I am intrigued by this whole tessellation stuff, as it's basically procedurally generated polygons, but the gpu must still have to render the polygons so there should still be the same performance hit of adding the polygons manually.
I probably need to read up more on how it works and why it's more beneficial to add detail to a model this way.
My score with CPU & GPU at stock speedsdoes it have scores or fps counter ?
My score with CPU & GPU at stock speeds
Says windows 32bit...But am running 64bit
Unigine
Heaven Demo v1.0
FPS: 45.3
Scores: 1141
Hardware
Binary: Windows 32bit Visual C++ 1500 Release Oct 22 2009
Operating system: Windows 7 (build 7600) 64bit
CPU model: Intel(R) Core(TM)2 Duo CPU E8500 @ 3.16GHz
CPU flags: 3166MHz MMX SSE SSE2 SSE3 SSSE3 SSE41 HTT
GPU model: NVIDIA GeForce 8800 GTX 8.16.11.9107 768Mb
Settings
Render: direct3d11
Mode: 1920x1200 fullscreen
Shaders: high
Textures: high
Filter: trilinear
Anisotropy: 16x
Occlusion: enabled
Refraction: enabled
Volumetric: enabled
Notice my sigive got a similar spec (excpet i have i7) so should be interesting![]()
i'm not that clued up either, but I imagine that there is an LOD benefit. As you approach an object it can smoothly up the pollys in their thousands.I am intrigued by this whole tessellation stuff, as it's basically procedurally generated polygons, but the gpu must still have to render the polygons so there should still be the same performance hit of adding the polygons manually.
I probably need to read up more on how it works and why it's more beneficial to add detail to a model this way.