Soldato
- Joined
- 12 Jan 2006
- Posts
- 2,547
Now these are benches for my system for a single instance of a single card, they are not to be taken as gospel for performance or for comparative purposes. if anything they are lower performance as my card is heavily bottlenecked.
Testbed:
Intel C2D [email protected]
DFI Ininity 975x Bios 12/01/06
GEIL 4GB PC6400 800MHz DDR2 Dual
HIS Ati 3870 X2 @ Stock - 8.3+Hotfix
WD5000AAKS 1TB Raid 0
OCZ GameXStream 700w Silent
Windows Vista 64 Ultimate SP1
GPU and CPU-Z:

Testing method:
Retail Fraps version 2.9.3 build 6914
running its benchmark for 600 seconds (later reduced to 300 for later games as it was taking too long to do each resolution) while playing the game.
No timedemos were done other than where it is stated.
Ati CCC Settings:
Standard Settings: Performance
Everything set to Application Settings except;
Catalyst AI: Advanced
Minimap Detail Level: Quality
Vertical Refresh: Always Off
Adaptive AA: Disabled
Other considerations:
A 42" Samsung HDTV is also plugged in via HDMI running at 1080p and set to extended mode, i dont know if this made a difference as i didnt test it.
GPU Scaling was disabled via CCC
Call of Duty 4 - Sniper Map
Original version, no patches applied
Max Settings: 1680x1050 - 4x AA, Extra Textures


Frames: 7049
Time (ms): 69063
Min: 46
Max: 123
Avg: 102.066
1680x1050 - No AA

Frames: 51023
Time (ms): 458737
Min: 57
Max: 311
Avg: 111.225
Crysis - dx10, x64
Original, no patches applied
Note: When the checkpoints happened the framerate dipped significantly which may account for the low minimum's, other than that there was no noticable slowdown to indicate the min fps error still occured in this game
1680x1050 - No AA Medium Settings


Frames: 33876
Time (ms): 600000
Min: 25
Max: 103
Avg: 56.46
1680x1050 - 8x AA, Medium Settings
Frames: 8676
Time (ms): 505730
Min: 7
Max: 55
Avg: 17.155
1680x1050 - No AA, High Settings
Frames: 28750
Time (ms): 600000
Min: 24
Max: 88
Avg: 47.917
1280x1024 - No AA, High Settings
Frames: 4433
Time (ms): 65683
Min: 23
Max: 114
Avg: 46.51
1280x1024 - No AA, Very High Settings
Frames: 7237
Time (ms): 600000
Min: 10
Max: 39
Avg: 24.123
The similar min fps on all the resolutions and settings (bar dx10) confirmed my suspicion that the min fps was due to checkpoint saves
Command and Conquer 3 - Tiberium Wars
No patches, 4 way Scrin battle as they have the biggest visual effects
1680x1050 - 3xaa (i think), Max settings

Frames: 8595
Time (ms): 283826
Min: 27
Max: 31
Avg: 30.283
FPS seems capped at 30, but no very low min fps so powerplay is not an issue
Mount and Blade 9.03 + High res texture pack
Tested in a 72vs80 troop skirmish (cavalry, footsoldier mix)
1680x1050.32bit - 4xAA, Ultra High Textures, Texture Details 100 (Shaders and Bump Mapping enabled)



Frames: 35646
Time (ms): 223742
Min: 84
Max: 200
Avg: 159.317
Testbed:
Intel C2D [email protected]
DFI Ininity 975x Bios 12/01/06
GEIL 4GB PC6400 800MHz DDR2 Dual
HIS Ati 3870 X2 @ Stock - 8.3+Hotfix
WD5000AAKS 1TB Raid 0
OCZ GameXStream 700w Silent
Windows Vista 64 Ultimate SP1
GPU and CPU-Z:

Testing method:
Retail Fraps version 2.9.3 build 6914
running its benchmark for 600 seconds (later reduced to 300 for later games as it was taking too long to do each resolution) while playing the game.
No timedemos were done other than where it is stated.
Ati CCC Settings:
Standard Settings: Performance
Everything set to Application Settings except;
Catalyst AI: Advanced
Minimap Detail Level: Quality
Vertical Refresh: Always Off
Adaptive AA: Disabled
Other considerations:
A 42" Samsung HDTV is also plugged in via HDMI running at 1080p and set to extended mode, i dont know if this made a difference as i didnt test it.
GPU Scaling was disabled via CCC
Call of Duty 4 - Sniper Map
Original version, no patches applied
Max Settings: 1680x1050 - 4x AA, Extra Textures


Frames: 7049
Time (ms): 69063
Min: 46
Max: 123
Avg: 102.066
1680x1050 - No AA

Frames: 51023
Time (ms): 458737
Min: 57
Max: 311
Avg: 111.225
Crysis - dx10, x64
Original, no patches applied
Note: When the checkpoints happened the framerate dipped significantly which may account for the low minimum's, other than that there was no noticable slowdown to indicate the min fps error still occured in this game
1680x1050 - No AA Medium Settings


Frames: 33876
Time (ms): 600000
Min: 25
Max: 103
Avg: 56.46
1680x1050 - 8x AA, Medium Settings
Frames: 8676
Time (ms): 505730
Min: 7
Max: 55
Avg: 17.155
1680x1050 - No AA, High Settings
Frames: 28750
Time (ms): 600000
Min: 24
Max: 88
Avg: 47.917
1280x1024 - No AA, High Settings
Frames: 4433
Time (ms): 65683
Min: 23
Max: 114
Avg: 46.51
1280x1024 - No AA, Very High Settings
Frames: 7237
Time (ms): 600000
Min: 10
Max: 39
Avg: 24.123
The similar min fps on all the resolutions and settings (bar dx10) confirmed my suspicion that the min fps was due to checkpoint saves
Command and Conquer 3 - Tiberium Wars
No patches, 4 way Scrin battle as they have the biggest visual effects
1680x1050 - 3xaa (i think), Max settings

Frames: 8595
Time (ms): 283826
Min: 27
Max: 31
Avg: 30.283
FPS seems capped at 30, but no very low min fps so powerplay is not an issue
Mount and Blade 9.03 + High res texture pack
Tested in a 72vs80 troop skirmish (cavalry, footsoldier mix)
1680x1050.32bit - 4xAA, Ultra High Textures, Texture Details 100 (Shaders and Bump Mapping enabled)



Frames: 35646
Time (ms): 223742
Min: 84
Max: 200
Avg: 159.317
Last edited:





