Hello all,
I have been working with using an oparray to construct multi-staged
envelopes, where each stage of the envelope is a different state in the
oparray. I guess my problem is more a lack of understanding on how
oparrays work, and when things are happening. As well as this, I have a
question regarding arrays and their indices.
On Arrays:
What determines the rate of an array? Is it the rate of the array,
itself? Or does the index also play a part on determining the rate?
On Oparrays:
I included the code here, and it is long. I marked the points I have
questions about.
---
kopcode eg_table(ksig release, ivar ratescale, table rates, table
levels) {
ivar stages, hold, icounter, r[10], l[10];
ksig kinit, triggered, startlevel, eg, stage, expire;
oparray kline[10];
if (1) { stages = ftlen(rates); }
hold = stages - 1;
while(icounter < stages) {
r[icounter] = tableread(rates, icounter) * ratescale;
l[icounter] = tableread(levels, icounter);
icounter = icounter + 1;
}
if (!kinit) {
kinit = 1;
expire = r[stage];
startlevel = l[stages - 1];
}
if (!release && !triggered && (stage < hold)) {
/////// #1
eg = (kline[stage](1, r[stage], 0) * (startlevel - l[stage])) +
l[stage];
if (itime >= expire) {
startlevel = l[stage];
stage = stage + 1;
expire = expire + r[stage];
if (stage == hold) {
eg = startlevel;
}
}
} else {
if (release && !triggered) {
triggered = 1;
expire = itime + r[stage];
startlevel = eg;
}
if (triggered) {
if (itime < expire) {
eg = (kline[stage](1, r[stage], 0) * (startlevel -
l[stage])) + l[stage];
} else {
eg = l[stage];
}
}
}
return(eg);
}
---
#1 - here is the call to the oparray. The thing wrong here is the rate
of the second parameter to kline is required to be ivar. My questions
here are:
When are each of the instances of kline created? If they are created
at i-time of the enclosing instrument/opcode, then my code would have to
be changed to define each instance at that time. I think I know how to
write the code to create the instances, but what do I write to access
them (what do I use for the parameters on the second oparray call)?
If on the other hand it is legal to create each instance as it used,
then the only limitation is that of the base opcode called. This would
mean that I can't do what I want, as all the parameters for kline are
ivar. To get around this, I would have to create discrete variables for
each stage, and this defeats the whole purpose of what I want to do. If
it were legal to use an array of ivar variables, and still allow
indexing with a krate variable, then my code above should work.
Thanks,
Mike
This archive was generated by hypermail 2b29 : Mon Jan 28 2002 - 12:11:07 EST