Hello everyone! Hope you are doing well.

Attached is a retake of Beethoven's 'Tempest' Sonata No 17. Just the 3rd movement for now... not quite ready with first two yet but hopefully soon.

Thanks to all for the great feedback on this piece from my posting a while back. While I have been spreading my practice time between a handful of different pieces (see additional postings later today), this one was certainly a big focus as I do so love the music. I hope you enjoy, and am very interested and appreciative of your feedback.

In addition to work on phrasing and other general corrections, here are some items I hope to have improved in particular:

* Articulation and evenness - it's certainly not "prefect" but this was something I put considerable time into
* Drive - I tried to maintain an aggressive push throughout and limited my use of rubato, although I couldn't resist a few particular places
* Trill corrections - thanks to feedback , I've adjusted my trill timing and emphasis in bars 42-46 and 270-274

Beethoven - Sonata No.17 in D minor, Op.31 No.2 "The Tempest", III: Allegretto

Well done! This is not an easy piece, to keep the drive, concentration and interest throughout the whole movement which is so completely dominated by a four note phrase. Articulation and evenness I think is fine.


A main concern is how to render this four note phrase. In my personal opinion you should not pedal the downbeat where the phrase ends. In the RH it is marked with a staccato dot - because of the slur it might be played more like portato, but should in any case neither be sustained nor carry an emphasis. In the LH is clearly marked a 16-note where only the following note is sustained, strongly suggesting that the first bass note should not be sustained. This suggests the phrase should be rounded off despite ending on a downbeat. When the phrase appears in the LH you do this better, though I would still not emphasize the downbeat as much.

For the trills in the second theme I would more emphasise the first note more, to get better stability of rhythm. In the following sequence with octaves in RH (sorry my edition has no bar numbers) you again emphasize and pedal the downbeat, breaking the flow. In fact this octave run is a paraphrase and extension of the original four note phrase, something that can be brought out much better. The subsequent passage with the tricky downward leaps in LH you do very well in this respect.

The development as you play it is pleasant to listen to (the issue of downbeat pedal remains though), but somehow I think Beethoven had something other than pleasant in mind. You could give it more of a sense of struggle, more dynamic variation with more abrupt fortes. It is completely obsessed with the key phrase, to an extent that is almost obscene. I see it as a series of failed attempts to get out and escape, which in the end creates an eerie feeling not very far from horror. At the end of the development the sad conclusion is that there is no alternative (I would not do a ritardando before the recapitulation). At the end of the coda the realization is that the only way out is to simply stop playing.

Your coda is fine. I would not do the rubatos towards the end, it breaks the pace.

Very interesting, we are agreed that he does it differently when the phrase is in the other hand, and that it should be more alike. Personally, though, I prefer his RH version to his LH, I would rather the LH staccato notes were a bit less short. They should be no less sustained than the way the RH played them earlier. Inasmuch as any difference is affected by use of pedal, then it ought to be applied fairly - if the RH gets it, the LH should get it too.
I completely agree. I suspect part of the idea of marking them staccato is to take some of the weight off them, and if one is going to compensate for the reduced note length by increasing the loudness of what's left, it's achieving an effect which is probably not intended. But we can only speculate.
The stability here (assuming you're speaking of bars 43-46) is already much improved from the previous version, but I agree. I think he is playing these "trills" (actually mordents) as two fast notes before the main note which in turn he plays on the beat, and is therefore accenting it; I reckon the first of the two fast notes should come on the beat and receive the main emphasis.

I think that in these four bars the hemiola aspect (this being in 3/8 time, but with an accent occurring every two beats it feels almost as though we had 6 bars of 2/8 instead of 4 bars of 3/8) is now overemphasized to the detriment of the underlying triple time, of which we should not completely lose sight here.

In bars 67-70 there is a curious effect as though some prankster were fooling around with the volume control, dimming out the middle of each bar. With the exception of a solitary occurrence in bar 15, this is the first time in the entire movement that the RH is silent for the entire second beat (and part of the first, due to that being staccato), and this is creating a "hole" in the overall sound level (well, four holes, since it happens in each of these four bars). Perhaps the middle two of the LH semiquavers in each bar could be beefed up a bit to compensate. Either that or a bit more pedal could help bridge the holes.

The 16th bar from the end (the bar after the chromatic scales in octaves) is still coming out as a 4/8 bar (note, rest, note, rest), it should be 3/8 (note, rest, note) with the 3rd beat note to be considered a definite upbeat to the following bar.

Matthew,
You have done a very good job of improving this! I didn't care for the tempo slowing and restaring around 3'00, but generally this is now a totally respectable rendition even if it still has opportunities for improvement. Congratulations! Here's one thing to think about that my collegues may or may not agree with: the measures that call for sudden f dynamic changes as in bars like m.99, 107, 243, should not change on the down beat of the measure (were the dynamic is indicated) but rather with the start of the four-note figure that begins three 16th-notes before, just like the original motif. If you study the score you'll see that Beethoven is not consistent with the metrical placement of the dynamic indications. Anyway, that's my interpretation when I have played this.

Again, Good job!

Best,
Eddy

Thanks so much for the quick review and feedback! I'm quite relieved, in particular, to hear this comment on the evenness - something I've been working on lately for sure.


Interesting. I remember working out the pedaling some time back, and in a few instances I even resorted to listening through a handful of recordings I found to get ideas. But it's been a while and I can't remember offhand my reasoning. I will certainly look back at this and do some experimenting, however - thanks for the alternate ideas!


I really like these sort of depictions - thanks so much for yours.

Thanks Eddy for listening and the feedback.

Regarding the brief [temporary] slowdown towards the end, that does seem to be the common annoyance to the folks that have reviewed this so far. I almost didn't put it in there, but at the last minute talked myself into it. I probably just need to go through a 12-step Rubato Addiction program because I've convinced myself it sounds good this way! But perhaps I'll concede this one because I have a feeling I'm in the minority on this point.

I like much this movement. It has the energy typical from Beethoven's music. Your rendition is much fine, especially the dynamics. The sound has too much boosted low end and is too dry for my taste. I would prefer the one from the processed file here attached. Great performance anyway!

Thank you for this. You play it very well and I enjoyed it.

You have done well with the dynamics and the range is very good.

I also like the lightness and precision of the staccato in the various places.

Thank you again.

Hello Matthew,

I've seriously worked on this one before, too. I like Joachim's description; that is how I also feel this movement. I think you could give it more 'struggle' and 'despair' - make it sound more desperate somehow. Easier said than done, I know. I was not satisfied with my own playing, gave up on it, and never recorded it. But I still like this Sonata very much though.

All in all, your playing here is good enough for the site, and Didier's magic makes your keyboard sound better to me. Just thought of something...maybe that because you are playing on a digital is the reason for the piece not sounding fully loaded....

Thank you for the comments Didier. Would you mind posting what you did to the file? I do like it better as well. Sounds like you added a longer reverb than I was using, and perhaps made some equalizer changes?

Thank you Mark for taking time to listen. I appreciate the nice remark on the dynamics - getting that "right" is something I certainly strive to do.

Thank you Monica for taking a listen! Very much appreciated.


Yes, this audio quality thing really bugs me and I continue to experiment. I'll keep working on it!

Yes just equalizing and adding reverb. The attached pictures shows the change of the spectrum before (white) and after (red) EQ and the reverberation setting.

What is your piano ? It sounds rather good, except for the unbalance in the lows. I appreciate especially the realistic stereo rendering as an opposite to most digital pianos where the highs and the lows are too much separated.

Good job there! Once I edited out the digitasl sound with my brain it was quite enjoyable. Others have commented on this extensively and I have nothing to add.

Thanks Didier for posting these settings. I'll be doing some experimenting based on this, but I don't have quite as fancy software as you... I use Nero WaveEditor.

My piano is a Kawai CN23 Digital Piano, however it only records mono. I then use the software to convert to stereo - figured that would be better somehow.. ??