S.T.A.L.K.E.R.: Clear Sky benchmark

[Mithrandir]

Excellent idea but you need to set some guide lines, like i said in my post, compare eggs with eggs, i never changed any settings on my "rated" run but it seems to have run at completely different settings to everyone else?? Maybe due to the resolution of my monitor or what i don't know.

Rather than take peoples "rated" score (as they all seem to be at different settings) run it at default preset / 1680x1050 / Enhanced full dynamic lighting DX10 / Advanced options all off?

Or whatever seems best?

Oh I see. I thought the rate bench ran at 1280x1024 with enchanced lighting on.

Not sure if its DX10 that runs on selected res.

[mac124]

I think the rated benchmark seems to only run at your windows setting, in my case 1920x1200 and wont change even if you set, for example, 1680x1050 in the settings for the benchmark prog.

I like the idea of a benchmark shoot out, ran a couple my self on other forums, it gets people a tweaking and overclocking for a few more points . Might be worth working out a few "rules" and starting a fresh "Stalker Benchy Shootout" thread if you are up for it? Basically what you have now but specify a detail and resolution, or just a detail level and have seperate charts for different res?

[shadowmaster]

Oh I see. I thought the rate bench ran at 1280x1024 with enchanced lighting on.

Not sure if its DX10 that runs on selected res.

The 'rated' benchmark doesn't allow you to change any settings except the different versions of DX9 or 10.

I propose we either keep the rated benchmark but limit it to 'DX9/10 Advanced' only, or use Plain Benchmark, Default, Enhanced full dynamic lighting, 1680*1050.

I think the rated benchmark seems to only run at your windows setting, in my case 1920x1200 and wont change even if you set, for example, 1680x1050 in the settings for the benchmark prog.

Thats weird. My one is stuck on 1280*1024 even though my monitor is set at 1680*1050.

[WhiteGiant]

Downloading now, will update.

[Mithrandir]

I think the rated benchmark seems to only run at your windows setting, in my case 1920x1200 and wont change even if you set, for example, 1680x1050 in the settings for the benchmark prog.

I like the idea of a benchmark shoot out, ran a couple my self on other forums, it gets people a tweaking and overclocking for a few more points . Might be worth working out a few "rules" and starting a fresh "Stalker Benchy Shootout" thread if you are up for it? Basically what you have now but specify a detail and resolution, or just a detail level and have seperate charts for different res?

Yup like shadowmaster my screen res is 1680x1050 but bench ran at 1280x1024.

[Mithrandir]

The 'rated' benchmark doesn't allow you to change any settings except the different versions of DX9 or 10.

I propose we either keep the rated benchmark but limit it to 'DX9/10 Advanced' only, or use Plain Benchmark, Default, Enhanced full dynamic lighting, 1680*1050.



Thats weird. My one is stuck on 1280*1024 even though my monitor is set at 1680*1050.

I personally prefer the rated one as it gives a score. We should limit to DX9 and DX10 advanced run on the 'rated' option.

[mac124]

Ah i see what it does, each step up increases the resolution, DX10 ultra is 1920x1200, high is 1680x1050, DX10 Advanced is 1280x1024 etc etc

Running again at DX10 Advanced will upload screenies when its done.

Advanced DX10 for gaming pc, system as under "my system"



And for Phenom rig

[shadowmaster]

I personally prefer the rated one as it gives a score. We should limit to DX9 and DX10 advanced run on the 'rated' option.

It looks like we have come to a general consensus , you should put the recommended settings in the opening post so that people know what to run.

[Main]

Well all this has done so far is remind me that I reeeeeeaaaally need an upgrade...cos Webby's DX9 results are in the region of twice as good

[Mithrandir]

It looks like we have come to a general consensus , you should put the recommended settings in the opening post so that people know what to run.

Yup, I'll update the first post

Ah i see what it does, each step up increases the resolution, DX10 ultra is 1920x1200, high is 1680x1050, DX10 Advanced is 1280x1024 etc etc

Running again at DX10 Advanced will upload screenies when its done.

Cheers for the info. Makes sense now.

[WhiteGiant]

Just restarted before running the benchy, not shut anything down.

[Webby]

If possible probably a good idea to have a DX9 and DX10 separate tables as the probably are differences to be seen (don't have Vista so limited to DX9 here). Can people running under Vista run in DX9 mode or not?

[shadowmaster]

If possible probably a good idea to have a DX9 and DX10 separate tables as the probably are differences to be seen (don't have Vista so limited to DX9 here). Can people running under Vista run in DX9 mode or not?

True, maybe if someone runs both modes then we can see if there is significant difference?

[mac124]

Main gaming rig DX10 1280x1024



Main gaming rig DX9 1280x1024



Quite a bit of difference.

[shadowmaster]

^Thanks mac124.

Ok we can either stick to rated benchmark and keep it at DX9 (because not everyone has dx10) or go onto the previous suggestion and use 'Plain Benchmark 1680*1050' and use the average 'day' frame rate as comparison?

Or alternatively we can have two tables, one for dx9 and one for dx10.

[Mithrandir]

^Thanks mac124.

Ok we can either stick to rated benchmark and keep it at DX9 (because not everyone has dx10) or go onto the previous suggestion and use 'Plain Benchmark 1680*1050' and use the average 'day' frame rate as comparison?

Or alternatively we can have two tables, one for dx9 and one for dx10.

I was planning on getting two tables up