Damper. Pedal, then key.

Again an issue I can't replicate, sorry.
Because you had chosen NY Steinway Model D I tested with this preset too (with factory defaults).

I hear the harp noise when I push the damper pedal. Releasing a key just triggers a sound like a hammer falling back. For me this is the expected behavior.

Pianoteq Standard
8.2.0/20240115 (Standalone)
Linux

Am I having another bug, when I can't replicate the bug that is on the todo list? ;-)

Keeping the damper pedal pushed and releasing a very softly pressed key (for example Note-On velocity 1) I hear just the virtual hammer falling back and the harp silently ringing (from the virtual "knock on wood" of the falling hammer I guess). What am I missing here? Or is the effect so subtle, that it can't be easily heard?

This needs some clarification. Are you talking about the sound of the "damper-on string" noise or the persistence of sustained notes?

Also, it's not clear to me what you mean by "during recording and subsequent playback". If it doesnt happen "in live play" I don't see how it would happen 'during recording" which involves live play. Or is it just on playback and export?

Also, if I understand correctly, this is all involving the standalone Pianoteq, not the VSTi, is that right?

Can you post recorded MIDI with a corresponding exported audio file that demonstrates the issue?

So far I have only tested it in the standalone version.  Under Linux.  I specially increased the noise volume of the hammers and dampers.  I used a standard preset with default settings.  I turned on the midi recording, pressed the pedal, slowly pressed the key.  Then, when played back in a midi player, it is discovered that the residual sound of the felt rustling across the strings when the pedal is pressed is cut off when the key is pressed slowly.  This occurs when playing back recorded midi and also when exporting to audio.  There should also be no rustling of the felt of the note being pressed when the pedal is pressed.  Example files are in the first post of this topic.

Okay, that's another story. You used a modified factory preset (in other words not a read-only preset from disk). I can hear it now when I max out Damper Noise (0 -> +24 dB) and Pedal Noise (0 -> +24 dB) and press a single note (Note 60, C4)

The characteristic of my pedal is a bit different from yours probably, because it has three fixed levels and not two levels as in so called "Half-Pedaling". The pedal noise is generated three times at level 38, 74 and 127. I remember that we had a discussion in another thread about the pros and cons of half, one-third and pseudo-continuous pedaling. A perfect solution does not exist currently if I remember correctly.

And yes, the odd thing is the recorded midifile behaves differently to live play (?!). On the recording the harp stops ringing when Note On velocity 1 arrives. My guess is just the single note is undamped in that moment - which shouldn't.

I would not expect this to make any difference in Pianoteq because its programming is presumably already instructing the model to undamp the string  before hammering it. And it doesn't hammer the string at velocity=1 so the attack generated by the second Note On should sound the same either way.

Also, I don't know what you've engineered into your keyboard, but my understanding of a typical triple-sensor implementation is that the middle sensor alone doesn't cause any message to be generated; it's just a reference point for resetting and measuring the velocity of a repetition. You should only get one Note On message from the first press of the key.

I interpreted him to be saying his self-designed triple-sensor controller sends two Note On events for each full (from the top) key press; one at velocity=1 as the key passes the middle sensor, and the second with velocity based conventionally on the transit time between the middle and bottom sensor. Becasue of the way Pianoteq is programmed, the first Note On will only release the damper, and the second will only hammer the string.The higher the middle sensor, the longer the time between damper release and hammer blow.