In certain points, I don't understand the initial posting. One thing is
clear to me - he doesn't want to construct a soundfonts ("Going by SF2
is going to be hard for developer to implement new soundfont"). It
appears he wants just to import sample files on button press, and start
playing them. That means for FS, the sample player shall construct the
soundfonts.
E. g., I'm not quite sure why *wave* support shall be added to
FluidSynth. SF2 is based on wave data files. Nowadays, we're able to
import 24bit wave samples into the soundfonts, so the sound quality as
such cannot be the reason for that request. So that is an obvious
misinterpretation.
I can understand the request for flac format, since it is lossless
(compared with wave), and saves about the half of hard disk space. It
might even be, once in future, FLAC will replace wave (if the CD
industry allows for this ;-).
FluidSynth is soundfonts player only, so in regards of FS, a request can
be only, to enable soundfonts contruction from native flac sample files
(IIRC Swami has planned so for future). FluidSynth as such could be made
fit for reading such soundfonts, indeed. But in that case, it would be
at least frome equivalent interest (in my eyes), to support soundfonts
files based on Ogg Vorbis or a similar HQ compression format. This would
be a real improvement, since the soundfonts could be reduced
impressively in file size, while keeping HQ sound.
David, does FS support playing soundfonts in a bit-ness higher than
16-bit? Another question in regards of FS, would be support for up to
96-bit soundfonts.
I've never seen any soundfonts where the constructor has has fully
reached the sf2 format's potential, and my point of view is that
soundfonts are future safe.
BR
Bernd.
Am 04.02.2015 um 20:53 schrieb David Henningsson:
On 2015-01-31 07:36, James Lei wrote:
Going by SF2 is going to be hard for developer to implement new
soundfont, Native Instrument has supported Wav and Flac files and there
are plenty audio loops (http://www.producerloops.com) on the web.
FluidSynth should consider implement support for these improvement.
I'm not sure if wave and FLAC files makes much sense for FluidSynth.
You would still need some kind of framework to tell how these wave
files should be played depending on what note you press and how hard,
etc. And the SF2 file format is just such a framework.
As for other frameworks (such as SFZ), I think it would be nice if we
could support that. But it would be substantial effort to implement
that; hence nobody has attempted it so far. Also, if we would bring
more than SF2 in, I'd like more people to chime in as well.
// David
_______________________________________________
fluid-dev mailing list
fluid-dev@nongnu.org
https://lists.nongnu.org/mailman/listinfo/fluid-dev
_______________________________________________
fluid-dev mailing list
fluid-dev@nongnu.org
https://lists.nongnu.org/mailman/listinfo/fluid-dev
