[PD] double precision + benchmarking

Sukandar Kartadinata sk at glui.de
Thu Apr 4 12:34:09 CEST 2002

>Yes, right, originally I wanted to stay close to the original,
>but it prooved to be to much hassle to do so, at the end
>the iPaq has other limitations too, not only floating point but
>screen size, fonts etc. ...

:) luckily I could strip those last two...
however, one option I had in mind was to _control_ the gluiph from a 
PDA, although my idea of a musical instrument really is not to stare 
at a GUI

>Well, I just replaced the time stuff int the scheduler loop and
>sys_getrelatime()  to int.
>So I left the clock_ functions the way they are, at least something
>I have tio think about yet.

hmmm, but if you stick with the default TIMEUNITPERSEC of 32*441000 
you have overflow after 304 seconds, no ?

>If I understood you right, your problem isn't floating point in general
>but double precision ?

yep, that's right, my floats are just as fast as my ints

>What can be seen by this setup is that, once cleaned up, it should
>theoretically be easy toswitch from float to double precision within
>computations. (Well not for the "magic" oscillators implementation).

indeed. and as I said, if I can help with a cleanup, let me know 
Miller. Partially I already had to do it anyway, and if it helps more 
people than myself, maybe it should find its way into the canonical 
(adopting CSound speak here) release.



sukandar kartadinata
   the glui performance hardware

More information about the Pd-list mailing list