process is still in its infancy, really".
The recording, sound-wise, is as good as Sony can give us, which is pretty impressive. I have listened to neither the regular stereo nor the binaural version.
Recommended only for the overly curious and/or those wishing (justifiably) to support Zenph's R&D. The potential of their technologies is, indeed, great!
Was this review helpful to you?