Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

To be honest, as a pianist and physicist I’m still sceptical. The only thing you can influence at a mechanical keyboard is the speed with which the hammer hits the strings because by the time contact happens the escapement mechanism has removed the connection between hammer and key. One cannot influence its weight, momentum, angle, etc. So if all there is is a single variable (speed of the hammer), I find it hard to believe that one can change timbre independently on its own.


You are right to be skeptical: the study in question only looked at 'perceived timbre' in the context of pieces performed by pianists. https://www.pnas.org/doi/10.1073/pnas.2425073122


They should have looked at spectrograms at least -_-


In piano performance you often can hear the mechanical sound of the keys themselves as well as the breathing and humming noises from the pianist. Not to mention a performance will be played in an environment where environmental sounds will be correlated with the perception of timber and the pianists physical movements. I don’t see mention if they controlled for all these things.


eg: the Keith Jarrett / David Helfgott recording inversion, performances where you can often hear the notes of the piano over the sounds of the pianist.


But why would you not expect the speed to change the timbre? Lower speed = less force of the hammer hitting the strings = less chaotic excitation of the string during the attack phase = less distortion in the upper harmonics, as well as less sympathetic resonance.


I expect change in timbre, but not independent of intensity and all that changes with a difference in hammer speed.

If all they are saying is that there is a change in timbre tout-court, then it’s trivial. If they’re saying one can change the timbre while keeping the intensity constant, then I have issues.




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: