[fluid-dev] Re: FluidSynth and DSSI

[Josh Green]

> On May 2, 2005, at 8:12 AM, Josh Green wrote:
> > As to the patch, I could look at it.  Your proposed changes sound OK
> to
> > me.  Currently I think FluidSynth is in need of a maintainer.  I 
> > haven't
> > had any time for it between my other projects (Swami/libInstPatch, 
> > etc).
> > If you send me the patch though, I'll do my best to review and
> commit 
> > it
> > to CVS.  Cheers.
> >       Josh Green
> 
> Excellent!  I've attached two patches -- the first is the stuff
> needed 
> by
> the DSSI plugin, and the second just fixes --enable-longlong for
> big-endian platforms.  Both apply cleanly against this morning's CVS.
> 
> Basically, the DSSI patch does these things:
> 
> - Adds the ability to change the polyphony limit after the synth is 
> instantiated.
>      new_fluid_synth() still allocates the number of voices indicated
> by 
> the
>      'synth.polyphony' setting, but then one may adjust the current 
> polyphony
>      limit from 1 up to that number of voices with the new
>      fluid_synth_set_polyphony() function.  new_fluid_synth() also 
> registers
>      a settings callback to handle when 'synth.polyphony' is changed.
>      Throughout fluid_synth.c, synth->polyphony is the current
> polyphony 
> limit,
>      and synth->nvoice is then only used as the total allocated voice 
> count.
> 
> - Exposes in the API fluid_synth_all_sounds_off() plus the two new 
> functions
>      fluid_synth_set_polyphony() and fluid_synth_get_polyphony().
> 
> - Changes fluid_cmd.c to allow setting polyphony from the cmd shell.
> 
> - Fixes fluid_synth_nwrite_float() to handle block lengths less than
> 64.
>      Without the patch, it assumes 'len' will always be 64 or
> greater, 
> though
>      it also happens to work if 'len' is always 32.
> 
> Let me know if you'd like me to tweak/rework anything.  Thanks,
> 
> -Sean
> 

Cool, I'll have a look at it and probably commit to CVS if Peter has no
objections :)

On a similar note, I'd like to add DSSI support to Swami as well.  This
is because it will in the future likely support multiple synthesizer
backends and other instrument formats (such as DLS, GigaSampler, etc) as
well.  Where can I find your DSSI work that you are doing?  I'd like to
see how it is done with FluidSynth so that I can perhaps use it as an
example when implementing this for Swami.  Cheers.
        Josh Green

signature.asc
Description: This is a digitally signed message part