RE: [info-performer] Mixing OpenGL and Performer

New Message Reply Date view Thread view Subject view Author view

From: Dorosky, Christopher G (christopher.g.dorosky++at++lmco.com)
Date: 06/12/2002 06:56:48


Down this road lies madness, until you get it right.

My solution, was to write a glState class, with a push, and pop.

Before doing GL anything, I would call state->push(), which
push every single GL state I could think of onto an internal stack,
state->clear() which made sure bizarre options were turned off,
and that all matrices were the identity, etc.
Then I do the GL stuff
Then I do a state->pop() and restore whatever performer was doing.

Just be aware that you can have 9 out of 10 things right, and still
get no picture, until you manage to fix that last thing.

You would think pfPushState and pfPopState would help, and I may
even call them in my push and pop, but they certainly don't do it all.

Chris

-----Original Message-----
From: David Geldreich [mailto:David.Geldreich++at++sophia.inria.fr]
Sent: Wednesday, June 12, 2002 3:55 AM
To: info-performer++at++sgi.com
Subject: [info-performer] Mixing OpenGL and Performer

    Hello pfers,

    perhaps is it a RTFM question, but I would like to mix OpenGL code to
the performer pipeline in
order to be able to use nVidia Geforce shaders capabilities. Is there a good
way of doing it ? Which
are the pitfalls (save/restoring contexts, projection, transformation
matrices, ...) I would have to
take care about ?

    Thanks in advance.


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 13 2002 - 17:39:16 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.