Post your Company of Heroes DX10 Benches

bolt14

Distinguished
Feb 19, 2007
51
0
18,630
:D COH has gone DX10!
So, how goes it for you all? Post the average fps and the min/max from the built in performance test and your settings.

Shaders: DX10
Everything on highest except trees which were on medium.
AA set at x4

AMD X2 5600+ at 3.08Ghz
2x 8800GTS 640Mb (Stock Speeds)
3GB Ram

Max - 65
Min - 1.9 :(
Avg - 41.3
 

Xazax310

Distinguished
Aug 14, 2006
857
0
18,980
For some reason with 1.7 FRAPS doesn't work any more =( but here are my scores.

Rig
E6600@ 3.44Ghz
2GB 800mhz Ram 5-5-5-12(1T)
8800GTX 620/1000mhz

Thats with Everything on High/Ultra with 2x AA

average = 44.4 FPS
maximum = 62.2 FPS
minimum = 9.9 FPS
 

locky28

Distinguished
Sep 19, 2006
478
0
18,780
I think what everyone's really after is comparisons between DX10 and DX9 regarding both your image quality and your Frames ;)
 

bolt14

Distinguished
Feb 19, 2007
51
0
18,630
Oops! Just realized, my score was with one 8800! I forgot to enable SLI!
SLI Scores: (The driver is a beta and is unstable, the game has lots of crashing issues w/ DX10 SLI)

Min - 12.1
Max - 73.2
Avg. - 50.7
 

bullaRh

Distinguished
Oct 6, 2006
592
0
18,980
can you tell me what fps u had with dx9? much better when you can compare em :d..
can u tell the difference on dx9 and dx10? maybe show some screens?
 

james_8970

Distinguished
Feb 27, 2007
169
0
18,680
Sorry, is the game itself crashing because of the beta drivers of DX10 or the beta drivers for SLI?
When did the DX10 patch come out, it's great timing building in one week now!
I want to see someone with a HD2900XT post benchies.
god are the mins ever low!
James
 

bolt14

Distinguished
Feb 19, 2007
51
0
18,630
DX9 was about 7 frames faster all around ON BOTH SINGLE CARD AND SLI MODES. Depending on the battle, I can tell the difference more or less. Rule of thumb - the more large explosions and tanks fighting the better DX10 looks. The German V-1 detonation is spectacular!

And it is especially satisfying when there are a couple enemy units caught in it. :wink:

EDIT: Weird problem with screenshots. Every shot in the folder is corrupt and Windows gives a BSOD when I try to open them. :? Will work on tonight and in the morning.
 

bolt14

Distinguished
Feb 19, 2007
51
0
18,630
Beta for SLI. With just one card, the game runs flawlessly. the single card drivers have been out for a while and seem stable. The NVIDIA 158.42 Beta DX10 SLI drivers are average at best. Worthless most of the time though. :wink:

The mins always are that low (atleast on my comp.) when ultra textures are enabled. I toned those down to high and the min shot up to 29.4fps.
The patch was released yesterday. From what I know from a friend the 2900XT is actually on a par w/ the 8800GTS in this case. With crossfire though, the game won't start. Figures. :roll:
 

bolt14

Distinguished
Feb 19, 2007
51
0
18,630
ACK! With the new drivers, the game wont even start in SLI!!!
Single card FPS performance is much better:
Max - 81.3
Min - 30.0
Avg. - 63.2
 

fishboi

Distinguished
Apr 25, 2006
1,415
0
19,280
Yeah - 1st thing I wanted when I saw this post was DX9 vs DX10 FPS comparisons, as well as screen shots. Plz post that.

Dont forget to those who try this to check out the FPS before you install the patch.
 

coolpurplefan

Distinguished
Feb 19, 2006
49
0
18,530
Uh, I don't know about you guys but it seems to me the water looks better and the performance is better in DX9.

Gee, this all seems kind of weird. Perhaps you people are right though, might be driver issues.

It kind of makes me wonder if people will prefer to play Unreal Tournament 3 in DX9 or DX10...?
 

dean7

Distinguished
Aug 15, 2006
1,559
0
19,780
Yeah... I kind of thought the same thing... and a 50% performance hit sucks.

But, like the article said this is their first stab at DX10, and DX9 is approaching the end of it's life cycle.
 

james_8970

Distinguished
Feb 27, 2007
169
0
18,680
agreed, but DX10 isn't making the hardware 100% more efficient. With better graphics we are to expect a hit in performance. Though it'll probably get better over time with drivers, I don't expect DX10 to have better performance then DX9.
James
 

fishboi

Distinguished
Apr 25, 2006
1,415
0
19,280
I wanna ask you guys something - whats the real benefit of DX10 over DX9 ???? What I understand, is that its more pixels per calculation because of the way the programming is handled (ie. more efficient so you can get better graphics for the hardware). With Core 2, it seems like our CPUs can do anything these days.

If a game is programmed to do a lot of calculations, and you have the chip to do it, then it should look nearly as good. I've seen the Crysis demos, and the DX9 looks nearly as good as the DX10 versions. Yes there is a diffs with repect to some lighting and cooler explosions, but if you look at DX9, the way objects move is similar because they are programmed to do so. The chips can handle it. Those physics visualizations are cool not because of DX10, but because the game is programmed to do them. My chip can do them (and DX10 cant hurt to help it along), but its damn close enough.

I think CPU capacity over ran the need for DX10, and thats why we havent seen mass adoption. When games/visualizations become so intense that our chips cant handle it and the calcs need to be more efficient, then DX10 will really come into play.

We havent even hit capacity limitations with Conroe. What about the new Quads dropping to $266 in July - everyone will get one. I think DX10 has a much longer time frame to go before we really need it. My $0.02.
 

Artmic

Distinguished
May 27, 2002
311
0
18,780
Uh, I don't know about you guys but it seems to me the water looks better and the performance is better in DX9.

Gee, this all seems kind of weird. Perhaps you people are right though, might be driver issues.

It kind of makes me wonder if people will prefer to play Unreal Tournament 3 in DX9 or DX10...?

I can see driver issues, but how much are the drivers going to improve? 200 or 300 percent?

I"m really starting to question DX10 performance, i figure this POS would be the same if it was displaying the same type of special effects in the game as the DX9 version.

The only way to know for sure is to have 2 versions of a game with the same features on the same hardware but one with DX9 and the other with DX10,

Then we'd know for sure how much of a drop-off is the result of the drivers and vista slowing down the game. IMO.