|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
73.222.105.47
In Reply to: RE: Oh my... posted by Ivan303 on July 11, 2016 at 19:11:22
Follow Ups:
I was just commenting on how long it was gonna take me to get through ALL of that Debussy if, as Rob claims, she plays the stuff REALLY slow.
I haven't listened to the entire set. And it does sound like a modern piano. I look forward to comparing it to the 1890 Erard. Unlike genungo, I like Satie and his eccentric, quirky humor.
And yes, she takes her time!
If you like Satie I think you will enjoy her take.
It's a great recording, and who knows?
Maybe even Chris will like the 1890 Erard? ;-)
. . . although that Zenph Rachmaninoff "Re-performance" album on Sony uses a restored 1904 Steinway AFAIR - not bad at all, especially via the multi-channel download.
From Satie to Rachmaninoff, and from an 1890 Erard to Zenph multi-channel download. Can Zenph be programmed to post pictures of babes and complaints about serialism? If so ...
"From Satie to Rachmaninoff" - Yup, bear with me. . .
"and from an 1890 Erard to Zenph multi-channel download." Uh, yeah - except that the logical progression was "and from an 1890 Erard to a 1904 Steinway" (which happens to be available as a multi-channel download - just like the 1890 Erard recording).
QED. ;-)
But so long as we're changing the subject from Satie and Ms. Ogawa, I will say that the Zenph Art Tatum did not impress me at all. It just did not sound like the real Art.
Nevertheless, I see Steinway now owns Zenph and is using the technology to make an electronic player piano. So maybe my comment about you becoming obsolete has become the real topic of this thread. ;-)
I inquired as to what degree that progress was hands-off computer number crunching and to what extent there was human intervention, and the answer I received was not at all satisfactory.
On many modern, computer/MIDI-based piano playback systems, there is the ability to "go back in" to the command files and literally clean things up--move events in time, change the nature of an event, etc.
My position on the matter is, sure, Rachmaninoff cut some piano rolls. But unless there is a computer program that takes what is on the piano roll and translates it to a modern playback piano system with no human intervention, then... we are hearing some fallible human's IMPROVEMENT or RE-INTERPRETATION of a piano roll.
Which I don't have a problem with. What I do have a problem with is the PR claim that what we are hearing is 100% Rachmaninoff.
Chris, what is your take on that?
John
So Zenph "re-recordings" are derived from actual recordings (mostly 78's, although their first release was the Gould Goldbergs from the LP mono era), which are subjected to computer analysis and processing which outputs a file which can be played on an actual piano from today, which can then be re-recorded without all the noisy artifacts (hiss, surface noise, etc.) and limited frequencies and dynamics of the original older recordings. I bought three of them: the Gould ('55) Goldbergs, the Rachmaninoff (in my post above), and the "Spanish Masters" album where Granados and de Falla "accompany" Isabel Bayrakdarian and Zuill Bailey. I consider the last of these kind of hokey (despite the fact that it has the best average rating on Amazon of the three Zenph classical releases - go figure!) because, far from accompanying or collaborating with with the cellist and the singer, the pre-existing recordings of Granados and de Falla place their collaborators in rhythmic and dynamic straight-jackets - just the opposite of an actual collaboration. (Kind of like playing with the old Music Minus One discs!)
The Gould recording was frustrating because the Zenph processing was not really needed - that '55 technology, although mono only, was just not that bad! Another issue to many listeners (but not to me) was that the Zenph "re-performance" was made on a Yamaha piano, whereas Gould used a Steinway on the original.
I guess Zenph took this criticism to heart, because they used a 1904 instrument for their Rachmaninoff release. I was originally very disappointed in this release (on CD - it was never available on SACD), because I felt that the piano was lacking in power and its tone quality just seemed wan. Subsequently, a 24/96 quad download was made available (on the iTrax site and other sites), and my listening to this incarnation changed my opinion of the sound quality and the success of the processing rather dramatically - another instance where one's perception of the performance itself can be changed by how well the engineering on the recording has been executed. I also must say that criticisms of this recording, such as Farhan Malik's one-star review on Amazon, strike me as surprisingly naive: Malik goes into conniptions over how the performances on the Zenph recording do not have the same "character" as the ones on the original 78's. Well duh!!! Of course it's not going to sound exactly the same: the Zenph recording is of a different piano (even one it's one from 1904), in a different acoustic, recorded with a different microphone, etc., etc., etc. How could it NOT sound different? Nevertheless, it still bears the stamp of Rachmaninoff's playing (at least as I understand it from having heard all of his 78's).
To me, the extant Zenph catalogue represents the squandering of an opportunity, in the sense that every one of their three classical releases has some strike against it which has nothing to do with the Zenph process itself. Nevertheless, I hope that the Zenph process will continue to be developed and refined, resulting in the release of many more historic recordings which IMHO need this kind of technology. If I ran the company, I would:
- choose VERY old recordings which need the Zenph technology the most (IOW, something recorded in 1925, not 1955)
- choose the very best sounding modern pianos to "re-record" with, and not bother with antiquarian fetishes
- make all of my releases available in hi-rez, multi-channel technology (whether discs or downloads)
So, despite all this verbiage, John, I see that I haven't really answered your question, which was whether I approved of human intervention in the processing and release of these kinds of recordings - processing and "intervention" which could make them even better than they originally sounded (or even better than the original player executed them!). Like you, I don't have a problem with any of this, but also like you, I would like the process to be totally honest and transparent. The companies should let their potential buyers know what each process entails. And the claim that we are hearing, say, 100% Rachmaninoff via this kind of processing should contain enough weasel words so the buyers know what's up! In a way, this is not so different from Madeline's (my wife's) wish that all recordings should come with some basic information, such as how many edits were used in the final master - IOW, truth in advertising!
OTOH, we know that some piano rolls themselves (I'm back to piano rolls again) were subject, even at the time, to ex post facto processing which evened out the timing, etc., and made the piano roll performances even more super-human sounding than the real performances from the celebrated artists who recorded them! So, even as we bemoan modern artists' and engineers' addiction to editing, we should also remember that, in some respects, it has always been thus! ;-)
I'd still rather have a clean original or a brilliant "original remaster". Or perhaps better yet, something completely new.
But to each his own, I guess.
I did buy the Tatum Zenph (but mostly because I wanted to hear it in binaural).
Thanks for the exhaustive reply though. As usual, you have shown that really know your stuff!
I've only heard the Zenph Glenn Gould and Art Tatum. As for Gould, I think you're exactly right -- it wasn't really necessary, though it sounded rather good to me. And didn't the real Glenn Gould switch to Yamaha at some point? But the Art Tatum was not as successful to me, somehow stilted and artificial. Maybe that's my prejudice, though I didn't have that feeling about the Gould.
It really goes back to John Marks' question -- aside from taking out the clicks and pops, just what does the Zenph program do, and how accurately does it reproduce the music without introducing estimations or approximations? Of course, introducing clarity and full, distortion-free dynamic range may be worth some sacrifice, but just what if anything is being sacrificed in the process?
First of all, thanks, Chris. Yes, I was asleep there, my memory misled by Tatum and Rachmaninoff. When years ago I asked the question and later discussed the response I received with JA, I knew that it was based on processing an audio recording, but, the question is the same:
Does the algorithm call all the shots, or, is there human intervention?
I am not at all an expert on piano technique, but, I do know that there are many subtle things a pianist can do in live performance (such as silently pressing a key down so that those strings can ring harmonically) that pre-hi-fi recordings might not catch. Just an example off the top of my head, and it might be faulty or false.
rbolaw, my understanding was that late in his career, Gould snitted at Steinway because he reportedly thought that his wishes were not being acted upon promptly. I do know that the 1982 Goldbergs "reconsideration" (not so if you know the 1954 CBC acetates) was recorded on a Yamaha.
BTW, I have never received a satisfactory answer to the presence of the "mouse orchestra" buried in one variation of the 1982 recording, but, one hypothesis is that it is leakage from the machine room where someone was rewinding or fast-forwarding a tape.
ATB,
John
Funny how one uses the Hoboken numbering and the other uses the Christa Landon numbering. ;-)
But, like you, I was favorably impressed with the Zenph Gould Goldbergs (but not everybody, even on this board, was similarly impressed) and, once I heard it in hi-rez quad, I liked the Rachmaninoff album too. So for me, whatever was lost or sacrificed was less than what was gained in the fullness of tone and dynamics as well as the freedom from noisy distractions.
Yes, but of course, one could get a brand spankin' new high rez Rachmaninoff recording by someone else rather than listen to cleaned-up pseudo Rachmaninoff himself. If they're audibly altering the performance, what's the point?
For example, I can see how switching Gould's 1955 Goldbergs to a Yamaha piano could annoy people. But perhaps because Gould himself ultimately switched to Yamaha, it doesn't overly offend my ears. Also, I suspect that when the source material sounds better, and the original Gould Goldbergs does sound pretty good, the Zenph result is better. But I'd guess the earlier, higher noise, higher distortion 78s might not work out as well, at least in terms of accurately reproducing the original, which is what many care about most for historical material.
And I think it's perhaps overstating it to call the Zenph versions "pseudo Rachmaninoff".
was, Zenph might produce an excellent-sounding result that nevertheless is somehow audibly different from the original in some systematic way or ways. Some would be happy with that, others may reject it. You and Farhan Malik might differ. I didn't mean to make a value judgment on something I haven't heard.
The Satie is available in 'FLAC 24-bit Surround 5.0' at e!classical below:
Total cost for the whole 24/96 multi-channel album during the promotional period: $9.10! Thank you, Robert - VERY much!
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: