Martin Naylor wrote: > HI, > Just out of curiosity I tried it with my Nvidia 7800 and got these errors? > cannot change Uniform type > cannot change Uniform type > cannot change Uniform type > cannot change Uniform type > cannot change Uniform type
Are you using the latest OSG from CVS? That shouldn't happen. > glLinkProgram "" FAILED > Program "" infolog: > Fragment info > ------------- > <stdlib>(3044) : error C5043: profile requires index expression to be > compile-time constant Ok, that is the kind of implementation variation we're looking for here, thanks. I'm wondering if we should start a wiki page to collect all the ways GPUs might constrain the use of arrays? Cheers -- mew > I have not followed the thread very well, so its probably something you > are aware of. > > Regards > > Martin. > > > > > > ------------------------------------------------------------------------ > > *From:* [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] *On Behalf Of *Edmond Gheury > *Sent:* 21 May 2006 02:38 > *To:* osg users > *Subject:* Re: [osg-users] GLSL uniform arrays v. GPU drivers > > > > Hi, > > I have exactly the same problem. IMO, it's not the uniform arrays > themselves thar are not supported. It's the indirect array indexing > feature, like Color2[i], which is not supported in HW and therefor > falling back to software mode. However, fixed value indexing works on my > X600 (like Color2[1]). I sent a mail to [EMAIL PROTECTED] > <mailto:[EMAIL PROTECTED]> with a test app to learn more about this > limitation, but they replied me that it should work on my GPU, which is > still not the case on mine with the latest drivers. Such a missing > feature is very surprising as it's commonly used when doing vertex > skinning for instance... > > A unpratical solution for small arrays would be to make a conditional > statement to select an array element: > > vec3 color; > if (i == 0) color = mix(Color2[0], Color1, colorSelect) * LightIntensity; > else color = mix(Color2[1], Color1, colorSelect) * LightIntensity; > > Best regards, > > Edmond > > > ------------------------------------------------------------------------ > > _______________________________________________ > 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/
