Archived

This topic is now archived and is closed to further replies.

Captian Goatse

VS TexCoord generation performance

Recommended Posts

Captian Goatse    122
What is the texcoord generation performance on older hardware? I''m talking about gf3 and gf4, at least gf4 gives pretty damn good performance, worse than vertex buffer, but a lot better than simple arrays. This is in OpenGL and the memory advatange from the dynamic generation seems really, really wonderful, because I''m paging stuff. Anyone got any prior experiences of this? I know new hardware can hack it, but what about the older one?

Share this post


Link to post
Share on other sites
duke    107
I use texture coordinate generation for my terrain and it seems to have a very minor performance hit. I use it on a GeForce2MX200 and MX400 as well as a GeForce4MX. To be honest, i am not sure if the performance hit is even measurable... but then it could have a lot to do with where the bottle necks in your rendering pipeline are. My GeForce2''s are defintely fill rate limited.

Share this post


Link to post
Share on other sites
WarAmp    750
I use it for raycasting cloud shadows onto my terrain, and for fog (Multipass blending + fog = fugly, TexGen fog + multipass blending = Perfect)

I don''t notice a perf hit at all, but I am bandwidth bound in my app.

gf2mx400 btw.

Share this post


Link to post
Share on other sites