My interpretation of Schubert Impromptu op 90 no 2

ranjit

This is finally getting to a point where I'm pretty happy with the interpretation. Some of the technique requires work. Please let me know what you think!

pseudonym58

How could I have gone through 60+ years of my life thinking this was a piece by Chopin?! 🤦🏻‍♂️

ranjit

pseudonym58 Haha, it does sound kind of like dramatic Chopin, in a way.

pianoloverus

This is quite good. 'Here are two things I think you can improve:

The triplet runs are not always even. Sometimes there seem to be hesitations and sometimes notes are run together.
Interpretively, I think the middle section should be much more dramatic and, in general, much louder. It starts out marked ff but you play it mostly at the same or only slightly greater dynamic than the A section.

pseudonym58

ranjit Especially like one of Chopin's Etudes.

TheBoringPianist

Very nice playing! You're doing many things well. I'm not going to talk about technique — you know where you need to practice — but here are some thoughts that could take it to the next level (I seem to recall you wanting more direct feedback — if that is not the case, I apologize). Of course, you're welcome to take or leave as much as you want:

I notice that you often speed up in spots where you're (a) more comfortable technically, (b) making a crescendo (m. 64), or (c) there's less going on in the LH (e.g. m. 35). Try to resist this temptation! It will help with the overall feeling of smoothness.
When you want to have a change in dynamic, you may need to set it up. For example, you're not really leaving yourself much room to go softer/darker at m. 25 because you've been playing soft the entire time so far. To give yourself that room, you can get louder in the preceding section e.g. starting at m. 17 so you have room to drop down at m. 25 (this also helps the preceding sections feel less repetitive). You could even una corda there for a color change as well.
While I think it's totally fine to slow down in m. 51/219, I would connect the last beat of that measure to the beginning of the next measure (like the upbeat at the beginning of the piece) so the transition feels more natural and momentum carries over
Agree with @pianoloverus that the middle section can be much more dramatic. I would probably also use less pedal. (Pedantic note: does your score have some LH downbeats marked with a staccato and not others? It may be worth considering why that might be, even if the conclusion you reach is that maybe he just got too lazy to write out all the staccatos) I also think you can bring out those crunchy LH chords more in m. 103/107. Even outside of the middle section, you can surge more when you have stuff like the fps in m. 37/39 or the fzs in m. 71–79, etc.
You lost all your momentum too early in m. 163–164. I would keep it growing in intensity all the way to m. 165 (which is the biggest moment of harmonic tension in the piece before we're rewarded with the return to the main theme)
In the coda, you've already started accelerating by m. 259... but the accelerando isn't marked until m. 267... so you were left with no more room to speed up. Generally speaking, if you have an accelerando/crescendo/etc., it'll sound most exciting if you increase the rate of speeding up over time (exponential curve rather than linear or log), so resist the urge to start speeding up right away

Hope at least something in there was useful!

ranjit

TheBoringPianist That's a lot of useful information, thanks a lot!

BartK

Great work! I really like this piece.

I think you really don't need to go that fast and it would sound much better at a slightly slower tempo but with more even triplets. Better clarity will make it sound faster than it really is.

If you haven't done so yet, watch this masterclass:

Nightowl

BartK I agree that this piece would sound better at a slower tempo. I changed the setting to 80% speed when I listened on YT and it sounded much better - but maybe that's just my preference and it's meant to be played at breakneck speed.

ranjit

Nightowl Interesting, I changed the tempo to 75% and you have a point. I don't think I really realized how fast I was playing, and you tend to get accustomed to how your own recording sounds, so I kind of missed how it might sound better a bit slower.

As I think about it more, my recording is almost exactly the same length as that of Zimerman (4 min 46 sec). That suggests that I was probably relying on my auditory memory of Zimerman's recording to a decent extent.

Nightowl

ranjit I'm glad I didn't offend you. 🙂

I'm in no position to judge such advanced playing and I know I could never play at such speed, but if you slow down slightly you'll be able to finesse your playing and improve little things that other advanced pianists notice.

pianoloverus

I agree with those who think the piece could go more slowly. You should definitely practice it much more slowly because, other than the interpretive change needed for the middle section which is easy to deal with, IMO the most serious issue is the unevenness and lack of articulation in the opening and closing sections.

If you are comparing timings with other pianists, you should probably just compare timings on the opening section. If you compare timings for the entire piece and your timing is similar to another pianist's timing, that might be because you're playing the opening section faster but the middle section slower. Or just compare metronome speeds.

ranjit

Nightowl Sure, that said when I perform I try to play it whatever tempo I fancy at that moment, and I will admit that I was trying to figure out if a faster tempo would work at the time I made this recording. I generally have a preference for faster tempos (I think they are underrated because most people think the only reason people play fast is to show off, which is far from the case imo). I still need to work on evenness etc. to make this work though. And I'll try to come up with a slower recording too, just to balance it out. 🙂