I’ve got a 500 Mhz machine with an ATI Rage 128 (don’t laugh, it works) and I’m trying my hand at openGL. The alarming thing I found is that rendering roughly 50 texture-mapped triangles takes a whopping 70 ms. If I turn off texture mapping, that goes down to about 10 ms (which I still consider pretty slow).
Granted, I’m using plain triangles instead of strips, no vertex arrays, et cetera. But is this kind of performance to be expected? I’ve seen software renderers on wimpy 75 mhz machines do better.
Hi !
Are you using the same texture for all the tringles, or do you change to different textures ?
Mikael
Originally posted by mikael_aronsson:
[b]Hi !
Are you using the same texture for all the tringles, or do you change to different textures ?
Mikael[/b]
It’s one texture. I’m trying a cheap-and-easy terrain mesh with a single 128x128 texture and a 5x5 height grid. Tex coordinates go from 0.0 to 1.0 in steps of 0.2.
You could have wait for vsyn enabled and an unlucky timing.
You could be using immediate mode (glVertex and stuff)
You could even be in software mode (microsofts software mode is pretty slow).
You could even be measuring incorrectly. It could be much. Maybe you could post some code?