[PD] GEM repeat + where is the bottleneck?

[chris clepper]

You are resetting the GPU matrices millions of times per second, which is
probably not in the design goals of the manufacturer (to say the least!).
Maybe you need to look into writing your own pix_ object or some of the
vertex_ objects.

On Sun, Oct 26, 2008 at 9:21 AM, rainer _ <jepoxy at hotmail.com> wrote:

>  Hi list,
>
> I´ve done a testpatch using repeat from zexy libary to generate full PAL
> resolution pixel by pixel in one rendercircle. Unfortunatly this is very
> slow on my machine (dualcore+windoze) about one frame every 4 seconds. I
> also have tried this on a brand new Power Mac 8core machine - suprisingly it
> was´nt much faster. CPU load on the mac was only at 10%....
>
> Any suggestions where the bottleneck is ? Does it all depend on the grafic
> card ?
>
> Note: you need zexy and maxlib - also you need to load maxlib before gem to
> use the maxlib`s scale object. This patch is using the old counter objects -
> you probably need to reconnect the counters. Connect the "done bang" to the
> inlet of the second counter.
>
>
> Thanks,
> Rainer
>
>
>
> ------------------------------
> Connect to the next generation of MSN Messenger  Get it now!
> <http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline>
>
> _______________________________________________
> Pd-list at iem.at mailing list
> UNSUBSCRIBE and account-management ->
> http://lists.puredata.info/listinfo/pd-list
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.puredata.info/pipermail/pd-list/attachments/20081026/d634097e/attachment.htm>