Hi, The patch looks good according to the EXT_separate_specular_color extension, but I'm not sure about the case where the vertex program writes result.color.secondary.
I'd like to get someone to confirm this before committing. On 9/8/07, Tommy Schultz Lassen <[EMAIL PROTECTED]> wrote: > Hi > > I have been playing around for some time with NeverwinterNights. That > has a problem with wrong colors on cloaks an some other parts of some > models. I fink i finally have found the problem. > > In the R300 driver's fragment program part the input COL1 component is > alleyways send to the program when the program wants it. But if wee > have a vertex program that uses the secondary input color for something > and wee don't having GL_SEPARATE_SPECULAR_COLOR. Then default fragment > program has COL1. But it while give us the wrong colors. > > Attached is a patch that tests for GL_SEPARATE_SPECULAR_COLOR. > > /Tommy > ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Mesa3d-dev mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
