Mozart piano sonatas

[Purusha]

I'm getting towards the end of my comparison, the first round of it anyway as I'll still have a number of cycles to listen to progressively after that.

After the combined K.475/457 versions, I listened to a few versions of each of the two works recorded separately. And they were all forgettable at most, except one version of K.457, by Annie Fischer (live in London, 1971, BBC Legends 4166). It's a very particular version, Fischer's piano is as usual very massive, romantic, powerful, schumannian... But it's also incredibly elegant, insightful, generous and spontaneous at the same time. I still rank it behind Moravec, Virsaladze and Van Immerseel, but would still recommend it among first choices.
No other recording of K.475 struck me as really unavoidable, even Edwin Fischer (but I only heard his 1941 version, couldn't find his 1947 recording apparently).

In K.533/494, the choice is not as varied as with most other sonatas but there are still very strong versions. But I still find a variety of issues even with the best of them. Peter Rösel (1982), for instance, plays it wonderfully, I mean, it is one of the most technically perfect Mozart performances on record, and it is detailed, subtle, and with gorgeous sound. It just sounds lacking a little spontaneity in the end. His sense of perfection is one of the strengths of Rösel, of course, but the result sounds just slightly too formal for my taste in Mozart.
Mikhaïl Pletnev's recording for Melodiya (1984) is very different, very personal, no other recording compares to it, and it is captivating, fascinating. But originality also leads to a few weaknesses, the result is original, creative, overwhelming, but it is not really Mozart anymore you're listening to. I would rank it as a really essential, indispensable version, but it can't be said a "first choice".
In the end, two versions are my personal nominees for best version. Sviatoslav Richter's live in 1989, in Como, is a wonderful performance, one of the top-notch Mozart performances by Richter. I still have to hear his 1966 Prague version though, I had forgotten it. And the other "best" version in my opinion is Jos van Immerseel (1996). Again, as in K.457/475, I was unsure about my fond memory of it. In the end, this opinion is entirely justified, it is a gorgeous and incredibly detailed and subtle recording. Every aspect of the sonata is rendered with colours, expressiveness, tenderness, but all contrasts are kept strong, nothing sounds mannered, on the contrary, the sense of quasi-improvisation is kept all along.
In addition to Richter 1966, I just saw I had forgotten Gilels 1972. Well, that's something I'll straighten out shortly.

The short K.545 allows for a speedy comparison, it's quite nice after K.533, which allows for the longest performances among Mozart sonatas. After hearing a half dozen, I'm sure Lili Kraus, at least, will be among my favourites.

Updated list of favourites :

Sonata 1 (K.279): Lili Kraus (1954), Daniel Barenboim (1984-85)
Sonata 2 (K.280): Lili Kraus (1954), Clara Haskil (1961)
Sonata 3 (K.281): Lili Kraus (1954), Emil Gilels (1970), Daniel Barenboim (1984-85)
Sonata 4 (K.282): Samuil Feinberg (1953), Lili Kraus (1954), Sviatoslav Richter (1989, live), Andreas Staier (2003), Elisso Virsaladze (2013)
Sonata 5 (K.283): Lili Kraus (1954), Sviatoslav Richter (1966, live), Dezső Ránki (1978-79), Dezső Ránki (1997, live)
Sonata 6 (K.284): Daniel Barenboim (1984-1985)
Sonata 7 (K.309): Sviatoslav Richter (1968, live in Prague), Daniel Barenboim (1984-85)
Sonata 8 (K.310): Dinu Lipatti (1950, studio in Geneva + live in Besançon), Emil Gilels (1971, live in Ossiach), Alexei Lubimov (198?), Sviatoslav Richter (1989, live in London)
Sonata 9 (K.311): Lili Kraus (1954), Dezső Ránki (1978-78)
Sonata 10 (K.330): Walter Gieseking (1953), Krystian Zimerman (live in Vienna, 2008)
Sonata 11 (K.331): Walter Gieseking (1953), Lili Kraus (1954)
Sonata 12 (K.332): Artur Schnabel (1946), Lili Kraus (1954), Andreas Staier (2004)
Sonata 13 (K.333): Lili Kraus (1954), Sviatoslav Richter (live in Prague, 1966), Ivan Moravec (1982), Vladimir Horowitz (1987, studio), Vladimir Horowitz (1987, live in Munich)
Fantasy K.475 + Sonata 14 (K.457): Ivan Moravec (1967), Elisso Virsaladze (live in Munich, 1995), Jos van Immerseel (1996)
K.475 recorded alone: -
K.457 recorded alone: Annie Fischer (live in London, 1971)
Sonata 15 (K.533/494): Sviatoslav Richter (live in Como, 1989), Jos van Immerseel (1996)
Sonata 16 (K.545): Lili Kraus (1954)
Sonata 17 (K.570):
Sonata 18 (K.576):

This list seems to be missing a certain Robert Casadesus.

Together with Jean-Bernard Pommier, another Mozart pianist i like which i rarely see mentioned  is Carl Seeman. They don't have any of his Mozart sonatas on youtube but there is a concerto:

http://www.youtube.com/v/JmRXR_2jfR0

I would also recommend Peter Rosel if you never heard of him:

http://www.youtube.com/v/BrdsMzyL728

[Cosi bel do]

I will not make further comments on pianists that are absent from this selection. Please, consider that I haven't heard any other versions than those mentioned, as I wouldn't like anyone feelings being hurt because I might have considered versions were better than others.

[Purusha]

The video refuses to play. The later Gulda recordings of which there are 4-5 sonatas in good sound: K 570, 576 and the c minor fantasy on DG and 331 and 333? on amadeo are the opposite of mannered. Rather straight and IMO natural and tasteful interpretations.

Are those studio recordings?

The only Gulda i have for Mozart are live performances of the first five sonatas he recorded sometime during the early 80s i believe. They are very good performances, and the sound quality is pretty good by the standards of live recordings. But i'm currently in the process of expanding my collection of studio recordings and Gulda is a very interesting musician. His Beethoven set on Amedeo (now on Brilliant?) is one of the best i ever owned.

[Purusha]

I will not make further comments on pianists that are absent from this selection. Please, consider that I haven't heard any other versions than those mentioned, as I wouldn't like anyone feelings being hurt because I might have considered versions were better than others.

Dude, no reason to get so defensive. I don't know what's going on here, i'm just posting recommendations.

[Mandryka]

This list seems to be missing a certain Robert Casadesus.

Together with Jean-Bernard Pommier, another Mozart pianist i like which i rarely see mentioned  is Carl Seeman. They don't have any of his Mozart sonatas on youtube but there is a concerto:

http://www.youtube.com/v/JmRXR_2jfR0

I would also recommend Peter Rosel if you never heard of him:

http://www.youtube.com/v/BrdsMzyL728

I was impressed by Pommier's 310 but haven't explored further, I don't own the set. Are there lots of really characterful and convincing performances in it?

[Mandryka]

This one is going to be wildly off topic...  Or maybe not...  I don't think piano music differs that much from other core repertoires in music.  At first sight, piano seems to be a percussive instrument; it seems piano playing is all about hammers, set into motion by the action of the fingers on the keys, and about strings, set into motion within the air by the action of those hammers onto them.  But piano music does not necessarily require such a percussive sound; actually, it seems to me quite the contrary, and the overwhelming majority of piano works (and pages thereof) require a variety of sounds that seems quite impossible to obtain from a keyboard.  (One of them being the percussive sound, of course.)  Through its performance, (piano) music presents itself, first and foremost, as a texture—even harmony and form or structure are particular cases of textures, from a sufficiently broad level of abstraction.  Therefore, IMO, the performer should excel at projecting out a wide, broad, and at the same time refined spectrum of textures, those that seem to him required by the music he hears or has heard within himself when he read and studied the score.

If one agrees with this opinion—and of course one needs not agree with it, though it certainly is my firm opinion—, then I don't think one can reduce, as it is quite often the case, piano playing to the art (or is it `ability', rather?) to put fingers at the right place and at the right time, whatever the tempo, whatever the nuance.  On the one hand, I really don't think that playing an F-sharp where an F-flat would have been in harmonic order spoils the music nearly quite as much as it is often said to harm it; these are merely incidents, mishaps, or hitches that just pass by and go (well, at least for me).  Artur Schnabel is reputed to have once said, clearly tongue in cheek, that one main problem of keyboard playing was that every correct key was midway between two incorrect ones...  But Schnabel also said—and this time he was serious about it—that music had nothing to do with security, and especially that piano music had nothing to do with fingers.  But I digress.  What I meant to say is that the `matter of the fact' definition of piano technique, namely the skill or craft to put the right fingers at the right place and at the right time, does not say much about real piano technique—actually, it does say almost nothing about it at all.  On the other hand, it is actually quite possible to play badly, and by this I mean also to play out of tune, while at the same time playing all the right notes at the right times on a perfectly tuned and harmonized piano.  For instance, a dominant chord is nothing special to marvel at in most piano music pieces (let's not even talk about a diminished seventh).  But within the realm of classical piano literature, a dominant chord is absolutely essential; it is not a benign event, it has `meaning' and at least it serves a purpose, because classical works draw a significantly large part of their sheer power from the harmonic tension between the tonic and its dominant.  Therefore, if a piano player plays the right notes within the right chords, but does not also play the harmonic function of the dominant chord inside the tonic region of the piece, then this piano player might be said to play badly, or rather even `out of tune', though he may be playing perfectly from the vertical, or letter, or score point of view, and on a perfect instrument.  The musical `effect' is spoilt.  There's a great (great because it is simple, almost obvious) example of such a dominant chord inside the tonic region at the extreme end of the theme of Beethoven's Arietta (Op.111); frankly, there's only one pianist I know who seriously acknowledges the dominant chord function, its crescendo, its sforzando, its subsequent piano (which is rather subito), and succeeds in projecting them all in a really convincing way, I mean in a way that makes one feel the (classical style) harmonic tension produced by the dominant, and then the C major diminuendo tonic resolution (this is at bars No.13-16).  Sadly, he is also a pianist whose performances of some classical pieces not infrequently get dismissed because they seem to `go to the extremes'—but I don't share the opinion that classical period music does not go to the extremes every bit as often as music from earlier or later eras does.

IMO, music pieces remain unique, spherical so to speak, and thus I expect the musician who plays them to project a unique work of art in each case.  Even when the piece is perfectly played from a strict vertical point of view, I think anyone is entitled to put forth strong, rational arguments against the performance if it does not raise itself to the level of uniqueness of the piece, if it seems to play the style or the genre or the period or (perish the thought!) the theory rather than the piece itself.  This is especially difficult (and important) in the case of music pieces from the classical era, because the material itself (including harmony and rhythm, of course) is rather circumscribed there, when compared both to earlier and to later pieces.  Piano may or may not be a percussive instrument, but one thing seems sure for me: it is a harmonic, polyphonic instrument; the ability to play the harmony through time is, IMO, much more important than that to avoid mishaps (and, as I already said above, I don't think mishaps spoil the harmony that much, and at least they don't harm it for long).  Not every emotion, characterization, and expressiveness comes from harmony and form, though; that's why I think textures are so important, especially in the case of piano literature.  I think that when uniqueness is not felt, something every bit as technical as a wrong note must have occurred, though it is certainly much more difficult (and interesting) to track it down than to point out a wrong note.  In the case of the piano, this is especially true when one is constantly reminded of the percussive nature of the instrument by the performer, and this is again on purely technical terms.

I don't think there can really be generic performances - the performer has to make too many decisions about all the nuances. What there can be is thoughtless, uncommitted, uninvolved performances. It's as if, in a worthwhile performance the spiritual and imaginative forces of musician interact somehow with the score, and the spiritual and imaginative forces of the listener interact with the concert.

I'm getting close to saying that this is all I will allow in evaluation. All that matters is the focus and imagination of the performer and the focus and imagination of the listener. That's part of the reason I'm beginning to feel a bit negative about Doscobolus's approach.

But these are difficult questions, I'm not clear. Time plays a role too - it's a listener and a performer in history. The composer drops out of the picture almost, quite rightly - he's done his work in leaving the score for living poets to interact with. End of.

Focus is what made Sofronitsky great. Is it him you're thinking of in op 111?
 
And I share your distaste for didactic performance. And I wonder what you think of very percussive pianists like Annie Fischer (in the Schumann Fantasie)

[Cosi bel do]

That's exactly the discussion I was thinking about for a new topic. Too late

[Que]

So what would be (exactly) the new topic? Please propose a title and I'll move all relevant posts to a new thread.

Q

[Cosi bel do]

Well, I don't know, either something very precise as "are there objective criteria to judge the quality of piano playing ?" or something more generic that could be used for different debates, as "Interpretation on the piano"... I was actually still checking if nothing of the kind exists already.

[Cosi bel do]

I'd also prefer the generic title. The term of performance generally implies, as I understand it, "live" performance, in front of an audience. Of course one could discuss if playing in front of microphones isn't also a performance, aimed at a certain audience, but it seems the title would be less clear about the content than the term "interpretation" would, and it would end as a discussion about piano concerts...

[Cosi bel do]

So I have finally listened to all 18 sonatas in a whole lot of versions, between 6 and 19 each. I'm not going to comment it further anymore, as I just want to underline the performance that are the best (not only "my favourites" then, but also which represent the best accomplishment on a technical level and other interpretative matters). If you don't see your favourite(s) in the list, just assume I don't know it, if that makes you feel better. It's weird how people are much more susceptible when it's about piano interpretation.

Updated list of best/favourite/recommended versions:

Sonata 1 (K.279): Lili Kraus (1954), Daniel Barenboim (1984-85)
Sonata 2 (K.280): Lili Kraus (1954), Clara Haskil (1961)
Sonata 3 (K.281): Lili Kraus (1954), Emil Gilels (1970), Daniel Barenboim (1984-85)
Sonata 4 (K.282): Samuil Feinberg (1953), Lili Kraus (1954), Sviatoslav Richter (live in London, 1989), Andreas Staier (2003), Elisso Virsaladze (2013)
Sonata 5 (K.283): Lili Kraus (1954), Sviatoslav Richter (live in Salzburg, 1966), Dezső Ránki (1978-79), Dezső Ránki (1997, live)
Sonata 6 (K.284): Daniel Barenboim (1984-1985)
Sonata 7 (K.309): Sviatoslav Richter (1968, live in Prague), Daniel Barenboim (1984-85)
Sonata 8 (K.310): Dinu Lipatti (1950, studio in Geneva + live in Besançon), Emil Gilels (1971, live in Ossiach), Alexei Lubimov (198?), Sviatoslav Richter (1989, live in London)
Sonata 9 (K.311): Lili Kraus (1954), Krystian Zimerman (1978), Dezső Ránki (1978-78)
Sonata 10 (K.330): Walter Gieseking (1953), Krystian Zimerman (live in Vienna, 2008)
Sonata 11 (K.331): Walter Gieseking (1953), Lili Kraus (1954)
Sonata 12 (K.332): Artur Schnabel (1946), Lili Kraus (1954), Maria Grinberg (1967), Andreas Staier (2004)
Sonata 13 (K.333): Lili Kraus (1954), Sviatoslav Richter (live in Prague, 1966), Ivan Moravec (1982), Vladimir Horowitz (1987, studio), Vladimir Horowitz (1987, live in Munich)
Fantasy K.475 + Sonata 14 (K.457): Ivan Moravec (1967), Elisso Virsaladze (live in Munich, 1995), Jos van Immerseel (1996)
K.475 recorded alone: -
K.457 recorded alone: Annie Fischer (live in London, 1971)
Sonata 15 (K.533/494): Sviatoslav Richter (live in Prague, 1956), Sviatoslav Richter (live in Como, 1989), Jos van Immerseel (1996)
Sonata 16 (K.545): Lili Kraus (1954), Sviatoslav Richter (live in Prague, 1956), Dezső Ránki (1978-78), Sviatoslav Richter (live in London, 1989)
Sonata 17 (K.570): Emil Gilels (Prague 1953), Lili Kraus (1954), Ivan Moravec (1963), Daniel Barenboim (1964, Westminster), Jos van Immerseel (1996)
Sonata 18 (K.576): Lili Kraus (1954), Jos van Immerseel (1996)

Here is also my ranking of the cycles I have listened to completely (but this is kind of deceptive, as some of the cycles in the middle might be actually very uneven, when others might just be completely unremarkable, for instance Ranki's part of the Hungaroton cycle is consistently magnificent, but Kocsis's half cycle really mediocre) :

1. Lili Kraus I (Discophiles français/Erato, 1954)
2. Maria João Pires I (Denon/Brilliant, 1974)
3. Daniel Barenboim (EMI/Warner)
4. Ranki & Kocsis (Hungaroton)
5. Mitsuko Uchida (Philips)
6. Christoph Eschenbach (DG)
7. Klara Würtz (Brilliant)
8. Glenn Gould (Columbia/Sony)

Now, I will listen to these sets during the next few months, but without clear order (and sometimes at the same time):
Horszowski / Haebler II (Denon) / Lubimov / Kraus II (Sony) / Arrau / Van Oort / Schoonderwoerd / Newman / Gieseking / Klien / Zacharias EMI / Larrocha

Plus a bunch of recordings of single sonatas.

Don't worry, I will also listen to other cycles afterwards (some of them I don't have at my disposal yet), at least : Haebler I (Philips) / Schiff / Bezuidenhout / Brautigam / Pires II / Pienaar / Gulda Tapes / Jando / Perlemuter / Engel.

At least this comparison has allowed me to realize how much I loved most of these pieces !

[Moonfish]

It has been quite interesting to follow your Mozart journey, Discobolus. Obviously a subjective realm, but your viewpoints have always been illuminating. Obviously, we all respond somewhat differently to these recordings. Personally, I lately shifted my favor to Lili Kraus (mostly because I never had the opportunity to listen to it before the new Erato set was issued). As you may know I championed Uchida before that although she did not fare well in your comparison (if I recall correctly). Regardless, the list (and thread) you created over the last month or so will always serve as a great starting point for diving into the Mozart sonatas. Where we go from there as individuals will of course be up to us. I suspect that most of us were pretty much entrenched with certain recordings in the first place. Perhaps "blind" listening is a way to avoid such preferences. Nostalgia and personal attachments to performances are always a factor (at least that is my own Uchida link). The thread makes great reading while pondering Mozart! 

[Cosi bel do]

It has been quite interesting to follow your Mozart journey, Discobolus. Obviously a subjective realm, but your viewpoints have always been illuminating. Obviously, we all respond somewhat differently to these recordings. Personally, I lately shifted my favor to Lili Kraus (mostly because I never had the opportunity to listen to it before the new Erato set was issued). As you may know I championed Uchida before that although she did not fare to well in your comparison (if I recall correctly). Regardless, the list (and thread) you created over the last month or so will always serve as a great starting point for diving into the Mozart sonatas. Where we go from there as individuals will of course be up to us. I suspect that most of us were pretty much entrenched with certain recordings in the first place. Perhaps "blind" listening is a way to avoid such preferences. Nostalgia and personal attachments to performances are always a factor (at least that is my own Uchida link). The thread makes great reading while pondering Mozart! 

Thanks ! The fact it's not a blind comparison is actually the main bias in my approach, but I didn't originally think I would rate Barenboim so high for instance, or Gould so low (even if I never really liked his Mozart I didn't think it was so bad before hearing it again).

Uchida is not that bad actually, she generally is a very reliable and comfortable reference, as is Eschenbach by the way. Rarely excellent (but sometimes, yes, Uchida is good in early sonatas and, even more, in K.311 and 330) but even more rarely really bad (except, for Uchida, in K.331, I'm not sure there's a worse recording than hers). But then, compared to Kraus, it is true than most cycles sound decidedly of limited interest...

[amw]

I'd actually appreciate any more detailed comments you might have on K576, it's my favourite of the sonatas and also one of the most difficult (technically and interpretively). Only if you're so inclined though.

[George]

Thanks for sharing your findings with us, Discobolus!

I am curious about something. Barenboim appears as a favorite in five sonatas, Pires I does not appear as a favorite in any works. Yet, she ranks higher than Barenboim in the complete sets. Can you say why that is?

[Mandryka]

I'd actually appreciate any more detailed comments you might have on K576, it's my favourite of the sonatas and also one of the most difficult (technically and interpretively). Only if you're so inclined though.

Something I look out for when I listen to that one is the development section of the first movement. It's very contrapuntal and I always like contrapuntal Mozart -- Mozart's severe style. It takes someone who's experienced at making that type of texture work in an interesting way, the voices relate to each other in an interesting way.

[Cosi bel do]

Thanks for sharing your findings with us, Discobolus!

I am curious about something. Barenboim appears as a favorite in five sonatas, Pires I does not appear as a favorite in any works. Yet, she ranks higher than Barenboim in the complete sets. Can you say why that is?

Well, yeah, that's because as I said Barenboim is really uneven, a few excellent sonatas but not everything is good in his set. While Pires is kind of a safe bet, she is never less than good. So if you want to hear a Mozart sonata, try Barenboim and you have a 50% chance (well, that's just to make myself clear) to be either disappointed or enchanted; try Pires and you'll certainly be charmed and satisfied (even if you are not as convinced or astounded in the end as you could be by more complete, or detailed, or tense (or whatever) performances).

[George]

Well, yeah, that's because as I said Barenboim is really uneven, a few excellent sonatas but not everything is good in his set. While Pires is kind of a safe bet, she is never less than good. So if you want to hear a Mozart sonata, try Barenboim and you have a 50% chance (well, that's just to make myself clear) to be either disappointed or enchanted; try Pires and you'll certainly be charmed and satisfied (even if you are not as convinced or astounded in the end as you could be by more complete, or detailed, or tense (or whatever) performances).

Ah, that makes sense! Thanks for explaining! And thanks again for your survey.

[Mandryka]

As you are already discussing K.475/475 I was still listening K.333 and trying not to get too impatient

As in K.332, in K.333 Schnabel and Kraus were early strong contenders... Despite its imperfections, Artur Schnabel (1943-44) is really very interesting. But Lili Kraus is even better, she is already captivating in 1948, and manages to reinvent and deepend every aspect of her interpretation in 1954. This version has actually no equal in the whole discography, and I still preferred it after hearing many other versions that left me quite disappointed by several of them, including Wanda Landowska (1955), really slow and mechanical, lifeless ; or even frustrated, as with Sviatoslav Richter (live, 1966) who puts somewhat a great performance in Salzburg but suffers from a horrible piano and is unable to cover it as he does, more or less, in the rest of the same concert. Even the almost perfect Dezso Ranki (1978-79) lacked a little of the irresistible spirit one can find when listening to Kraus.
Then came Ivan Moravec (1982). And even if I think I still prefer Kraus, I can't deny Moravec is kind of perfect, elegant, sometimes almost mannered but with such a natural style that it never sounds artificial or out of place. With Moravec, the music breathes, everything is alive with an extreme spontaneity and flexibility. With Kraus 1954, Moravec is my second "5 stars" version.
A few names were remaining, that were not necessarily among my favourites, but I could expect them to do better than they really did, including Daniel Barenboim or Mitsuko Uchida, or Fazil Say (while those I feared I wouldn't like, mainly Würtz and Brendel, did indeed make me suffer, the latter succeeding in giving a paradoxically utterly flat, disgracious and pretentious performance at the same time).
But in the end a third name joined Kraus and Moravec on my personal list of final favourites, and it is the most unpredictable one : Vladimir Horowitz. He had taken me by surprise in K.281, and completely disappointed me in K.330. Here, the two versions I heard, the studio in Milan in 1987 (recorded with concerto K.488 with Giulini), and the live in Hamburg (the last concert), were surprisingly different in many ways, but both completely fascinating in their alliance between detail and subtelty, without ever sounding too mannered, thanks to a complete and touching sincerity...
These three big versions are quite different but they share, in the end, a few qualities. Lili Kraus, Ivan Moravec and Vladimir Horowitz were three very generous figures, sensitive artists with unique personalities, and a sincere humility. And these are, I guess, essential components of a true mozartian artist...

Updated list of favourites :

1 (K.279): Lili Kraus (1954), Daniel Barenboim (1984-85)
2 (K.280): Lili Kraus (1954), Clara Haskil (1961)
3 (K.281): Lili Kraus (1954), Emil Gilels (1970), Daniel Barenboim (1984-85)
4 (K.282): Samuil Feinberg (1953), Lili Kraus (1954), Sviatoslav Richter (1989, live), Andreas Staier (2003), Elisso Virsaladze (2013)
5 (K.283): Lili Kraus (1954), Sviatoslav Richter (1966, live), Dezső Ránki (1978-79), Dezső Ránki (1997, live)
6 (K.284): Daniel Barenboim (1984-1985)
7 (K.309): Sviatoslav Richter (1968, live in Prague), Daniel Barenboim (1984-85)
8 (K.310): Dinu Lipatti (1950, studio in Geneva + live in Besançon), Emil Gilels (1971, live in Ossiach), Alexei Lubimov (198?), Sviatoslav Richter (1989, live in London)
9 (K.311): Lili Kraus (1954), Dezső Ránki (1978-78)
10 (K.330): Walter Gieseking (1953), Krystian Zimerman (live in Vienna, 2008)
11 (K.331): Walter Gieseking (1953), Lili Kraus (1954)
12 (K.332): Artur Schnabel (1946), Lili Kraus (1954), Andreas Staier (2004)
13 (K.333): Lili Kraus (1954), Ivan Moravec (1982), Vladimir Horowitz (1987, studio), Vladimir Horowitz (1987, live in Munich)

I'm quite happy to get to the final sonatas, I had difficulties to maintain the rythm of my listening during K.330-333 : more versions, for works I don't really adore...
For K.475/457 I'll start with all versions where both pieces are played (and recorded) together, and listen other versions of each afterwards.

I ended up prefering  Horowitz's Hamburg 333, to me it comes across as less flashy, more rapt. And more natural. But you're right about the touching sincerity for sure. (The PC with Giulini is well worth seeing.)

[Sergeant Rock]

I ended up prefering  Horowitz's Hamburg 333, to me it comes across as less flashy, more rapt. And more natural.

And now, you'll never listen to it again?

But once you've heard what they do, and you've grokked the consequences, why bother to hear it again? You already know it, understand it.

Sarge