[CM] duration equivalent on snd-rt

[Juan I Reyes]

Thanks a lot Kjetil!  I thought of the scheduler but I had rather you
commenting the issue.

I might be generalizing but Physical Modeling gurus often take durations
between half seconds and five. This might work for delay lines or at
least to get things working.

Looks like I have to sit down and read your LAC-2008 paper because
syntax is a bit different. And speaking of code simplification, wow!.

 --* Juan

> 
> Just in case there is a misunderstanding: Snd-rt has a scheduler, so the 
> length of the instrument can be set.
> 
> 
> (<rt-out> :dur 1 3
>            (oscil))
> 
> is the same as:
> 
> (definstrument os (start dur)
>    (let ((o (make-oscil)))
>      (run
>        (loop for i from (* *srate* start)
>                    to   (+ (* *srate* start)
>                            (* *srate* dur))
>           do (let ((val (oscil o)))
>                (outa i val)
>                (outb i val))))))
> 
> (with-sound () (os 1 3))
> 
> 
> 
> > Do I average or is there
> > like a system handling mechanism which takes care of this issue?. 
> > The reason for this question is that in the case of delay-lines, lengths
> > are pretty much function of durations and sampling-rate.
> >
> 
> I guess that depends on the situation...