8800GTX Shadermark results

[Geeforcer]

Looks like we have our first sampling of shadermark on G80:

video mode / device info
(1600x1200) X8R8G8B8 (D24X8) vram used 168820736
HAL (pure hw vp): NVIDIA GeForce 8800 GTX (Anti-Detect-Mode: off, gamma correction: DAC)



options

pixel shader version: Pixel Shader 3_0

results:
shader 1 ( Pixel Shader Precision Test): pixel shader 1.1 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 1.4 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 2.0 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 2.0 partial precision: s23e8
( Pixel Shader Precision Test): pixel shader 3.0 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 3.0 partial precision: s23e8

shader 2 ( Per Pixel Diffuse Lighting - ps_3_0): 1860 fps 0.5376 mspf 9304 rendered frames
shader 3 ( Per Pixel Directional Light Shader (Phong) - ps_3_0): 1426 fps 0.7011 mspf 7134 rendered frames
shader 4 ( Per Pixel Point Light Shader (Phong) - ps_3_0): 1375 fps 0.7272 mspf 6878 rendered frames
shader 5 ( Per Pixel Spot Light Shader (Phong) - ps_3_0): 1178 fps 0.8487 mspf 5894 rendered frames
shader 6 ( Per Pixel Anisotropic Lighting - ps_3_0): 1455 fps 0.6875 mspf 7276 rendered frames
shader 7 ( Per Pixel Fresnel Reflections - ps_3_0): 1303 fps 0.7672 mspf 6520 rendered frames
shader 8 ( Per Pixel Car Surface Shader - ps_3_0): 967 fps 1.0339 mspf 4838 rendered frames
shader 9 ( Per Pixel Environment Mapping - ps_3_0): 2188 fps 0.4571 mspf 10942 rendered frames
shader 10 ( Per Pixel Environment Bump Mapping - ps_3_0): 1984 fps 0.5041 mspf 9922 rendered frames
shader 11 ( Per Pixel Bump Mapping - ps_3_0): 1335 fps 0.7491 mspf 6677 rendered frames
shader 12 ( Per Pixel Shadowed Bump Mapping - ps_3_0): 808 fps 1.2372 mspf 4043 rendered frames
shader 13 ( Per Pixel Veined Marble Shader - ps_3_0): 929 fps 1.0762 mspf 4648 rendered frames
shader 14 ( Per Pixel Wood Shader - ps_3_0): 1152 fps 0.8679 mspf 5763 rendered frames
shader 15 ( Per Pixel Tile Shader - ps_3_0): 720 fps 1.3883 mspf 3603 rendered frames
shader 16 ( Per Pixel Refraction and Reflection Shader with Phong Lighting - ps_3_0): 630 fps 1.5884 mspf 3149 rendered frames
shader 17 ( Per Pixel BRDF-Phong/Anisotropic Lighting - ps_3_0): 995 fps 1.0054 mspf 4975 rendered frames

performance impact of heavy alpha blending

shader 18 ( Fur Shader With Anisotropic Lighting - ps_3_0): 101 fps 9.8773 mspf 507 rendered frames

performance impact of multiple shaders switches

shader 19 ( Combination Effect - ps_3_0): 446 fps 2.2400 mspf 2233 rendered frames

performance impact of dynamic flow control

shader 20 ( Dual Layer 8x8 PCF Shadow Mapping without Flow Control - ps_3_0): 168 fps 5.9452 mspf 842 rendered frames
shader 21 ( Dual Layer 8x8 PCF Shadow Mapping with Flow Control - ps_3_0): 201 fps 4.9859 mspf 1004 rendered frames

performance impact of floating point filtering

shader 22 ( High Dynamic Range Shader - low quality version without filtering - ps_3_0): 192 fps 5.2104 mspf 960 rendered frames
shader 23 ( High Dynamic Range Shader - high quality with fp filtering - ps_3_0): 192 fps 5.2124 mspf 960 rendered frames
shader 24 ( High Dynamic Range Shader - high quality without fp filtering - ps_3_0): 192 fps 5.2104 mspf 960 rendered frames

performance impact of multiple render targets

shader 25 ( Per Pixel Edge Detection And Hatching Shader using 1 RT and 2 Passes - ps_3_0): 255 fps 3.9185 mspf 1277 rendered frames
shader 26 ( Per Pixel Edge Detection And Hatching Shader using 2 RTs and 1 Pass - ps_3_0): 257 fps 3.8911 mspf 1286 rendered frames

 

[Davros]

how does that compare to other cards ?

 

[Geeforcer]

how does that compare to other cards ?

IDK how Shadermark scales with resolution, if at all - and 1600x1200 scores are not around. I guess if anyone wants to run it at 1600x1200 on their card, we will find out.

 

[SugarCoat]

X1800XT @ 1280x1024 just for giggles. Careful though, shader tests dont match up completely, different versions.

video mode / device info
(1280x1024) X8R8G8B8 (D24X8) vram used 147849216
HAL (pure hw vp): Radeon X1800 Series (gamma correction: DAC)

shader 1 (Per Pixel Diffuse Lighting - ps_3_0): 896 fps 1.1156 mspf 4496 rendered frames
shader 2 (Per Pixel Directional Light Shader (Phong) - ps_3_0): 595 fps 1.6797 mspf 2986 rendered frames
shader 3 (Per Pixel Point Light Shader (Phong) - ps_3_0): 649 fps 1.5409 mspf 3255 rendered frames
shader 4 (Per Pixel Spot Light Shader (Phong) - ps_3_0): 493 fps 2.0298 mspf 2471 rendered frames
shader 5 (Per Pixel Anisotropic Lighting - ps_3_0): 601 fps 1.6630 mspf 3016 rendered frames
shader 6 (Per Pixel Fresnel Reflections - ps_3_0): 581 fps 1.7224 mspf 2912 rendered frames
shader 7 (Per Pixel Car Surface Shader - ps_3_0): 429 fps 2.3318 mspf 2151 rendered frames
shader 8 (Per Pixel Environment Mapping - ps_3_0): 1164 fps 0.8591 mspf 5838 rendered frames
shader 9 (Per Pixel Environment Bump Mapping - ps_3_0): 941 fps 1.0624 mspf 4721 rendered frames
shader 10 (Per Pixel Bump Mapping - ps_3_0): 602 fps 1.6619 mspf 3018 rendered frames
shader 11 (Per Pixel Shadowed Bump Mapping - ps_3_0): 433 fps 2.3092 mspf 2172 rendered frames
shader 12 (Per Pixel Veined Marble Shader - ps_3_0): 424 fps 2.3603 mspf 2125 rendered frames
shader 13 (Per Pixel Wood Shader - ps_3_0): 543 fps 1.8433 mspf 2721 rendered frames
shader 14 (Per Pixel Tile Shader - ps_3_0): 373 fps 2.6793 mspf 1872 rendered frames
shader 15 (Per Pixel Refraction and Reflection Shader with Phong Lighting - ps_3_0): 379 fps 2.6356 mspf 1903 rendered frames
shader 16 (Per Pixel BRDF-Phong/Anisotropic Lighting - ps_3_0): 437 fps 2.2882 mspf 2192 rendered frames

performance impact of heavy alpha blending

shader 17 (Fur Shader With Anisotropic Lighting - ps_3_0): 223 fps 4.4903 mspf 1117 rendered frames

performance impact of multiple shaders switches

shader 18 (Combination Effect - ps_3_0): 160 fps 6.2617 mspf 801 rendered frames

performance impact of dynamic flow control

shader 19 (Dual Layer 8x8 PCF Shadow Mapping without Flow Control - ps_3_0): 86 fps 11.6642 mspf 430 rendered frames
shader 20 (Dual Layer 8x8 PCF Shadow Mapping with Flow Control - ps_3_0): 114 fps 8.7533 mspf 573 rendered frames

performance impact of floating point filtering

shader 21 (High Dynamic Range Shader - low quality version without filtering - ps_3_0): 204 fps 4.8981 mspf 1024 rendered frames
shader 22 (High Dynamic Range Shader - high quality with fp filtering - ps_3_0): 227 fps 4.4035 mspf 1139 rendered frames
shader 23 (High Dynamic Range Shader - high quality without fp filtering - ps_3_0): 158 fps 6.3329 mspf 792 rendered frames

performance impact of multiple render targets

shader 24 (Per Pixel Edge Detection And Hatching Shader using 1 RT and 2 Passes - ps_3_0): 129 fps 7.7641 mspf 646 rendered frames
shader 25 (Per Pixel Edge Detection And Hatching Shader using 2 RTs and 1 Pass - ps_3_0): 135 fps 7.4306 mspf 675 rendered frames

 

[Geeforcer]

Sugarcoat, are did you just run those? If so, could you try a different resolution (preferebly 1600x1200, but anything will do) to see what, if any, difference does it make?

 

[SugarCoat]

would have if i could, LCD an all though.

Re-run on 1024x768 does give a pretty substantial increase in all scores though, in many cases close to 100%, so resolution certainly plays a part.

 

[BlizzardOne]

I hope this is of use/interest

Single X1950XT-X

video mode / device info
(1600x1200) X8R8G8B8 (D24X8) vram used 169869312
HAL (pure hw vp): Radeon X1950 CrossFire Edition (Anti-Detect-Mode: off, gamma correction: DAC)



options

pixel shader version: Pixel Shader 3_0

results:
shader 1 ( Pixel Shader Precision Test): pixel shader 1.1 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 1.4 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 2.0 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 2.0 partial precision: s23e8
( Pixel Shader Precision Test): pixel shader 3.0 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 3.0 partial precision: s23e8

shader 2 ( Per Pixel Diffuse Lighting - ps_3_0): 1067 fps 0.9376 mspf 5366 rendered frames
shader 3 ( Per Pixel Directional Light Shader (Phong) - ps_3_0): 846 fps 1.1824 mspf 4255 rendered frames
shader 4 ( Per Pixel Point Light Shader (Phong) - ps_3_0): 862 fps 1.1598 mspf 4338 rendered frames
shader 5 ( Per Pixel Spot Light Shader (Phong) - ps_3_0): 721 fps 1.3864 mspf 3629 rendered frames
shader 6 ( Per Pixel Anisotropic Lighting - ps_3_0): 846 fps 1.1827 mspf 4254 rendered frames
shader 7 ( Per Pixel Fresnel Reflections - ps_3_0): 806 fps 1.2414 mspf 4053 rendered frames
shader 8 ( Per Pixel Car Surface Shader - ps_3_0): 526 fps 1.9029 mspf 2644 rendered frames
shader 9 ( Per Pixel Environment Mapping - ps_3_0): 1046 fps 0.9556 mspf 5265 rendered frames
shader 10 ( Per Pixel Environment Bump Mapping - ps_3_0): 868 fps 1.1518 mspf 4368 rendered frames
shader 11 ( Per Pixel Bump Mapping - ps_3_0): 810 fps 1.2347 mspf 4075 rendered frames
shader 12 ( Per Pixel Shadowed Bump Mapping - ps_3_0): 456 fps 2.1932 mspf 2294 rendered frames
shader 13 ( Per Pixel Veined Marble Shader - ps_3_0): 354 fps 2.8234 mspf 1782 rendered frames
shader 14 ( Per Pixel Wood Shader - ps_3_0): 536 fps 1.8641 mspf 2699 rendered frames
shader 15 ( Per Pixel Tile Shader - ps_3_0): 557 fps 1.7956 mspf 2802 rendered frames
shader 16 ( Per Pixel Refraction and Reflection Shader with Phong Lighting - ps_3_0): 514 fps 1.9463 mspf 2585 rendered frames
shader 17 ( Per Pixel BRDF-Phong/Anisotropic Lighting - ps_3_0): 635 fps 1.5747 mspf 3195 rendered frames

performance impact of heavy alpha blending

shader 18 ( Fur Shader With Anisotropic Lighting - ps_3_0): 60 fps 16.6048 mspf 303 rendered frames

performance impact of multiple shaders switches

shader 19 ( Combination Effect - ps_3_0): 260 fps 3.8524 mspf 1306 rendered frames

performance impact of dynamic flow control

shader 20 ( Dual Layer 8x8 PCF Shadow Mapping without Flow Control - ps_3_0): 95 fps 10.5037 mspf 479 rendered frames
shader 21 ( Dual Layer 8x8 PCF Shadow Mapping with Flow Control - ps_3_0): 105 fps 9.5651 mspf 526 rendered frames

performance impact of floating point filtering

shader 22 ( High Dynamic Range Shader - low quality version without filtering - ps_3_0): 165 fps 6.0472 mspf 832 rendered frames
shader 23 ( High Dynamic Range Shader - high quality with fp filtering - ps_3_0): 190 fps 5.2738 mspf 954 rendered frames
shader 24 ( High Dynamic Range Shader - high quality without fp filtering - ps_3_0): 120 fps 8.3161 mspf 605 rendered frames

performance impact of multiple render targets

shader 25 ( Per Pixel Edge Detection And Hatching Shader using 1 RT and 2 Passes - ps_3_0): 106 fps 9.4395 mspf 533 rendered frames
shader 26 ( Per Pixel Edge Detection And Hatching Shader using 2 RTs and 1 Pass - ps_3_0): 106 fps 9.4750 mspf 531 rendered frames



X1950XT-X in Crossfire:

video mode / device info
(1600x1200) X8R8G8B8 (D24X8) vram used 164626432
HAL (pure hw vp): Radeon X1950 CrossFire Edition (Anti-Detect-Mode: off, gamma correction: DAC)



options

pixel shader version: Pixel Shader 3_0

results:
shader 1 ( Pixel Shader Precision Test): pixel shader 1.1 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 1.4 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 2.0 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 2.0 partial precision: s23e8
( Pixel Shader Precision Test): pixel shader 3.0 full precision: s23e8
( Pixel Shader Precision Test): pixel shader 3.0 partial precision: s23e8

shader 2 ( Per Pixel Diffuse Lighting - ps_3_0): 1783 fps 0.5609 mspf 8970 rendered frames
shader 3 ( Per Pixel Directional Light Shader (Phong) - ps_3_0): 1465 fps 0.6826 mspf 7371 rendered frames
shader 4 ( Per Pixel Point Light Shader (Phong) - ps_3_0): 1485 fps 0.6734 mspf 7471 rendered frames
shader 5 ( Per Pixel Spot Light Shader (Phong) - ps_3_0): 1269 fps 0.7882 mspf 6383 rendered frames
shader 6 ( Per Pixel Anisotropic Lighting - ps_3_0): 1465 fps 0.6828 mspf 7369 rendered frames
shader 7 ( Per Pixel Fresnel Reflections - ps_3_0): 1402 fps 0.7133 mspf 7053 rendered frames
shader 8 ( Per Pixel Car Surface Shader - ps_3_0): 994 fps 1.0065 mspf 4999 rendered frames
shader 9 ( Per Pixel Environment Mapping - ps_3_0): 1787 fps 0.5597 mspf 8989 rendered frames
shader 10 ( Per Pixel Environment Bump Mapping - ps_3_0): 1648 fps 0.6068 mspf 8292 rendered frames
shader 11 ( Per Pixel Bump Mapping - ps_3_0): 1430 fps 0.6992 mspf 7196 rendered frames
shader 12 ( Per Pixel Shadowed Bump Mapping - ps_3_0): 902 fps 1.1087 mspf 4538 rendered frames
shader 13 ( Per Pixel Veined Marble Shader - ps_3_0): 688 fps 1.4537 mspf 3461 rendered frames
shader 14 ( Per Pixel Wood Shader - ps_3_0): 966 fps 1.0352 mspf 4860 rendered frames
shader 15 ( Per Pixel Tile Shader - ps_3_0): 781 fps 1.2812 mspf 3927 rendered frames
shader 16 ( Per Pixel Refraction and Reflection Shader with Phong Lighting - ps_3_0): 997 fps 1.0028 mspf 5017 rendered frames
shader 17 ( Per Pixel BRDF-Phong/Anisotropic Lighting - ps_3_0): 1145 fps 0.8736 mspf 5759 rendered frames

performance impact of heavy alpha blending

shader 18 ( Fur Shader With Anisotropic Lighting - ps_3_0): 111 fps 9.0166 mspf 558 rendered frames

performance impact of multiple shaders switches

shader 19 ( Combination Effect - ps_3_0): 402 fps 2.4846 mspf 2025 rendered frames

performance impact of dynamic flow control

shader 20 ( Dual Layer 8x8 PCF Shadow Mapping without Flow Control - ps_3_0): 83 fps 12.0653 mspf 417 rendered frames
shader 21 ( Dual Layer 8x8 PCF Shadow Mapping with Flow Control - ps_3_0): 90 fps 11.1311 mspf 452 rendered frames

performance impact of floating point filtering

shader 22 ( High Dynamic Range Shader - low quality version without filtering - ps_3_0): 103 fps 9.6755 mspf 520 rendered frames
shader 23 ( High Dynamic Range Shader - high quality with fp filtering - ps_3_0): 112 fps 8.9049 mspf 565 rendered frames
shader 24 ( High Dynamic Range Shader - high quality without fp filtering - ps_3_0): 84 fps 11.9507 mspf 421 rendered frames

performance impact of multiple render targets

shader 25 ( Per Pixel Edge Detection And Hatching Shader using 1 RT and 2 Passes - ps_3_0): 119 fps 8.4276 mspf 597 rendered frames
shader 26 ( Per Pixel Edge Detection And Hatching Shader using 2 RTs and 1 Pass - ps_3_0): 119 fps 8.4276 mspf 597 rendered frames

 

[BlizzardOne]

Put X1950, X1950CF and 8800GTX results into a pretty graph for those who like graphs and stuff..



Very odd that some of the CF scores are lower than the single card scores though..

 

[Skinner]

Put X1950, X1950CF and 8800GTX results into a pretty graph for those who like graphs and stuff..



Very odd that some of the CF scores are lower than the single card scores though..

Thanx, I got dizzy from comparing the numbers

 

[Razor1]

Put X1950, X1950CF and 8800GTX results into a pretty graph for those who like graphs and stuff..



Very odd that some of the CF scores are lower than the single card scores though..

depending on the type of shadow maps used, could be a small burden on the CPU don't know if it would bottleneck CF though. Wierd HDR results for CF, on the otherhand, the gf8 seems to not have any performance hit or gain on some of the similiar shaders special HDR ones.

 

[BlizzardOne]

depending on the type of shadow maps used, could be a small burden on the CPU don't know if it would bottleneck CF though. Wierd HDR results for CF, on the otherhand, the gf8 seems to not have any performance hit or gain on some of the similiar shaders special HDR ones.

Well, I just re-ran for both single and CF setups, this time with Cat A.I. set to Advanced, and pretty substantial improvements, especially in the later tests:

 

[SugarCoat]

shader replacements are a wonderful thing.

 

[BlizzardOne]

Just looking over that second set of numbers again, and it's interesting/uncanny how many of the Single to CF scores are exactly double..

 

[woundingchaney]

Amazing performance out of the single 8800gtx, I was wanting to wait to see ATIs answer but it may be near impossible to do this.

 

[Ailuros]

shader replacements are a wonderful thing.

For all three compared cases or just one?

 

[ChrisRay]

shader replacements are a wonderful thing.

I am willing to bet it has more to do with AFR being used rather than tiling in the case of crossfire.

 

[Geeforcer]

Something seems to be off with the shaders 22-24 (the 192 fps ones) - it's clearly hitting some sort of a wall.

 

[Ailuros]

Something seems to be off with the shaders 22-24 (the 192 fps ones) - it's clearly hitting some sort of a wall.

They obviously didn't have enough time to replace all shaders

 

[Ftn]

They obviously didn't have enough time to replace all shaders

Hehe. Seems like those tests are somehow limited at 192 fps, and 192*2 on sli/crossfire.

 

[SugarCoat]

I am willing to bet it has more to do with AFR being used rather than tiling in the case of crossfire.

Doesnt make sense you'd have to adjust Cat AI for that to work at optimal, unless it wasnt using AFR before and that forced it on. Seems to be acting like the Doom 3 Engine shader replacement/memory optimization to me. Even a couple of the single card tests got a noticable kick.

Amazing performance out of the single 8800gtx, I was wanting to wait to see ATIs answer but it may be near impossible to do this.

i'm of the opinion now where i think i can hold out until Vista launches. I'll try to anyway, before i go upgrading my graphics card. To be honest there arent any games coming out that i would need such a power house yet, resolution limit and all, plus theres always the fear of getting obsoleted in couple months .

I'm far more interested in the technical break down of the new architecture to give me an idea of what to expect in the future and checking out some new tech vids then i am on spending hundreds for the actual card cause theres no graphics goal yet for me to meet. Now when crysis launches, it will be a different story.