But that is not the way csound works. The istor
parameter only regulates whether the internal
memory buffer is cleared or not. It doesn't know
anything about previous note parameters etc.
When an instrument starts, it either allocates a
new block for itself or takes up an existing block.
The init routine can then make sure that its
dataspace is reset or not. The istor parameter
lets the user decide which.

This mechanism is mainly directed at 'tied'
notes, so that one instance can follow an existing
one gracefully.

Victor

>
> I was thinking that value was directly assigned, rather
> than Boolean. I wonder if it would be useful or possible
> to have it assignable; so that, using tab opcodes perhaps,
> one could use the previous value of note 1 for note 3 and
> that of note 2 for note 4, or whatever. -Chuckk
>
>
> On 11/11/07, Victor Lazzarini 
> > wrote: instruments take up existing allocated memory, if
> > there is any. So if you don't reset the dataspace,
> > previous values will still be there.
> >
> > >
> > > Reading some of the documentation on writing opcodes
> > > in C. Looking at the rms code in ugens5.h and
> > > ugens5.c.  Not very far yet. Question for anyone who
> > > knows: In rmsset I see
> > >
> > >     if (!*p->istor)
> > >       p->prvq = 0.0;
> > >
> > > But I don't see anything actually assigning the value
> > > of *p->istor to p->prvq.  Does this happen someplace
> > > else? Because in the rms function, p->prvq is assigned
> > > to q unconditionally, and this is used for
> > > computations.  The RMS struct (declaration?
> > > definition?) in ugens5.h doesn't initialize prvq
> > either. >
> > > I used this orc code:
> > > asig    init    20000
> > > krms    rms    asig, 10, 20000
> > >     printk2    krms
> > >
> > > and the output climbed to 20000 as it did when I left
> > > out iskip, rather than starting there as I expected.
> > > I didn't check if the output was exactly the same.
> > >
> > > -Chuckk
> > >
> > > --
> > > http://www.badmuthahubbard.com
> > >
> > > Send bugs reports to this list.
> > > To unsubscribe, send email sympa@lists.bath.ac.uk with
> > > body "unsubscribe csound"
> >
> >
> > Send bugs reports to this list.
> > To unsubscribe, send email sympa@lists.bath.ac.uk with
> body "unsubscribe csound" >
>
>
> --
> http://www.badmuthahubbard.com
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with
> body "unsubscribe csound"