3DMark05 Benchmark Results

[DMA]

i really hope that the 1000 point boost ati cards get is full of ****...cause the nvidia is getting hammered to the floor again.

8.07 Beta is Approved by Futuremark. :)
http://www.futuremark.com/community/drivers/?approved

[Blacklash]

Lately I have been running my rig at moderate settings. After serious number crunching I didn't really see the sense in driving my hardware in the dirt for benches.

This is my CPU at 3.4 and my card at 440/1.15, driver 66.51 >5426,

http://service.futuremark.com/compare?dm05=6564

EDIT Here's some other numbers with the video card remaining at 440/1.20, CPU@3.2|2-2-2-5 timings=5352, CPU@ 3.7|3-3-2-6 timings=5531.

One last add, my entire system at flat stock speed with AA/AF Application controlled and the 66.51 driver set to quality=5034.

[Spliffstarr]

http://service.futuremark.com/compare?dm05=3422

66.70
426core/1111mem
P4E 2.8@3.2
BFG 6800GT OC
:nanahump: :nanahump: :nanahump:

[PeterJensen]

Well i simply dont trust you can pull 1000 marks out of nowhere, and that it only affects ONE benchmark. There are no score increase in 3dmark03 with those Ati "beta" drivers....

[aAv7]

Yes I know that, it's the games that really matter it always is, but I wanted to get scores that are higher than ATI, I just bought the card.

I agree...it is somewhat dissapointing that a x800 beats a gt....I hope this isnt a sign of things to come.

Something that always bothered me in 2K3 was the 5-10fps difference in the vertex shader test. My old x800 pro would get 35fps in the test, while my gt on a good set of drivers gets 29fps. I've seen other pro's with up to a 10fps lead on gt's in the vertex shader test. Why nvidia lacks 3d3 performance, I dunno. They have such great opengl support...but the problem is...for every 50 games released...maybe 2 are opengl. They need to improved d3d performance imo. And not just based on this benchmark, but also based on in-game performance. I miss my old x800 pro sometimes....damn sm3.0 ..I bought into the hype =P

[Danhill]

I agree...it is somewhat dissapointing that a x800 beats a gt....I hope this isnt a sign of things to come.

Something that always bothered me in 2K3 was the 5-10fps difference in the vertex shader test. My old x800 pro would get 35fps in the test, while my gt on a good set of drivers gets 29fps. I've seen other pro's with up to a 10fps lead on gt's in the vertex shader test. Why nvidia lacks 3d3 performance, I dunno. They have such great opengl support...but the problem is...for every 50 games released...maybe 2 are opengl. They need to improved d3d performance imo. And not just based on this benchmark, but also based on in-game performance. I miss my old x800 pro sometimes....damn sm3.0 ..I bought into the hype =P

Yes, agree with you there regarding the SM 3.0. But maybe Nvida is having something up their sleeves.

[jbirney]

Well i simply dont trust you can pull 1000 marks out of nowhere, and that it only affects ONE benchmark. There are no score increase in 3dmark03 with those Ati "beta" drivers....


Its been doucmented that it was for AGP only that saw that increase due to poor texture memory management on the 256 mb card when you have more than 256 mb worth of texture data. Before crying foul why dont you see if there are any other cases (ie games) where this does not work (meaning the AGP version does not see any increase using 256 mb cards with more than 256 mb of texture memory in DX). I am not trying to say that is the case...just until you/we find a case where it does not work, then we have take ATIs word for this....

[Sazar]

CPU: a64 3400+ @ 2.2ghz
Motherboard: Msi Neo Fis2r, bios rev == 1.8
Memory: 2x512 Corsair XMS, ddr 400, timings == 2.5/3/3/7
Video Card: ATi AIW 9700 pro, 324/310, cat 4.9's
Hard drives: Dual Maxtor DM9's, 80gig each

score == 2150
http://service.futuremark.com/compare?dm05=4638

I will attempt a run with the 4.10's when the debut next month and will update accordingly...

[aAv7]

Yes, agree with you there regarding the SM 3.0. But maybe Nvida is having something up their sleeves.


Eh...i've been hearing the "wait and see, something up their sleeves" since June when the products first showed weakness in d3d. =/ ( No offense to you)

I'm starting to think this is about as good as it gets with nv40. All those games that allegedlly had or were going to have sm3.0 suddenly have dissapeared or been delayed. Battle for middle earth and the painkiller expansion are the only sm3.0 titles for 04. (now that splinter cell and stalker have been delayed) And sm3.0 in those games only speed up shaders...something ati can easily do with 2.0b just as well, if not better. All the I.Q changes they led us to believe could and would exist just arent there.

Even the upcoming farcry patch will only have HDR enabled...to my understanding...r300 cards could do that. (which again shows they haven' used true sm3.0...just simply paid off companies to use features specifically for the nv40 that in reality can be done on r420 or lower just as easily)

I gotta find someone with a x800 pro they want to trade.

[Omega53]

I gotta find someone with a x800 pro they want to trade.

Check pm ;)

[bobmanfoo]

xp2800+ 233x10, 1g pc466, GT 440/1.1ghz

5439 points

[Smokey]

Me cries, I think the highest I saw my fps was 8! Total of 1135, cpu @3000+ and gpu @425/925.

[JD-Scorpion]

My score with 66.70 Forceware driver: :bleh:
http://12.22.230.41/MicroTech/Hosted/Images/2005.jpg
Video score: 5374
CPU score: 4587

[lordtyranus2]

4594

http://service.futuremark.com/servlet/Index?pageid=/orb/projectdetails&projectType=12&projectId=3271

[joeuser65536]

Man...today is the day 3dmark05 is released, but my card also died today.

Please help me running 3dmark05 by clicking here: http://www.nvnews.net/vbulletin/showthread.php?p=437261#post437261

:lame:

[EMunEeE]

I tell ya, the 5900...
(mag)

[Warden13]

Well i simply dont trust you can pull 1000 marks out of nowhere, and that it only affects ONE benchmark. There are no score increase in 3dmark03 with those Ati "beta" drivers....

Well you can say that but it is Futuremark approved and ATI has managed to consistently produce Fm approved and WHQL approved drivers all year long. Nvidia has been promising WHQL drivers for ages and there still aren't any on their site (at least as of a couple days ago).

Warden

[Dr Keyzer]

Me cries, I think the highest I saw my fps was 8! Total of 1135, cpu @3000+ and gpu @425/925.

Maybe this will cheer you up :) http://service.futuremark.com/compare?dm05=4419


XT PE beat NVidia's SLI?: http://service.futuremark.com/compare?dm05=5340

[strk]

I'm getting 3992 which i think is low for my system any help will be appreciated for help in settings maybe i have something wrong.

[bobmanfoo]

I'm getting 3992 which i think is low for my system any help will be appreciated for help in settings maybe i have something wrong.
dunno, but i notice all the a64s are scoring lower than the OCed XPs and p4s

[Mr^Yahtzee]

61.71 driver core=370 mem=1000 score=3,975

[|JuiceZ|]

XT PE beat NVidia's SLI?: http://service.futuremark.com/compare?dm05=5340

I don't think NV mentioned o'cing the cpu or gpu's in their press release. Not that the world will end because an ATi card scores higher in this benchmark though.

[Dr Keyzer]

I'm getting 3992 which i think is low for my system any help will be appreciated for help in settings maybe i have something wrong.

Perhaps you've forced AA- or AF-settings in the NV-cpl?

[Sazar]

dunno, but i notice all the a64s are scoring lower than the OCed XPs and p4s

you are noticing hte exact opposite of what I am seeing :)

stock a64's are beating out other stock cpu's...

oc'd a64's are generally beating out other ocd cpu's... unless its a massively oc'd p4... then its closer...

wrt strk's score... its lower than the average score for the cpu I have seen...

i ran my benchmark with azureus and other apps running in the background.. didn't seem to affect the cpu score too much...

[strk]

no there off and i have it at performance but trillnear and anisotropic optimization are On.