WTC 1 and 2 Complete

[johnlewisgrant]

Heavily sequenced and ultimately reproduced on the Hans Zimmer Piano sample using the Kawai RM3 controller. 

Book 1
https://soundcloud.com/johnlgrant/sets/bach-well-tempered-clavier-bk
Book 2
https://soundcloud.com/johnlgrant/sets/bach-well-tempered-clavier

Really, this experiment--which I've posted here quite a while back--has been useful to me on my REAL piano, my Hailun 218 (which I love).  It's Books 1 and 2 as, ideally, I would LIKE to hear myself play them.  A interpretation entirely "in my mind," inotherwords, which serves for me as an interpretive target at which to aim on my real piano (the Hailun 218).

It's interesting that in mocking these up at the Kawai controller, each vst (because I experimented with other VSTs besides the HZP and I go through them on the Hailun as well) forces a different approach to the music.
 

[anacrusis]

I'm not really sure I understand. Did you create a digital recording with MIDI somehow? What is a controller?

[johnlewisgrant]

Sorry. I forget to provide the applicable links!  Here they are again.

Bach WTC Book 1 (Complete):
https://soundcloud.com/johnlgrant/sets/bach-well-tempered-clavier-bk

Bach WTC Book 2 (Complete)

https://soundcloud.com/johnlgrant/sets/bach-well-tempered-clavier

An example of a "controller" is any keyboard that can be attached to a computer.  (A crude but functional definition.) 

[lelle]

This sort of thing should probably be posted in the Audition Room and not in Performance I will see if I can find some time to listen to some of it later. When you say it's "A interpretation entirely "in my mind,"", do you mean that you have edited the midi?

[johnlewisgrant]

Yes, I've midi-edited these.  Although I've played all the WTC over the years (started piano at 5, and I'm 67 now), it would be disingenuous of me to pretend that I'm doing anything other than producing, by hook or crook, a recording that satisfies me personally, and in my case, that definitely rules out the live home recording approach.

With the advent of Yamaha's most recent piano player systems, more and more high end recording studios are using the technology.  It is cheaper, easier, and way more precise than editing with multiple takes and audio splicing. 

This represents a sea-change in the industry.  We all know that unless explicitly denied ALL modern classical recordings are edited, some quite heavily.  Now pianists have the option of editing their performances using MIDI. I'm told that it's ALREADY the norm in many classical piano recording studios.  Remember, we're not talking about "MIDI" in the usual sense, but a Steinway D, typically, equipped with the latest piano replaying technology, which is now so completely accurate that the original performance cannot be in any way distinguished from the replayed performance.  The original and the replay are for all intents and purposes identical.   

Therefore, it is obvious that using MIDI to "correct" errors a studio recording is a much cheaper and more accurate way to do post performance editing.  I don't know for a fact whether the technology has, at this point, trespassed the sphere of classical piano recordings. But there's no way of knowing, really, unless someone "confesses," so to speak.

So, from my point of view, we have essentially 2 types of recording: LIVE recordings which are truly "authentic" in the sense that the pianist is using his hands and fingers and NOTHING ELSE; and NON-LIVE recordings, where not only are minor technical mistakes corrected but also entire conceptions of a piece can be fundamentally altered. 

The latter has always been the case, I think, even with the relatively innocent technique of audio splicing.  Take Hewitt's WTC (there are 2 recordings); I've heard Hewitt play the entire cycle twice, and the live performances are completely unlike her studio recordings.  That's not a bad thing at all.  It's just to say that retakes and editing, big or small, can fundamentally alter an interpretation. 

For some listeners this distinction matters.  For others it does not.  The MIDI edited or audio spliced/multiple retake studio recording attempts to get the listener as close as possible to what the pianist IDEALLY thinks a good interpretation should consist of.  We are peering into the performer's mind, not necessarily into the operation of her hands and fingers!  (Hands and fingers alone may not convey the performer's concept of the piece.) 

The concert hall is (or certainly CAN be) something very different, calling attention  arguably to audience/artist communication and spontaneity, not to a perfected conception of the music. Gould famously argued that the former--the interpretation in the performer's mind perfected by heavy editing if necessary--was what mattered. The drama of live performance was not his bag.

BOTH live and recorded/edited are equally valid and interesting from my standpoint. Covid has been a real downer for me, because I live for live classical concerts. 

My point is that as far as studio recordings are concerned, I think we have to acknowledge that the game is changing.  I wouldn't be at all surprised if ALL studio recordings of classical repertoire eventually succumb to ex post facto MIDI editing. 

When that happens, for better or worse, the floodgates will open.  The sharp distinction that pianists (especially classically trained pianists) like to draw between a) music that is made with the hands and fingers and b) music that is made sitting in front of a computer monitor may become perilously weak.

[anacrusis]

Yes, I've midi-edited these.  Although I've played all the WTC over the years (started piano at 5, and I'm 67 now), it would be disingenuous of me to pretend that I'm doing anything other than producing, by hook or crook, a recording that satisfies me personally, and in my case, that definitely rules out the live home recording approach.

With the advent of Yamaha's most recent piano player systems, more and more high end recording studios are using the technology.  It is cheaper, easier, and way more precise than editing with multiple takes and audio splicing. 

This represents a sea-change in the industry.  We all know that unless explicitly denied ALL modern classical recordings are edited, some quite heavily.  Now pianists have the option of editing their performances using MIDI. I'm told that it's ALREADY the norm in many classical piano recording studios.  Remember, we're not talking about "MIDI" in the usual sense, but a Steinway D, typically, equipped with the latest piano replaying technology, which is now so completely accurate that the original performance cannot be in any way distinguished from the replayed performance.  The original and the replay are for all intents and purposes identical.   

Therefore, it is obvious that using MIDI to "correct" errors a studio recording is a much cheaper and more accurate way to do post performance editing.  I don't know for a fact whether the technology has, at this point, trespassed the sphere of classical piano recordings. But there's no way of knowing, really, unless someone "confesses," so to speak.

So, from my point of view, we have essentially 2 types of recording: LIVE recordings which are truly "authentic" in the sense that the pianist is using his hands and fingers and NOTHING ELSE; and NON-LIVE recordings, where not only are minor technical mistakes corrected but also entire conceptions of a piece can be fundamentally altered. 

The latter has always been the case, I think, even with the relatively innocent technique of audio splicing.  Take Hewitt's WTC (there are 2 recordings); I've heard Hewitt play the entire cycle twice, and the live performances are completely unlike her studio recordings.  That's not a bad thing at all.  It's just to say that retakes and editing, big or small, can fundamentally alter an interpretation. 

For some listeners this distinction matters.  For others it does not.  The MIDI edited or audio spliced/multiple retake studio recording attempts to get the listener as close as possible to what the pianist IDEALLY thinks a good interpretation should consist of.  We are peering into the performer's mind, not necessarily into the operation of her hands and fingers!  (Hands and fingers alone may not convey the performer's concept of the piece.) 

The concert hall is (or certainly CAN be) something very different, calling attention  arguably to audience/artist communication and spontaneity, not to a perfected conception of the music. Gould famously argued that the former--the interpretation in the performer's mind perfected by heavy editing if necessary--was what mattered. The drama of live performance was not his bag.

BOTH live and recorded/edited are equally valid and interesting from my standpoint. Covid has been a real downer for me, because I live for live classical concerts. 

My point is that as far as studio recordings are concerned, I think we have to acknowledge that the game is changing.  I wouldn't be at all surprised if ALL studio recordings of classical repertoire eventually succumb to ex post facto MIDI editing. 

When that happens, for better or worse, the floodgates will open.  The sharp distinction that pianists (especially classically trained pianists) like to draw between a) music that is made with the hands and fingers and b) music that is made sitting in front of a computer monitor may become perilously weak.

Fascinating. My preference has always been for recordings that are as unedited as possible. A performance is birthed in the moment, and you are always reacting off of what you did before. I love hearing an authentic, unedited performance, even if there are many mistakes, if the spirit and character is right. I wonder if editing in detail is ever going to be able to reproduce the feeling of something being birthed, organically, as a whole.

[johnlewisgrant]

The spontaneity of a live performance.... or the retakes and splicing of the recording studio the purpose of which is to make a performance "better", in some sense: remove technical errors, or perhaps make a performance more coherent, more detailed, more dynamic, etc., etc., etc...  Gould's preference was for the latter type of music-making; but it seems to me hard to argue definitively for the inherent superiority of live and spontaneous over edited and preconceived, or vice versa..  A matter of personal taste, in the end.

Occasionally one gets both, however: a performance that is completely coherent and structured and yet absolutely spontaneous; one perfect take in the studio, or a recording of a memorable live performance.  It could be argued, I think, that only live performances can have both characteristics; that you can't produce a heavily edited performance or interpretation of a piece that is also spontaneous, but that you can do a live performance that, if luck or talent would have it, ends up being wonderful in every respect. This position, if correct, would require an argument that shows that editing and spontaneity are incompatible.

I don't know the answer. Personally, I'm a product over process listener: I don't want to know HOW the music was made; the knowledge might very well prejudice my appreciation.

[lostinidlewonder]

Midi or any sequencing is not an improvement over real recordings. Juxtapose a sequenced recording to a real one and we will see how much superior 100% unedited human playing with fingers only on a real piano keyboard is.

[johnlewisgrant]

Midi or any sequencing is not an improvement over real recordings, you tried to convince us years ago of that. Shall we debate it again? Round... what would it be... like 5. Juxtapose a sequenced recording to a real one and we will see how much superior 100% unedited human playing with fingers only on a real piano keyboard is.

Actually, I completely agree with you on the first point, that “sequencing is not an improvement over real recordings.”  If you can show me where I say midi is “better” I’ll happily retract my words.

I suggested above that sequencing could be an efficient alternative to splicing in corrections or modifications, the way in which editing is done in recording studios. But that’s not saying that sequencing is “better” than real piano-playing.

On the second point we may have a legitimate disagreement. If you mean that real recordings are better than sequenced ones in the sense that they are “more beautiful” or “more interesting” in some way, then I’m forced to say that for me at least judgments of of this sort are ultimately matters of personal taste. I find many sequenced “performances” unlistenable, but my opinion is subjective, not factual. For me it’s like saying that Picasso is better than Rembrandt or Mondrian. It boils down to personal taste.

[lostinidlewonder]

You said

"My point is that as far as studio recordings are concerned, I think we have to acknowledge that the game is changing.  I wouldn't be at all surprised if ALL studio recordings of classical repertoire eventually succumb to ex post facto MIDI editing."

So are you saying recording in studio without any midi editing will be inferior? This is where I disagree. You say you completely agree that it is no improvement over real recordings then think all studios will use it eventually. Even used to edit one note here and there it is going to be negligible improvement and in many cases sound unusual and artificial since you are trying to edit an audio file in midi format which always be an estimated value.

If you disagree with the juxtaposition then just post some examples side by side, midi will always lose, people always hear the difference and appreciate humans are still winning by a far margin. Pianistic mannerisms in Bach music don't really tell us much interesting as Tovey supported, the over use of pedal and forcing volumes up and down won't tell us much interesting about Bachs writing or the music of his period. Sure people can have personal taste but the tradition needs to be also well known or we lose it.

[johnlewisgrant]

"So are you saying recording in studio without any midi editing will be inferior?"

I'm sorry that I can't make my point more clearly.  Midi-editing is neither superior or inferior; it is merely a more efficient way (in my view) of accomplishing what has ALWAYS occurred in the recording studio: audio editing.

I mean this in the following sense:

"Editing" is almost always "splicing", removing the unwanted sections and substituting the wanted ones.  I understand from the engineers that do this for a living that as many takes as the pianist can AFFORD (studio time is money) will be made, and more often than not the final recording will consist of a splicing together of these takes. 

The process used to be accomplished with physical splicing of tape.  That method was replaced by digital splicing.  Now there are studios equipped with MIDI recording Steinway, Bosendorfer, and Yamaha player pianos; so the process can be done even more precisely via MIDI coupled with piano-replaying the "corrected" performance
(not pianist-replaying but the midi recording being replayed by the player-piano).
 
Do I think that the "corrected" performance is "better"?  Not necessarily, obviously, regardless of which method is employed: tape splicing, digital manipulation of wave files, or MIDI coupled with replaying the performance via Disklavier.

The question of "is the edited performance "better"?" is up to the performer, engineer, and producer, all of whom I assume have to approve the final result before the CD is pressed, or the performance is digitally released online.

The really important distinction, I think, is between LIVE and RECORDED.  The latter I think we can say with certainty, is ALWAYS, of necessity, something ARTIFICIAL.  Again, that's the case regardless of which of the above editing methods is employed. 

One could go out on a limb, I suppose, and say that the MORE editing or splicing that is done, the more ARTIFICIAL the recording, again, regardless of the method of editing.

Personally, I'd want to be a bit clearer about what "artificial" means before making that particular philosophical leap.

A final note: even "live" recordings appear to get edited from time to time.  Bad cell-phone events are edited out.  But performance snafus are as well.

Case in point: Angela Hewitt in Toronto, a number of years ago, playing all of the WTC live, in concert.  She got completely gummed up, completely lost, in the latter half of the Book 1 B minor fugue (#24). In the CBC (Canadian Broadcasting Corp) radio version of the live concert (which I attended), the damage was somehow repaired!

Postscript: Working with MIDI and a piano SAMPLE, which is what I've done with the WTC in this instance is ultimately a kind of MIDI super-editing. It is NOT a performance.  That raises the question many people find perplexing.  If I were to have the benefit of expensive microphones and, say, a concert Fazioli or Hamburg Steinway equipped with the latest and greatest player system, I could have recorded my work using exactly the same equipment that can be used to MIDI edit and splice piano performances at Disklaviers.

So the question arises: at what point does a "performance" that has been super-edited become something OTHER than a performance by the original pianist?  For some listeners, ANYTHING other than a bona fide LIVE performance is musically suspect.  For others, it depends how much ex post fact editing has been done.

These are questions of PROCESS, however; not questions of PRODUCT.  If I hadn't attended the Hewitt WTC concert, it would never have occurred to me that editing had occurred.  The "product" seemed just fine, but I was ignorant of the "process."

These things don't matter to me, personally.  But they matter a lot to others.

Concerning the accuracy of player pianos.  I saw a performance replayed by a Yamaha with a small group of listeners on stage at Koerner Hall in Toronto.  I couldn't tell the replay from the original (which had been performed minutes earlier).  Impossible, and that  was many years ago.  MIDI "2" has, I'm told, over one thousand velocity layers, not the old "127".  And the player mechanisms are much more sophisticated.  So I'm not convinced that accuracy of reproduction is a limiting factor.