I couldn't initially reproduce the issue playing the file with MIDI sent from Cakewalk via physical MIDI loopback path to standalone Pianoteq 8 so it would appear as real-time MIDI input, But then I started experimenting with lower durations and found that very short, soft notes would not consistently trigger the key noise. With the velocity of all notes reduced to 2 the key noise became progressivley more intermittent below a duration of 120 ticks (960 ppq at 120bpm) until it was pretty consistently missing below a duration of about 100 ticks. I thought this might be related to Pianoteq's modeling of repetition level function, but the notes aren't close enough together to be considered a repetition (start times left at the approximately quarter-note intervals in the file). The sme thing happens with the Pianoteq 8 VSTi hosted inside Cakewalk with MIDI routed directly to it.

But, again, I don't hear the problem with notes of longer duration as in the original MIDI file regardless of velocity.

Thanks for sharing.

It's also related to internal sample rate. When internal sample rate exceeds 48kHz, you'll hear hammer noise losses using my midi file.

But, again, I don't hear the problem with notes of longer duration as in the original MIDI file regardless of velocity.

Here is audio samples and the midi file:

https://forum.modartt.com/uploads.php?f...corded.mp3

https://forum.modartt.com/uploads.php?f...ported.mp3

https://forum.modartt.com/uploads.php?f...20Test.mid