Re: [sc-dev] demand rate

[Charlls Quarra <charlls_quarra@xxxxxxxxxxxx>]

The control cycle latency would not be necessary if the synth is placed _later_  in the execution hierarchy than the spawning node.

 

The spawner code could look in general terms like this: (i dont remember the specific macros, sorry for that)

 

float *stackarray

for (i=0;i<blocksize;i++)

{

  if (prevtrig<=0 && *trig>0)

  {

     trig_synths++;

     stackarray[idx++]=i; //sample offset

     for (j=0; j<synthargs;j++)

     {

         stackarray[idx++]=**(zin+1+j); //setting synth arguments

         ++*(zin+1+j); 

     }    

  }

  else for (j=0;j<synthargs;j++ , ++*(zin+1+j));

 

}

 

//then you schedule all the synths with their offsets at the end of the block

(this assumes each synth can spawn only a given synthdef, but maybe its not difficult to make it work with any valid synthdef)

while (trig_synths-->0)

 {

   WorldCallToSpawnNodes->createNodeAt( group/node ID¿?, tail/head flag¿? , stackarray);

 }

 

stefan kersten <steve@xxxxxxxxxx> escribió:

an ar trigger could have a latency of one control cycle to
queue the synth(s) for the next period, but there's also the
possibility of a demand rate, subsample accuary scheduler
based on delta times (which would also have a latency, of
course).

---

1GB gratis, Antivirus y Antispam
 
Correo Yahoo!, el mejor correo web del mundo
 
Abrí tu cuenta aquí