Bug or real modelling? (I guess bug)

Perhaps these are due to the upper ranges of pianos being undamped, so they will resonate both with and without the sustain pedal being depressed.

Yes. This is a normal behavior for acoustic piano.

Well, I didn't find how I can upload midi and preset options. Here is the secong recording, it seems more to be a bug. If you press sustain, then with it sustenuto without releasing sustain, the sound becoms too resonant for a while
https://soundcloud.com/axisforward/2020...11-seconds

Ant. Petrof prelude modified (room size 19 m)

I'd like to give an example where you want to use 2 pedals simultaneously.  Watch Chopin's nocturne 48 n1 https://www.youtube.com/watch?v=-7mntyrW3HU time code 1:50  4-5 bar. You sostenuto C-C octave whilst sustaning thirds, then you release sustain to shut down thirds and play C-E-G-E on the refreshed space and lasting C-C octave as Chopin intended. There is another way to do it but I like this one.

I'll clarify that I release all pedals before I press some notes that have this reverb effect for some time. You can hear in the second link that the first playing chord before all pedals is clear and shuts down quickly after lifting up the keys.

I have found that not only notes that are obertones have this reverb slowly dying effect. This is definetly not the natural reverb or something that was supposed to be, as this dying reverb sound is very loud and can not be found before I do this trick with the pedals. Here is 1 more example https://soundcloud.com/axisforward/2020...12-seconds

I did not say "two pedals simultaneously never make sense". In fact, if you press a chord, then press sostenuto, thus holding those specific dampers up, then use sustain for lifting/dropping all the other dampers while holding sostenuto all the time to keep the original chord's dampers up, that can absolutely make sense. But pressing the sostenuto pedal while/after the sustain pedal is pressed down does not make much sense, as it just acts as a copy of the sustain pedal, i.e. it will just be holding all the dampers up.

That said, I have identified an artifact I too would classify as a bug, but it depends on using the two pedals in exactly the bizarre fashion I described:
1. Press the sustain pedal
2. Press the sostenuto pedal while still pressing sustain
3. Release the sustain pedal
4. Play a chord; sostenuto acts just like sustain now, which is to be expected
5. Release the chord and the pedal
6. The main string sounds are dampened immediately, but the sympathetic resonance simulation actually seems to linger for about half a second before it too is abruptly cut off

This behaviour is indeed incorrect, as in this specific scenario releasing sostenuto should sound exactly like releasing sustain. But I still maintain this is a bizarre corner case of little practical import. It would be nice if it got fixed by Modartt, but I wouldn't call it exactly urgent either.

Addendum: I have two sets of Kontakt-based sample instruments (Session Keys and Native Instruments), and I played around with them for a bit and neither is getting pedalling completely right. The SK instruments have a completely broken implementation that actually drops the dampers on all the strings when you press sustain while sostenuto-ing specific keys (i.e. play/hold a chord, press/hold sostenuto, release the keys, press sustain (while still holding sostenuto) and immediately hear all notes being dampened, it's quite fascinating). The SK implementation also cuts off repeat notes when pressing sustain after the first key press, which is a very old sustain bug many electronic instruments used to have. Both implementations also ignore the dampers being raised by sustain, i.e.: press/hold sustain, play/hold a chord, press/hold sostenuto, release keys, release sustain and magically only the dampers for the chord stay up, which is not what an actual physical sostenuto pedal would do, as it IMHO simply cannot for mechanical reasons. Maybe this is what he OP expected to happen.

Without addressing your issue surrounding unwanted "obertones", as a matter of historical fact & interest Chopin never had a sostenuto pedal on any of the instruments he used, and thus obviously never used it in any of his works.  If you're using it in the playing of his music then I'd suggest that there's something seriously amiss with your interpretation/technique, unless you believe that you're somehow a superior pianist to Chopin, in which case there would seem to be other issues at play...

edit:
PS by "you" I don't necessarily mean you, Romariozen, per se but rather anyone (including Arthur Rubinstein) who thus mangles Chopin.
PPS for anyone who's interested, here's an informative thesis on the development of the "modern" (ie 1875 Steinway-stye) pedaling setup: https://etd.ohiolink.edu/!etd.send_file...1428503747, as well as Sandra Rosenblum's critical discussion of Rowland's History...: https://scholarship.claremont.edu/cgi/v...ontext=ppr

OT we can agree on, but that's about it.
At issue is not "intention", but rather an understanding, or an apparent lack there of, of the parameters (as given by some 'musical author') by which one might arrive at a reasonably musically-informed interpretation of said author's own works.  "The right of an artist" is just the sort of juvenile, self-entitled approach that reveals either, at best, a gross lack of education or, more likely, a willful ignorance born out of hubris, both of which seem profoundly disrespectful of the true authority which such masters as Chopin, particularly as concerns the performance of their own works, have been so good as to bequeath us...  unless you feel that Chopin (et al) and his instructions as we have them (ie the score) are no real authority at all?  that all those indications provided by the score (slurs, dots, wedges, etc) are really just window dressing which he 'randomly put in there, and your own sensibilities are actually superior?  Or perhaps it's just the notion of authority at all which rankles you?  Further, your comparative example from the world of literature is apples-to-oranges... but in that vein, what if "my interpretation" tells me that Goethe's or Shakespeare's text should be changed—instead of "to be, or not to be (etc)" to something like "yo, izzle or no-shwizzle, that's the jamb"—because it's my "right" as an "artist"... 'that OK with you?  I realize it's the 21st century and, in the spirit of being "woke", anything and everything is fair game in these decadent days, and the more irreverent the better it would largely seem, but (call me old-fashioned) I prefer to study the works of the great composers and see what they can tell me, and not the other way around.  As well, your notion that trying to respect and learn from the composer's instructions beyond mere propaedeutic intervalic & rhythmic content somehow ossifies the work, inhibiting its >Immerwerdenheit< (as Adorno would have it), is simply quatsch.
As for Bach, I'm not at all sure that "pretty much everybody agrees" on appropriate choice of instruments, unless by "everybody" you mean the swath of "musicians" who still, surprisingly (or not?), have their prejudiced heads stuck back in the 20th century, marketing victims of the notion that the 1875 Steinway is the end-all-be-all of keyboard development...

Emphatically agreed.

1 more
https://soundcloud.com/axisforward/2020...10-seconds

https://forum.modartt.com/uploads.php?f...281%29.mid

Hi all,

We have released Pianoteq version 6.7.2, which should fix this issue with the sostenuto pedal.

Well... we have updated version 6.7.2 again because the previous fix broke the sostenuto in a different way -- sorry for that.