9800 GX2 Benchmark is out!

[Squirrely]

well i jsut ran 3d mark again, same settings but @ 1600x1200 and scored 8281...

maybe they just wrote the res down wrong?

I don't think they wrote the wrong res down. As here is the scrennie of their 3dmark06 scores:
http://bp2.blogger.com/_veswGS4uNJQ/R8z-yeav_yI/AAAAAAAACDM/ZEBeRgCiqsk/s1600-h/Hardware-Aktuell_GeForce_9800_GX2_3DMark06.jpg

From the review which is on the first page. Its the correct res which they claim...hmm

 

[wolf]

well i just reran the test on the original settings (1280x768, 4xAA, 16X AF) and this time cranked the AA Quality to 7 and scored 10,398 .... down from the ~10,800 i pulled before, and with more ROPS and Shaders and Pixel fillrate the GTX and Ultra should have more AA grunt that my GT anyway.....interesting results.... can post a screenie if anyone so desires.

 

[magibeg]

I'll just have to wait and see before i make any judgements on nvidias new card. I would hope though that it performs a little better than that, however it does just seem to be a die shrink and such. Not sure why everyone is expecting miracles from it.

 

[reviewhunter]

Guys , reviewhunter has brought up a good point in the bench i ran earlier , about 4aa runinng at level 7 , which makes more sense to me now ..................... i think we will have diff score numbers after i run another benchmark .............. =)

Hmmm, my testing shows only a 5% fall in score when AA quality is set to level 7.

 

[mandelore]

Hmmm, my testing shows only a 5% fall in score when AA quality is set to level 7.

in that case then the 3870X2 is a beast!

I was worried that it may have turned out that setting 7 really crippled the performance

still waiting for an ultra to give us a proper estimate of a good score at those settings

 

[InnocentCriminal]

Someone will have to whack on the 8.3's and see if they're is any increase with the X2's.

 

[wolf]

Hmmm, my testing shows only a 5% fall in score when AA quality is set to level 7.

my test showed a drop in performance just Under 4% by changing from AA quality 0 to 7

 

[DrunkenMafia]

3dmark06 only:

http://lly316.blogspot.com/2008/03/nvidia-geforce-9800-gx2-benchmark.html

come on please...

Do you really think that those cards score that low at such a low res, even with some AA etc turned on..

I really can't believe this for a second :shadedshu

 

[InnocentCriminal]

Hence why I'm holding my judgement until we see some legit reviews. Which should be in the next couple of weeks I'd expect.

 

[wolf]

techreport, guru3d... some reviews ill actually think have a glimmer of being right. and TPU ones naturally.

 

[mandelore]

come on please...

Do you really think that those cards score that low at such a low res, even with some AA etc turned on..

I really can't believe this for a second :shadedshu

hence us wanting a tpu member with one to run at the same settings with AA lvl 7 enabled

 

[Nitro-Max]

The test rig was doing 3.2ghz as well!! My quad @ 2.4ghz stock was getting over 12k with my 3870x2

 

[wolf]

well its 2 full 128 shader G92's i cant really see it doing much worse than 8800GT in SLi right? i mean it makes sense. should be around G92GTS in SLi, man i cant wait for it already!

 

[zOaib]

my test showed a drop in performance just Under 4% by changing from AA quality 0 to 7

i am not being able to have the option to select level 7 , When i try to do 4aa sampling ..... the box below is not higlighted to make the change , does this has to do with limitations b/w professional and advanced versions ??? thx

really wanna run this benchmark right i am itching bad here

 

[EastCoasthandle]

90% faster then the 8600 GTS

 

[zOaib]

90% faster then the 8600 GTS

lol , you know i didnt think nvidia can lay such a lame comparision for this beast of theirs ......

 

[DrunkenMafia]

^^^

Haahaaa. Is that picture serious!!! It might as well say 90% faster than 6100se if that is the case. 8600 isnt exactly a fast card. I would expect just 1 of these so called 9800GTX cards to be at least 200 percent faster than a 8600, let alone a GX2 version....

that pic is a mock up isn't it???? Its gotta be

 

[wolf]

they were probably suppost to write 8800 GTS, it could be a typo, ya know, it being "up to" 90% faster than one 8800 GTS its pretty downright logical considering what it is. considering it IS two 128 shader G92's....

naturally it wont always scale at 90%, but under "some" conditions it "may" be "up to" 90% faster

 

[bud951]

I get over 16k in 3dmark06 with my 2x 8800GT's in SLI overclocked of course. This seems abit dissapointing to me. I was thinking of getting one 9800gx2 and hopfully getting the same performance without as much heat and noise but I may be dreaming about the heat anyway. I might as well wait for the next gen. 9900, 10000? Who knows?

 

[reviewhunter]

this is confirmed,
the 9800 GX2 is retailing at slightly less than US$600

 

[yogurt_21]

they were probably suppost to write 8800 GTS, it could be a typo, ya know, it being "up to" 90% faster than one 8800 GTS its pretty downright logical considering what it is. considering it IS two 128 shader G92's....

naturally it wont always scale at 90%, but under "some" conditions it "may" be "up to" 90% faster

such mistakes rarely make it to cebit.

 

[Tatty_Two]

I get over 16k in 3dmark06 with my 2x 8800GT's in SLI overclocked of course. This seems abit dissapointing to me. I was thinking of getting one 9800gx2 and hopfully getting the same performance without as much heat and noise but I may be dreaming about the heat anyway. I might as well wait for the next gen. 9900, 10000? Who knows? [/QUOT

That does seem a little low, there are people here who score over 16000 with one 8800GT.

 

[candle_86]

90% faster then the 8600 GTS

thats gotta be a typo, my 8800GS box says that

 

[Tatty_Two]

yeah I am sure it should have read 90% faster than an 8400GS

 

[reviewhunter]

That does seem a little low, there are people here who score over 16000 with one 8800GT.

That's cool, I only hit 12k+ with a C2Q @ 3.2ghz & an o/c (un-modded) 88gt 512