This thread reminds me: I would
really like to release GeneralUser GS under an open-source license
(GPL, etc.). Right now the license is just one I made up, but I
understand that prevents GeneralUser from being included with
open-source projects, so I would like to change that.
I am very new to the whole licensing thing. Does anybody have any
recommendations for a good license to use? If there are future
contributers to GeneralUser, I would still like to be able to have the
final say in what changes make it in or not. I also know that some of
the samples in GeneralUser are borrowed from the sample banks
Creative/E-MU has provided for free (shipping with the sound cards, or
on their website as free downloads), and while many other samples come
from free banks online, there is no way for me to be 100% sure that
some of them didn't come from copyrighted sources. I don't know how
this plays into an open-source license.
-~Chris
address@hidden wrote:
Quoting David Henningsson
<address@hidden>:
address@hidden skrev:
The thread concerning using FluidSynth as
an internal sampler, made me
think of something. FluidSynth should really try and auto-configure
itself as far as the audio drivers are concerned (if none are
specified). If one fails, try another, until one works. It should
also
have a decent default GM/GS SoundFont which is used by default if
nothing is specified.
But if we did, which one? I just updated the
http://fluidsynth.resonance.org/trac/wiki/SoundFont
page to list two soundfonts we know of, but I think they're both too
large to be included in a standard release, and I doubt that a smaller
one will sound good enough.
Perhaps we could just have FluidSynth look in a standard path. I see
on Ubuntu that the fluid-soundfont-gs and fluid-soundfont-gm installs
FluidR3_GS.sf2 and
FluidR3_GM.sf2 in /usr/share/sounds/sf2. We could add other typical
paths and SoundFont files present in other distros as needed. We'd
probably only want it to fallback to loading a default in the case
where MIDI file playback is requested without a SoundFont, since the
FluidR3_GM.sf2 is pretty big, if for some reason it got loaded when the
user didn't want it to. Perhaps we could add some API too.
A system wide config would also help. This
would
allow for FluidSynth to be more automatic, when used by other
applications which just want to play MIDI events, etc. Just a thought.
May need some API additions.
As for API additions, I'll prefer just an "auto" audio driver. The
problem, however, is to detect whether a sound driver is actually
working or not. Or perhaps, we could just check if
new_fluid_audio_driver fails or succeeds.
Sounds good. I think most audio drivers will just fail, if for some
reason they aren't available. The init could happen in a certain order
(better and/or more common audio drivers first).
// David
Josh
_______________________________________________
fluid-dev mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/fluid-dev
|