• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

CRYSIS vs. The Modern GPU (Part 1)

Big.Wayne

Big.Wayne

Big.Wayne

Soldato
Joined
11 Sep 2003
Posts
14,716
Location
London
crysisimages13editxe8.jpg


Well the cold winter evening are here, the TV and newspaper advertisements are reminding us that Christmas is around the corner and its time to start making out our xmas card list and shopping for presents . . . . but hold on a sec . . . . . CRYSIS is going to be released any day now? . . . . . wOOt!!!!!!!!!

Right well the xmas plans can just go on hold for the moment as any self respecting gamer knows, Christmas eve is meant for all of that stuff, right now is time to game! ;)

The Crysis demo has been out for a few weeks now and 8800GTX owners world wide have been gasping in disbelief as their frame-rates have *crumbled*

I'm sure there were one or two people who maxed their graphic settings before loading the game for the first time, only to be greeted by a high-def slide show!

So in this review I want to find out how this game plays on a fairly modern system, using a few different resolutions and different amounts of Anti-Aliasing (AA).

This is the first time I have tried to publish a set of FPS results in graph form and while it is easier than i thought it would be, its not as easy as I would like lol! :D

Therefore to keep things simple (i.e. avoiding data overload) I have kept the in-game graphic settings at either *Medium* or *High* and nothing has been changed in the graphics cards drivers. The only exceptions to this rule is Audio setting which is on *high* (X-Fi baby!) and the Shader setting which turns itself to *high* once any Anti-Aliasing is used.

In future instalments I will look at the effect of overclocking the graphics card, also comparing newer more refined drivers and maybe even SLI.

Now a quick peek at the test bed specs then onto some bench results . . .

spec1ff7.jpg


spec2jc1.jpg
 
Just wanted to add a little about the test results. The in-game benchmark was left to run its course (four loops), I then discarded the results from the first loop and averaged the remaining three.

resultsdr1.jpg



Robbie G said:
Something wrong with the 2AA at the lower resolutions though...it has a lower min FPS than 4AA??!
Click to expand...
yup thats the sorta thing that happens when you bulk bench, sometimes the results don't add up. In some cases its worth running the test again to see if its a hiccup or perhaps a bug in the driver/game, time permitting of course!

Heres the original log from those particular tests. . . .

results2kp8.jpg
 
Last edited:
Duke said:
Great stuff Big.Wayne. It shows some people talk out their **** :)
Click to expand...
No I wouldn't go that far, it just shows that what people consider silky smooth can vary a great deal. This isn't an attempt to shame any of my fellow forum members but instead to take away any opinions at all and just look at bare fps facts. Personally I aim for 1920x1200with a minimum of 30fps and as much goodies as I can get away with, others may prefere an average of 30fps.

I also don't mind switching down resolutions to get the right balance!

marscay said:
good stuff wayne - i think that's an indication that SLI'd GT's are the way to go for 1920x1200 action unless you want to play on medium.
Click to expand...
Thanks marscay, for the way I like a game to play I tend to agree with you, from some early results I have seen this game does scale well using SLI, now the only problem I have is not owning a SLI compatible motherboard! :o

carlazai said:
Only criticism is that all the graphs should be the same scale. Otherwise great demonstration from a real life user
Click to expand...
Thank you again and criticism welcomed and duly noted! :)
 
Last edited:
silversurfer said:
How did you collect the fps data, fraps?
Click to expand...
No just used the games in-built benchmark. The results are dumped to a text file.

However I just found this cool new app today, wish It had been out a few weeks ago!

Crysis Benchmark Tool 1.05 Final

Click image to download

Also allowed me to pick the widescreen resolutions that were missing in-game.
 
Last edited:
Nice to see people taking the time to share their results, just one thing though it would make for easier comparison in this thread if you bench at either *All Medium* or *All High* :cool:

[edit] lol silversurfer beat me to it by 17 seconds!

Also if you have an X-Fi just leave audio settings at high throughout.
 
Last edited:
Sir Random said:
Looks like XP is the OS for Crysis
Click to expand...
That's what I thought when I first looked at the results, but look again and u can see its actually DX10 and not Vista. DX9 codepath run under Vista produces very similar results to DX9 on Windows XP.

There is extra DX10 eye candy (so I read and am told) being rendered there, that seems to be the reason for the lower FPS.
 
Carter said:
28/11/2007 10:09:38 - XP
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 800x600, AA=No AA, Vsync=Disabled, 32 bit test, FullScreen
Demo Loops=3, Time Of Day= 9
Global Game Quality:
Custom Quality Values:
VolumetricEffects=
Texture=
ObjectDetail=
Sound=
Shadows=
Water=
Physics=
Particles=
Shading=
PostProcessing=
GameEffects=
Click to expand...
Hi Carter,

am I being thick or is the quality settings missing there?
 
You must log in or register to reply here.
Back
Top Bottom