51 lines
2.1 KiB
Text
51 lines
2.1 KiB
Text
--------------------------------------------------------------------------------
|
|
Running performance on file test\p30_clouds.ps
|
|
-------------------- NV40 --------------------
|
|
Target: GeForce 6800 Ultra (NV40) :: Unified Compiler: v65.04
|
|
IPU0 ------ Simplified schedule: --------
|
|
|
|
Pass | Unit | uOp | PC: Op
|
|
-----+--------+------+-------------------------
|
|
1 | SCT0 | div | 2: TEXh h0.xyz, f[TEX1], TEX0;
|
|
| SCT1 | mov | 0: MOVr r1.w, const.---x;
|
|
| TEX | tex | 2: TEXh h0.xyz, f[TEX1], TEX0;
|
|
| | |
|
|
2 | SCT0 | div | 3: TEXh h1.xyz, f[TEX2], TEX1;
|
|
| TEX | tex | 3: TEXh h1.xyz, f[TEX2], TEX1;
|
|
| SCB0 | add | 4: ADDh h1.xyz, h0, h1;
|
|
| | |
|
|
3 | SCT0/1 | div | 5: MOVh h0, f[TEX0];
|
|
| SCB0 | mul | 6: MULh h0.xyz, h1, h0;
|
|
| | |
|
|
4 | SCT0 | mov | 7: TEXh h1.x, r1.wwww, TEX2;
|
|
| TEX | tex | 7: TEXh h1.x, r1.wwww, TEX2;
|
|
| SCB0 | mul | 8: MULh h0.xyz, h0, h1.xxx-;
|
|
|
|
Pass SCT TEX SCB
|
|
1: 75% 100% 0%
|
|
2: 50% 100% 75%
|
|
3: 100% 0% 75%
|
|
4: 0% 100% 75%
|
|
|
|
MEAN: 56% 75% 56%
|
|
|
|
Pass SCT0 SCT1 TEX SCB0 SCB1
|
|
1: 100% 100% 100% 0% 0%
|
|
2: 100% 0% 100% 100% 0%
|
|
3: 100% 100% 0% 100% 0%
|
|
4: 0% 0% 100% 100% 0%
|
|
|
|
MEAN: 75% 50% 75% 75% 0%
|
|
Cycles: 4.00 :: R Regs Used: 2 :: R Regs Max Index (0 based): 1
|
|
--------------------------------------------------------------------------------
|
|
Running performance on file test\p30_clouds.ps
|
|
-------------------- NV40 --------------------
|
|
Target: GeForce 6800 Ultra (NV40) :: Unified Compiler: v81.95
|
|
Cycles: 4.00 :: R Regs Used: 1 :: R Regs Max Index (0 based): 0
|
|
Pixel throughput (assuming 1 cycle texture lookup) 1.60 GP/s
|
|
--------------------------------------------------------------------------------
|
|
Running performance on file test\p30_clouds.ps
|
|
-------------------- G70 --------------------
|
|
Target: GeForce 7800 GT (G70) :: Unified Compiler: v81.95
|
|
Cycles: 4.00 :: R Regs Used: 1 :: R Regs Max Index (0 based): 0
|
|
Pixel throughput (assuming 1 cycle texture lookup) 2.40 GP/s
|