Ahh, and I thought that was going to be the golden commit.
Quoting Rui Nuno Capela <[email protected]>:
re. svn trunk r278
fluid_synth_unset_program() seems to do its deeds now, but ...
that fallback logic needs work still. or else. for example, simple
soundfonts which define fewer instruments/presets than the number of
available channels, are now set initially unassigned, as reported by
fluid_synth_get_channel_info().
I'm not seeing that. I just loaded up a SoundFont in QSynth 0.3.4
with only one preset and it was assigned to every channel, except 10
which is unassigned due to lack of percussion preset. Program/preset
0/0 should get assigned to every channel except 10. Am I
misunderstanding what you are seeing?
furthermore, any attempt to assign any
other bank/program to a channel, via fluid_synth_bank_select() and
fluid_synth_program_change(), are strangely refused or will fallback to
unassigned or default preset quite immediately.
Weird. I'm not seeing that in QSynth 0.3.4 though. I tried a
SoundFont which has 2 banks of instruments and was able to select from
other banks just fine, at least according to the channels dialog.
one other thing, more like a question, is fluid_synth_program_reset()
really supposed to reset all channels to the default bank/program
assignment and disregard any explicit or discrete bank/program change
made so far? i am not sure, but i think that is in complete disruption
with older behavior or semantics from pre 1.1.0
fluid_synth_program_reset() is supposed to work just as before, that
is (from the API reference):
Resend a bank select and a program change for every channel.
So it should just resend the bank and program which is already
assigned, which might cause a new preset to become assigned if the
loaded SoundFont stack or bank offsets have changed. If you are
seeing something else, then that is a bug.
cheers
--
rncbc aka Rui Nuno Capela
[email protected]
Do you have some updated QSynth code I could test with? Perhaps there
is something going wrong with some of the new functions or something.
Or maybe I didn't understand what exactly you are experiencing. Do
you have any reproducible steps I could use to observe the problems
you are seeing?
Best regards,
Josh
_______________________________________________
fluid-dev mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/fluid-dev
