Quote:
I've done some homebrew for the DC. I can get the DC to max out at about 4.1 million polygons per frame with ~30% CPU idle time left over. At that point, the PVR can't go any faster.
My test was drawing lit, transformed, textured, anisotropicly filtered tori at 60 FPS. Lighting was a dot, clamp to >0.0, and ambient add. There were 144 tori, each was a single triangle strip containing 480 vertices, or 478 polygons. All of these tori were rotating around in front of the camera. You could probably get a higher polygons-sent-to-TA number if you were sending polygons that are off screen or something, and can be culled by the PVR.
Le Mans definitely does not do 5 million polygons per second. My 4.1 million at 60 FPS uses up almost all of the DC's VRAM for vertex buffers and frame buffers. Le Mans runs at 30 FPS. Doing 4.1 at 30 FPS would require twice as large vertex buffers, since you send twice as many polygons per frame, but that's not possible because there's no VRAM left for larger vertex buffers. 5 million is even less likely to have been done. I know of a few complicated tricks to reduce how large of a vertex buffer is required, but I doubt they've ever been used.