NEW - 3DMark 11 Compilation

[Bjorn_Of_Iceland]

Originally Posted by ShiBDiB

i crash during the physics test pretty sure its cause i have the physx drivers installed with no compatible card.... hmm guess it gives me an excuse to use my old 8600gt as a physics card.

Try unchecking the scan systeminfo check box in the help tab. Worked for me.

[fullinfusion]

Originally Posted by ShiBDiB

i crash during the physics test pretty sure its cause i have the physx drivers installed with no compatible card.... hmm guess it gives me an excuse to use my old 8600gt as a physics card.

That don't do shit mate!

I'm using a 8800gt and there's no difference having it enabled or disabled

[alexsubri]

Okay okay here I go!

Quick Question I have the Free Version, am I excluded or included ? :-P

I will buy the $20 version soon.....

Alexsubri | ATI 5850 Crossfire | 900/1250 MLAA Enabled | 6341 | AMD BE 965 (c3) @ 3.6ghz



Am I doing it right?

[Bjorn_Of_Iceland]

Originally Posted by ShiBDiB

i crash during the physics test pretty sure its cause i have the physx drivers installed with no compatible card.... hmm guess it gives me an excuse to use my old 8600gt as a physics card.

Its using Bullet physics middleware. Its not even touching a single byte of any physx related runtime code. Like saying 'ah crap, Ive webcam driver installed without my webcam, its going to hurt my score'

[HammerON]

HammerON | EVGA GTX 470 | 860/1000 | 5709 | Intel i7 920@3.9 GHz

[Fitseries3]

I have 24cores at 4.5ghz and 4gtx480s scoring 56xxpts. This is bullshit!

[Fitseries3]

its weird, the physx test wont run unless i specify cpu only physx, and gpus show 0% use throughout the entire bench runs. i cant get more than 30fps

i can run a 48k pts vantage run right after but only get 5600pts in 3dm11



EDIT: says sli is disabled in 3dm11.... WTF?

[puma99dk|]

i ran 3DMark11 on my i3 setup without oc'ing my cpu bcs i got problems with my connection and it turns out it's not my pc, so i haven't enabled OC Genie again so it's a stock I3-540 and the rest u can see in my System Specs and on the screenshot

[heky]

@puma99dk|
I get 3709 on my GTX460 @ 850/1700/1025 and q6600 @ 3520mhz.

[puma99dk|]

Originally Posted by heky

@puma99dk|
I get 3709 on my GTX460 @ 850/1700/1025 and q6600 @ 3520mhz.

mine is just a Dual-Core with a split personality it got HT (Hyper-Threading) so Windows finds it as a Quad-Core and heky my CPU is not overclocked atm, but i think i will be needing more than just the stock cooler and some hours of reading how to oc on LGA1156 platform if OC Genie won't crack my little i3-540 on a great Thermalright cooler up to 4ghz

[Live OR Die]

LiveOrDie|EVGA 480|800/1000|5909|i7 920 @ 3.6ghz

[heky]

@puma99dk|
I know it is a dual core, i just wanted to make the comparison becouse people said the benchmark is very GPU dependent. And since we both have a 460, but yours is clocked 50mhz higer on the core, 100 on the shaders and i still get a higher score, its an interesting comparison.

I say you go for it and oc that i3 to the max, and report the score then.

[bear jesus]

I don't want to be a pain in the butt, but could i ask to have my results changed to what i list below as these are my absolute top scores that i could get and can't improve on.

single card
bear jesus|6870|1045/1125|4555|phenom II 965 @ 4.03ghz


dual card
bear jesus|6870 crossfire|1025/1120|7155|phenom II 965 @ 4.07ghz


both scores the cpu was pushed a tiny bit higher on the core speed but i think with the crossfire score it was pushing the HT, core, memory and NB that helped out as my phenom holds back a pair of 6870's

[douglatins]

Unigine is really consistent, a lot more that i have seem from fullinfusion

[Rado D]

Originally Posted by Live OR Die

LiveOrDie|EVGA 480|800/1000|5909|i7 920 @ 3.6ghz

http://img694.imageshack.us/img694/4498/73278608.png

Sorry budy,now I'll just have to beat your score

[MoonPig]

I've edited post #3 to show how to apparently enable SLi.

[Live OR Die]

Originally Posted by Rado D

Sorry budy,now I'll just have to beat your score

bring it

[Athlonite]

Originally Posted by jpierce55

I used the original 5770 driver first time and it done the same thing. The current and new Catalyst worked fine, and has all the options the old one did.

errr it's not the first time I've run 3DMK11 it only went DERP when I OC'd my CPU to 3.4GHz I didn't change anything else

[Rado D]

HEY GUYS!!! Now Im going to cause some debating here!!!

I dont know about you,but I've just found out trying some random stuff,that The overall Image Quality Setting in your NVIDIA control panel or Catalyst does have a huge impact on the score!!

Until now I thought that 3dMark11 entirely handles its own settings to make the scores more objective,but it seems that I was wrong...Now I think we have to make some changes in the rules(like adding one which tells us to use the highest image quality settings,cos that seems the only way to make it as objective as possible)..but I leave this for AthlonX2 and to rest of you guys to verify this by yourself..

EDIT:experiment results:i7 950 @ 4.140GHz + GTX 480 835core/1000memory with NVIDIA control panel image settings at quality score P5851
control panel image settings at balanced score P5929

EDIT:control panel image settings at performance score P6130

Now this is ridiculous!!! We all know that the control panels for VGA cards have the control over the overall performance but should be a benchmark with such a big name as objective as possible....well according to my test objective my arse...they should make the benchmark settings override the control panels settings to one same settings everybody would use...

[EastCoasthandle]

Originally Posted by Rado D

HEY GUYS!!! Now Im going to cause some debating here!!!

I dont know about you,but I've just found out trying some random stuff,that The overall Image Quality Setting in your NVIDIA control panel or Catalyst does have a huge impact on the score!!

Until now I thought that 3dMark11 entirely handles its own settings to make the scores more objective,but it seems that I was wrong...Now I think we have to make some changes in the rules(like adding one which tells us to use the highest image quality settings,cos that seems the only way to make it as objective as possible)..but I leave this for AthlonX2 and to rest of you guys to verify this by yourself..

And solid proof for this?

[Rado D]

Originally Posted by EastCoasthandle

And solid proof for this?

just edited my previous post

[EastCoasthandle]

Originally Posted by Rado D

just edited my previous post

EDIT:experiment results:i7 950 @ 4.140GHz + GTX 480 835core/1000memory with NVIDIA control panel image quality at quality score P5851
control panel image quality at balanced score P5929

Ok, this is interesting. Which drivers are you using?

[Athlonite]

I'll check it out on the HD5770 see what happens

[Rado D]

Originally Posted by EastCoasthandle

Ok, this is interesting. Which drivers are you using?

I've added another score to my first post...Im using the latest drivers made for GTX580,the 263.09 with modded .ini file to be able to use this package with my GTX 480

[Athlonite]

with CAT AI on high I get P2638 3DMarks with CAT AI on quality I get P2757 3DMarks