DX10 games benchmarked again

Asus

Newbie
Joined
Aug 22, 2003
Messages
10,346
Reaction score
0
Recent benchmark of some DX10 games and DX10 cards.
This review uses Call of Juarez, Company of Heroes and Lost Planet. Would be nice to see some Bioshock numbers though. Any out there?
edit: Here is one only using the HD2900 and 8800GTX OCed (no GTS).

And they didn't get numbers from the HD2900 when using AA in Lost Planet. Driver or game issues? And they didn't benchmark with AA at all in Company of Heroes.

Not surprising, drivers need a lot of work and the HD2900 is more future looking than the 8800 due to the six months extra it had. Both are really poor in DX 10. I really want to know when drivers will mature for these cards and when Vista gets sp 1 then, more games with DX 10 and should give us better performance with the current cards.
 
Here is another quick Bioshock performance look but it is just with Nvidia hardware.

I wouldn't worry about the DX10 performance in Bioshock as much. They used the extra specs and power to add better effects, whether you notice it or not. hehe
The difference in shadows (side by side) is a bit more noticeable than the dynamic water ripples or smoke.

Really need to see a comparison. So far DX10 hasn't added much to games. Like in World of Conflict, DX10 helped with getting better looking smoke and that was about it.
 
There is a comparison on the Firing Squad review with pictures. While it is hard to tell a difference and it may not look 'better', they still used more shader instructions and it will have an impact on performance. So you can't compare DX9->DX10 performance directly in Bioshock since they are not running the same effects.

They said
Epic's own DX10 additions to Unreal Engine 3 are focused on performance optimizations, but Irrational has chosen to go in a different direction for BioShock, using DX10 shaders to enhance graphics. According to the game's manual, the DX10 enhancements include "dynamic water ripples, soft edges for particles [such as fire and smoke - Ed], and crisper shadow edges."

DX9 shadow.
DX10 shadow with a crisper shadow.
 
theres arent really any new special effects in dx10 beside geometry shading,but it does most of what dx9 does now except if you ran the dx10 effects in dx9 then it would suffer performance loss.
 
8800gts 320mb seems like a good choice at least for this game, as long as I don't go to *checks numbers* 2560x1600...even then it did alright in that one graph.
 
My only gripe with the Bioshock is that you cant perform a anti-aliasing pass from within the game
 
Hardocp has a pretty good review over Call of Juarez DX10 vs DX9.
Looks like they put a lot of effort into their DX10 version. Able to squeeze out enough performance in DX10 to do those extra effects. The framerate must have been similar because Hardocp lists the max playable settings as roughly the same for DX9 and DX10 with each card.

Rock texture.

Big hit on the 8800 320mb vs 640mb. Definitely not playable at the same settings in this game.
They list the 8800 320MB with a max-playable setting of 1024x768 @ NoAA/8xAF vs 8800 640MB with 1280x1024 @ NoAA/16AF in DX10.
 
Hardocp has a pretty good review over Call of Juarez DX10 vs DX9.
Looks like they put a lot of effort into their DX10 version. Able to squeeze out enough performance in DX10 to do those extra effects. The framerate must have been similar because Hardocp lists the max playable settings as roughly the same for DX9 and DX10 with each card.

Rock texture.

Big hit on the 8800 320mb vs 640mb. Definitely not playable at the same settings in this game.
They list the 8800 320MB with a max-playable setting of 1024x768 @ NoAA/8xAF vs 8800 640MB with 1280x1024 @ NoAA/16AF in DX10.

Wow, that is quite a difference in textures. Let's see how the the GTX faired
 
Back
Top