The most significant changes between 1.1 and 1.1.1-rc1, whcih might be relevant, are:
1) Changes to osg::State so the the function pointers to the vertex array calls glClientActiveTexture,
instead of these being statically initialized, they are now initialized to 0 in the constructor of each
osg::State and set when first required. Perhaps there is a path where they aren't initilized?
2) Changes to the way the per context texture objects/dislpay lists and extensions are managed,
changing the code so that it scales better and avoid race conditions that were causing problems under OSX.
Items 1 would mean just reverting include/osg/State and src/osg/State.cpp, while item 2 would be much more extension in changes... so I'd revert the item 1 changes first test things out, if that doesn't change things they look item 2 changes to Drawable/Texture etc.
Robert.
On 8/27/06, Farshid Lashkari <[EMAIL PROTECTED]> wrote:
Hi Robert,
On 8/26/06, Robert Osfield <[EMAIL PROTECTED]> wrote:
> No clues the cause, but I can reproduce the problem here. My guess is that
> some state is no being set by the text in the HUD, and being inherited from
> the state in the scene.
I didn't notice this problem in version 1.1. Were there any changes
made to osgText or RenderStage since then? I commented out the code in
osgcubemap that applies the TexEnvCombine attribute to the stateset
and the text appears fine. Do you think this attribute is somehow
being inherited by the text? In my application the text objects are in
a completely separate subgraph, so shouldn't they not be affected by
this attribute?
-Farshid
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
_______________________________________________ osg-users mailing list [email protected] http://openscenegraph.net/mailman/listinfo/osg-users http://www.openscenegraph.org/
