[PD] GEM questions

chris clepper cgc at humboldtblvd.com
Wed Feb 15 23:54:25 CET 2006


You can try http://taproot.dyndns.org/~cgc/downloads/Gem.dll but I
make no guarantees that it will work better (or at all).  On my test
box it does have the QT crash fixed, pix_record and pixel shaders
working.  You might want to stay away from openpanel because that
makes Pd evaporate for some reason (not crash, just disappear).  Good
luck.

As for point c) - a 0 or 1 will turn pix_objects off and back on.  The
default is on.

cgc

ps - where's the gig?

On 2/15/06, David Powers <cyborgk at nocturnalnoize.com> wrote:
> Hello everyone,
>
> I am working with GEM on a Windows XP, using Hans' extended release.
> This Saturday, some guys from Perlon record label will be here, and I am
> going to do some visuals. I have some stuff made in a program called
> Visual Jockey that I can use, but I would really like to do some stuff
> in Gem. So, I have these questions:
>
> a) What known issues are there, that might make patches crash, that
> would be best to avoid using in patches? GEM does crash more than PD! As
> far as I know, I'm probably best off NOT using any movie clips, as
> loading clips seems to be the biggest issue causing crashes and hogging
> CPU also (both AVI and QUICKTIME as far as I can tell)...
>
> b) Does anyone have advice on organizing patches and rendering chains
> for a show? This is the hard part for me, I can get individual rendering
> chains that look cool, but I'm worried about organizing and switching in
> some kind of meta-patch. Is it better to have one big patch with sub
> patches, or to open seperate patches at once?
>
> c) On a related note, I notice that effects I might put in the render
> chain don't have ON/OFF, there is just ON/OFF for entire rendering
> chain. So, suppose I have a pix_multiimage that I want to run through a
> couple of different effects - not simultaneously. Is the best strategy
> to just have the same pix_multiimate in two different rendering chains,
> with a selection interface?
>
> Thanks,
> David




More information about the Pd-list mailing list