Re: [LAU] Fluidsynth: map each channel to individual jack audio output

Hi Tito,

On 24/04/16 14:15, Tito Latini wrote:
> On Sun, Apr 24, 2016 at 03:13:42AM +0200, Lorenzo Sutton wrote:

[...]
>>
>> However the behaviour isn't as expected at least according to this (very
>> old post):
>>
>> https://lists.nongnu.org/archive/html/fluid-dev/2004-02/msg00027.html
>>
>> Ideally is there a way to have each channel routed to a separate audio
>> output on jack these days? Any help appreciated.
>
> '-o synth.audio-groups=4' fails: `fluid_settings_getint' returns 1
> after `process_o_cmd_line_option' (also tested with a printf).
>
> sh> fluidsynth -a jack -o synth.audio-groups=4 <<< settings 2>/dev/null \
> | grep audio-groups
> synth.audio-groups 1
>
> Idem with "-o synth.audio-channels=4 -o synth.audio-groups=4",
> but "-G 4" works:
>
> sh> fluidsynth -a jack -G 4 <<< settings 2>/dev/null | grep audio-groups
> synth.audio-groups 4
>
> There is a comment in `src/fluidsynth.c:594':
>
> /* The 'groups' setting is only relevant for LADSPA operation
> * If not given, set number groups to number of audio channels, because
> * they are the same (there is nothing between synth output and 'sound card')
> */
>
> tested with:
>
> sh> fluidsynth -a jack -L 4 <<< settings 2>/dev/null | grep audio-groups
> synth.audio-groups 4
>
> The follow example works (audio.jack.multi=no by default):
>
> sh> fluidsynth -a jack -L 4
> load /usr/share/sounds/sf2/FluidR3_GM.sf2
> select 0 1 0 0
> select 1 1 0 20
> select 2 1 0 30
> select 3 1 0 40
>

Thanks for testing... However I'm not really sure what to make out of
that apart from assuming that the use case isn't directly covered by
fluidsynth?

I imagine the easiest workaround would be starting multiple fluidsynth
instances, maybe lowering poliphony (not sure what impact this has on
resource consumption), possibly scripting it...

Lorenzo.

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-user
Received on Sun Apr 24 20:15:02 2016