Physically Modeled Piano

[TristezaOrange]

If it has no samples how come the download is 168MB? - that doesn't seem right.

Maybe it physically models samples being played by a sampler.

[robogone]

If it has no samples how come the download is 168MB? - that doesn't seem right.

Maybe it physically models samples being played by a sampler.

Maybe they're not very efficient coders...

[aMUSEd]

Doesn't this make it static sounding? I mean shouldn't the IRs actually be based on some mathematical formula for true physical modelling?

No it shouldn't - if the IR's are used as exciters they just provide the initial force that is then shaped by the physical model - the fact that they are sample based shouldn't make a difference necessarily. That is if they are using them the way they state - I did have my initial concerns too (as the above post shows) but in principle how they state they are using them is plausible in relation to what I know of physical modelling, mainly from discussions with people like Harm Visser. Note, this is not the same as a hybrid model.

[voltara]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

Here's a modified redc1.wyc file with all samples stripped. Instead, both velocity layers of the first key range (A0 and A#0, the lowest two notes on an 88-key keyboard) sample were replaced with a recorded female voice speaking the phrase "rear left". The rest of the keys are silent.

Download link: redc1.wyc (http://www.2shared.com/file/6461839/b8b76011/redc1.html)

[ZergFood]

No it shouldn't - if the IR's are used as exciters they just provide the initial force that is then shaped by the physical model - the fact that they are sample based shouldn't make a difference necessarily. That is if they are using them the way they state - I did have my initial concerns too (as the above post shows) but in principle how they state they are using them is plausible in relation to what I know of physical modelling, mainly from discussions with people like Harm Visser. Note, this is not the same as a hybrid model.

Yes but Impulse Responses are always static even for exciters -- in this case, the exciters would be static. (although it wouldn't make much difference since then the physical model will take over and probably introduce real acoustic variations etc)

PS by "static" I mean that when you play the same note it always sounds the same (even if there are other notes at the same time etc)

[aMUSEd]

No it shouldn't - if the IR's are used as exciters they just provide the initial force that is then shaped by the physical model - the fact that they are sample based shouldn't make a difference necessarily. That is if they are using them the way they state - I did have my initial concerns too (as the above post shows) but in principle how they state they are using them is plausible in relation to what I know of physical modelling, mainly from discussions with people like Harm Visser. Note, this is not the same as a hybrid model.

Yes but Impulse Responses are always static even for exciters -- in this case, the exciters would be static. (although it wouldn't make much difference since then the physical model will take over and probably introduce real acoustic variations etc)

That was my point

[voltara]

That was my point

Which is moot anyway, since the modified wyc file I uploaded demonstrates that it's just plain old sample playback, with maybe a little bit of filtering.

(Sorry to resurrect a dead topic.)

[Midiworks]