You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The group=x polyphony=1 is useful in particular for preventing build-up of release samples during long trills in baroque keyboard music. As I now understand it, SFZ format specifies polyphony as a group code but clearly Sforzando can also interpret it just fine within a region and how I've seen it used in other instruments.
I now have a potential workaround from Paul by using "master" to create hierarchy enabling use of "group" per release note but it would be great for sfizz to be able to read existing patches that work flawlessly with Sforzando.
The text was updated successfully, but these errors were encountered:
What I do not understand is that when you do trills, you altenately release each note, so their release sample shouldn't "build up" but be played in turn ? Or are the release samples long enough that you really need the new release sample to steal the voice of the previously playing one (for the same note) ?
Just to be sure: accepting this change would make note_polyphony=<value> a synonym of group=<non-used-group-shared-by-all-notes-of-the-same-key> polyphony=<value> (ignoring note_selfmask).
That being said, I agree that it is a bug in sfizz and sfzformat. In the polyphony specification, you interpret "group" as <group> where I would interpret it as "all regions with the same group opcode" (which can be defined at <group> level, but doesn't need be).
Yes, the release samples are quite long...it is the characteristic soundboard resonance that can sound over several seconds, hence the crazy build-up on a trill and need for stealing per individual note. You have it exactly right!
I seem to remember trying note_polyphony= and it not behaving as I would expect (in Sforzando) so that's why I moved to group= polyphony= having seen the same method used elsewhere. I believe I need polyphony not note_polyphony because my harpsichord is not chromatically sampled so includes more than one pitch per group. I read something similar here: https://sfzformat.com/opcodes/polyphony
With Sforzando, I can use the following line without issue:
<region> sample=/Lower8/release/31.wav lokey=g1 hikey=g#1 pitch_keycenter=g1 offset=9594 group=1 polyphony=1
The
group=x polyphony=1
is useful in particular for preventing build-up of release samples during long trills in baroque keyboard music. As I now understand it, SFZ format specifies polyphony as a group code but clearly Sforzando can also interpret it just fine within a region and how I've seen it used in other instruments.I now have a potential workaround from Paul by using "master" to create hierarchy enabling use of "group" per release note but it would be great for sfizz to be able to read existing patches that work flawlessly with Sforzando.
The text was updated successfully, but these errors were encountered: