[chuck-users] trying to implement a midi sequencer in chuck

Sat Feb 12 15:55:25 EST 2022

As for the timing/yielding questions, I already wrote up this little
example so I thought I would share it. Maybe it clarifies some things,
maybe it doesn't.

// plays each note in notes[] for the given duration
function void play(int notes[], dur length) {
  SinOsc s => dac;
  for (0 => int i; i < notes.size(); i++) {
    Std.mtof(notes[i]) => s.freq;
    <<< "playing note", notes[i], "and waiting for", length, "samples
before code execution continues" >>>;
    length => now;
    <<< length, "samples have gone by, so now we continue" >>>;
  }
}

spork ~ play([60, 62, 64, 66, 68, 70, 72], 500::ms);

// uncomment this spork to see how the shreds run in parallel
// spork ~ play([72, 71, 70, 69, 68, 67, 66, 65, 64, 63, 62], 250::ms);

<<< "waiting here for an arbitrary amount of time because if we don't, the
sporked shreds won't get a chance to do anything before this process ends",
"" >>>;
day => now;

On Sat, Feb 12, 2022 at 12:41 PM d <dchuck at yurfish.com> wrote:

> THANKS Dana!
> I've been struggling with this for a couple
> of weeks and you fixed it in one post!
>
> I'm planning to explore "conducting"
> midi scores that I produce in lilypond.
> Not sure where that will go, but this gives
> me a lot of possibilities to mess with the
> events before they go out.
>
> The MidiScoreReader in the source looks interesting
> but i don't know to access it. Track metadata
> would be nice to have.
> I may just read tracks into arrays and work with that.
>
> davud
>
>
>
>
> On Sat, 12 Feb 2022 12:23:55 -0800
> Dana Batali <dana.batali at gmail.com> wrote:
>
> > How about moving the MidiMsg object into the track function?
> > Of course it could always be a bug in MidiOut (I haven't used it.)
> >
> > On Sat, Feb 12, 2022 at 12:18 PM d <dchuck at yurfish.com> wrote:
> > >
> > > Im using jack with fluidsynth.
> > >
> > > I had some confusion about whether each
> > > spork had its own "now" but I did a
> > > test script and that seems to be the case.
> > >
> > > I have tried adding me.yeild in various
> > > places but no joy.
> > >
> > > I created a midi file with a triad in
> > > in three tracks ..8 quarter notes.
> > > The events seem to be correct
> > > but the output is only the last
> > > spork. Perhaps its a problem
> > > with my setup.
> > > If i play the file with fludidynth
> > > I get what I expected.
> > >
> > > Any suggestions appreciated.
> > > thanks
> > > -----
> > > MidiFileIn min;
> > > MidiOut mout;
> > > MidiMsg msg;
> > > string filename;
> > > me.arg(0) => filename;
> > > if( !min.open(filename) ){
> > >    <<< "unable to open MIDI file:'",filename >>>;
> > >     me.exit();
> > > }
> > > mout.open(0);
> > > spork ~ track(1);
> > > spork ~ track(2);
> > > spork ~ track(3);
> > > 5::second=>now;
> > >
> > > function void  track(int t) {
> > >     10::ms =>now;
> > >     while( min.read(msg,t))
> > >     {
> > >   <<<t,msg.when,msg.data1&0x0f,msg.data2,msg.data3 >>>;
> > >       10::ms =>now;
> > >          msg.when=> now;
> > >          mout.send(msg);
> > >          me.yield();
> > >     }
> > > }
> > >
> > > 1 0.000000 0 75 0
> > > 2 0.000000 1 75 0
> > > 3 0.000000 2 75 0
> > > 1 0.000000 0 75 0
> > > 2 0.000000 1 75 0
> > > 3 0.000000 2 75 0
> > > 1 0.000000 0 60 90
> > > 2 0.000000 1 64 90
> > > 3 0.000000 2 55 90
> > > 1 18000.000000 0 60 0
> > > 2 18000.000000 1 64 0
> > > 3 18000.000000 2 55 0
> > > 1 0.000000 0 60 90
> > > 2 0.000000 1 64 90
> > > 3 0.000000 2 55 90
> > > 1 18000.000000 0 60 0
> > > 2 18000.000000 1 64 0
> > > 3 18000.000000 2 55 0
> > > 1 0.000000 0 60 90
> > > 2 0.000000 1 64 90
> > > 3 0.000000 2 55 90
> > > 1 18000.000000 0 60 0
> > > 2 18000.000000 1 64 0
> > >
> > >
> > > On Sat, 12 Feb 2022 10:22:17 -0800
> > > Forrest Curo <treegestalt at gmail.com> wrote:
> > >
> > > > So how is this going to change their timing? It shouldn't, unless
> > > > some process somewhere keeps running and won't yield when the
> > > > wait ends (and the waiting spork wants to start from there.)
> > > >
> > > > Jack or fluidsynth -- or what?
> > > >
> > > > On Sat, Feb 12, 2022 at 9:55 AM Curtis Ullerich
> > > > <curtullerich at gmail.com> wrote:
> > > >
> > > > > "advancing time" means "yield for this long." so the other
> > > > > shreds will be executing, or yielding, as their own timing
> > > > > dictates.
> > > > >
> > > > > On Sat, Feb 12, 2022, 06:40 Forrest Curo <treegestalt at gmail.com>
> > > > > wrote:
> > > > >> "Advancing time on a spork" means to "Wait until this amount of
> > > > >> time has elapsed."
> > > > >> If one spork is waiting for time to elapse, the others will be
> > > > >> doing... what?
> > > > >>
> > > > >> On Fri, Feb 11, 2022 at 1:36 PM d <dchuck at yurfish.com> wrote:
> > > > >>
> > > > >>>
> > > > >>> Yes,thats the one I modeled mine after.
> > > > >>> I removed the ugen stuff and replaced it with
> > > > >>> send (midiout).
> > > > >>> I have fluidsynth on jack The behaviour is strange.
> > > > >>> It appears that advancing time on a spork advances
> > > > >>> time in others and events get missed
> > > > >>> But I could be wrong. I have notes with
> > > > >>> no note off and CC that never happens
> > > > >>>
> > > > >>> It works fine for single track files.
> > > > >>>
> > > > >>> On Fri, 11 Feb 2022 13:10:20 -0800
> > > > >>> Dana Batali <dana.batali at gmail.com> wrote:
> > > > >>>
> > > > >>> > Hi Davud,  have you seen this example?
> > > > >>> >
> > > > >>> >
> https://cannerycoders.com/docs/chuck/examples/midi/midifile-play.html
> > > > >>> >
> > > > >>> > One obvious point is that min.read accepts a track number.
> > > > >>> > _______________________________________________
> > > > >>> > chuck-users mailing list
> > > > >>> > chuck-users at lists.cs.princeton.edu
> > > > >>> > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
> > > > >>> >
> > > > >>>
> > > > >>>
> > > > >>> On Fri, 11 Feb 2022 13:10:20 -0800
> > > > >>> Dana Batali <dana.batali at gmail.com> wrote:
> > > > >>>
> > > > >>> > Hi Davud,  have you seen this example?
> > > > >>> >
> > > > >>> >
> https://cannerycoders.com/docs/chuck/examples/midi/midifile-play.html
> > > > >>> >
> > > > >>> > One obvious point is that min.read accepts a track number.
> > > > >>> > _______________________________________________
> > > > >>> > chuck-users mailing list
> > > > >>> > chuck-users at lists.cs.princeton.edu
> > > > >>> > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
> > > > >>> >
> > > > >>>
> > > > >>> _______________________________________________
> > > > >>> chuck-users mailing list
> > > > >>> chuck-users at lists.cs.princeton.edu
> > > > >>> https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
> > > > >>>
> > > > >> _______________________________________________
> > > > >> chuck-users mailing list
> > > > >> chuck-users at lists.cs.princeton.edu
> > > > >> https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
> > > > >>
> > > > > _______________________________________________
> > > > > chuck-users mailing list
> > > > > chuck-users at lists.cs.princeton.edu
> > > > > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
> > > > >
> > >
> > > _______________________________________________
> > > chuck-users mailing list
> > > chuck-users at lists.cs.princeton.edu
> > > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
> > _______________________________________________
> > chuck-users mailing list
> > chuck-users at lists.cs.princeton.edu
> > https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
>
> _______________________________________________
> chuck-users mailing list
> chuck-users at lists.cs.princeton.edu
> https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/chuck-users/attachments/20220212/cf63da34/attachment-0001.html>