DIRT 2 PC Benchmark results

Windows 7 32bit
Catalyst 9.11(official) with hotfix.
DX9
All max 8xAA
With rapture3D
Result: 12954, av_fps=54.443100 at 1280x1024

 

Not bad at all,

my Q6600 clocked safely and cool at 3Ghz(tape mod) with my 8800GTS512 cards at stock clocks in SLI mode in my 8GB RAM rig with every setting maxed out except resolution and AA which are 1920x1200 and 4xMsAA produce....

samples: 8582
min fps: 29.502872
avg fps: 36.338413

And the game plays very very well at these settings on this hardware!!!

Rapture 3d is installed and I don't know where to see in what DX mode the benchmark ran.

Nforce driver version = 195.62
Vsync = ON

 

Dude, it's a nice score but......

without stating at what resolution and what the AA settings are it's worthless. Come on, share with us

 

From where u guys got the Demo?

 

I found the link in this thread and I found a fast torrent on Google

 

Looking at those specs i think bozotheclown fps a little low...doesn't seem right IMO...

 

Resolution and options are visible in screenshots.

 

Hm maybe, I don't know. I forgot to add my vsync setting to the post. I corrected that now.

I ran the benchmark with Vsync = ON I'll re run the benchmark with Vsync = Off to see what result that produces.

 

I quoted Liniuxs whom hasn't posten any screenies nor details about resolution, AA, Vsync etc.

 

I don't seem to find a place that will let me DL at more than 15 KB/s.

If someone finds a fast link for USA please gimme!

 

win 7 64bit
195.62 whql
high settings, 1280x1024, 4xmsaa, no vsync
rapture 3d i think

<benchmark>
<average min_fps="37.898987" av_fps="54.668789" min_fps_ms="26.385929" av_fps_ms="18.291973" />
<track>
<settings name="morocco_trail" route="route_0">
<car name="ect_00" />
<car name="ect_01" />
<car name="stt_01" />
<car name="crt_03" />
<car name="n1t_01" />
<car name="bmt_01" />
<car name="sot_04" />
<car name="35t_04" />
</settings>
<results samples="12885" min_fps="37.898987" av_fps="54.668789" min_fps_ms="26.385929" av_fps_ms="18.291973" />
</track>
</benchmark>

 

Here is the result with Vsync=Off

my Q6600 clocked safely and cool at 3Ghz(tape mod) with my 8800GTS512 cards at stock clocks in SLI mode in my 8GB RAM rig with every setting maxed out except resolution and AA which are 1920x1200 and 4xMsAA produce....

samples: 8926
min fps: 32.880615
avg fps: 38.314781

And the game plays very very well at these settings on this hardware!!!

Rapture 3d is installed and I don't know where to see in what DX mode the benchmark ran.

Nforce driver version = 195.62
Vsync = Off

I must say, Vsync is not needed as there is no visible tearing.

 

From memory:

Ultra settings, 1920x1200, 4xAA gives me an average of 65 fps with a minimum of 50 fps. The same settings and resolution with 8xQCSSAA (or whatever it's called) gives me an average of 49 fps with a minimum of 40 fps. It's silky smooth with no judder whatsoever even when the screen is packed with cars so it's evidently a well optimised engine for the PC (just like GRID's was). I'm using the 195.62 WHQL drivers.

Nice performance for DX9 and the game looks very good but, IMO, not that much better than the 360 version I played in the summer bar the higher framerate and resolution/AA. I noticed that grass and bushes as well as reflections in water tend to pop into view suddently due to an erratic draw distance but that's about the only gripe I have with the visuals.

 

And judging by WhiteLightning's post, is this another game that is limiting effects for lesser DirectX users? I'm referring to the Post Process setting which doesn't go higher than Medium with DX9?

Seeing as there doesn't seem to be all that many visual differences between the DX9 and DX11 versions that I've seen from footage and screenshots so far, it makes me wonder what the DX10 version would have looked like had Codemasters bothered to code it. I mean I'm guessing it would even closer to DX11 quality perhaps without the severe performance loss. From the PCGameHardware benchmarks I saw HD5870 owners lose 50% performance going from DX9 to DX11. Crowds still look like they're from a last-generation game too despite tessellation.

Seems to me that's it's the whole DX10 thing all over again where games looks marginally better at the cost of a lot of performance. Of course, it's a choice that's nice to have but it's not one that makes me want to dash out and buy a DX11 card.

 

Here

http://hotfile.com/dl/18934388/8f50dec/Dirt2.part1.rar.html
http://hotfile.com/dl/18934522/d24f514/Dirt2.part2.rar.html
http://hotfile.com/dl/18934373/9c5bc1b/Dirt2.part3.rar.html
http://hotfile.com/dl/18934505/f150153/Dirt2.part4.rar.html
http://hotfile.com/dl/18934378/48fffce/Dirt2.part5.rar.html
http://hotfile.com/dl/18934525/8e744a4/Dirt2.part6.rar.html
http://hotfile.com/dl/18934627/b194b99/Dirt2.part7.rar.html

 

And that is with sli enabled:3eyes:, could you confirm or anyone that sli works well...


Edit: By the way: SLI is still not working, even with the Geforce 195.62 WHQL, although the new driver fixes the game crashes originating with the 195.55.
Source : http://www.pcgameshardware.com/aid,...DX-11-Update-Radeon-HD-5970-results/Practice/

 

You generally don't see much tearing in racing games where 90% of the game has you hurtling forward at high speeds. However, you can see tearing very clearly when cornering (moving laterally).

Tearing isn't the only reason to enable v-sync anyway as capping the framerate to your display's refresh rate means less work for your graphics card, which equals less heat, and also means that inevitable drops from 60 fps to 50 fps (16.67% drop), say, are going to be less obvious than drops from 100 fps to 50 fps (50% drop).

It's obviously your choice but there's no real need for a racing game to run above 60 fps IMO; it certainly doesn't make the game feel any faster nor does it make much difference to the responsiveness of the controls (not like 30 fps to 60 fps does anyway).

 

Muchas gracias