9800 GX2 Benchmark is out!

[Tatty_Two]

Got your numbers a bit mixed up mate, you need to divide the GX2's by the ultra's, works out at about a 22% boost. When the X2 is 50fps faster than the ultra in some benchmarks, shows that it's got a lot of potential. At the moment it seems my X2 is cpu limited(q6600 @ 3.9ghz) so it's plenty fast for me. Thing is though is that we aren't seeing the usual generational performance leap, it's usually about 2x last gen's high end, and we're only seeing 20% at the moment, i'm sure some of you are a bit shocked. That's why i reckon nv have been caught off guard i think they weren't expecting the X2 to surpass their ultra by much.

Yeah but it's not too difficult to pass the Ultra in some things.....my G92 8800GTS 512MB at 830 core does that in some benches, just like the single G92 GTS beats the 3870x2 in a couple of benches dont get me wrong, I am a huge admirer of the x2, might even get me one or two but I actually do think that the GX2 will be pretty quick, just how quick remains to be seen, if the price leaks are correct, it needs to be at least 25% quicker than the 3970x2 otherwise noone will buy it for the increased costs over the x2, if it's just 5% faster....then thats fine as long as it's no more than 5% more in cost otherwise again, not many will buy it.

 

[Tatty_Two]

hehe. sry guys, I know its early for the gx2, and it will obviously get some improvements with driver revisions, but 10k from Gx2 vs 14.5K from X2 kinda makes me grin. (with Gx2 on faster cpu 3.2ghz york compared to 3ghz regular quad)

BUT, again, we need both cards ran at hi rez to really see how big their balls are

We all know even a 8800GTX and the 8800Ultra get more than that so I wouldnt put to much store in a site that mentions "blog"

 

[Tatty_Two]

mind providing some screen-shots?

And I get 17,211 with my single 8800GTS 512MB G92.......so the points is? (I would guess the point is that these figures are most likely to be cr*p)..........but then again....who knows

 

[mandelore]

And I get 17,211 with my single 8800GTS 512MB G92.......so the points is? (I would guess the point is that these figures are most likely to be cr*p)..........but then again....who knows

yeah but at what cpu speed and with 4xaa and 16xaf???

 

[Tatty_Two]

yeah but at what cpu speed and with 4xaa and 16xaf???

I suppose my point is....anyone can throw around benchmarks, it dont mean a lot, and I have gotta say (despite me just ordering a 3870x2) that that reviews seems pretty dodgy to me......I dont like "blogs"

 

[mandelore]

I suppose my point is....anyone can throw around benchmarks, it dont mean a lot, and I have gotta say (despite me just ordering a 3870x2) that that reviews seems pretty dodgy to me......I dont like "blogs"

I order you to break into an NV lab and steal one!!

then we would have reliable info

 

[das müffin mann]

ill do it if you pay my legal fees

 

[reviewhunter]

I suppose my point is....anyone can throw around benchmarks, it dont mean a lot, and I have gotta say (despite me just ordering a 3870x2) that that reviews seems pretty dodgy to me......I dont like "blogs"

It came from a German tech website.

 

[farlex85]

Yea I'm not real sure why people are throwing out other benchmarks and saying their cards are better. The main thing to look at with this (if its even to be trusted) is the comparison with the other nvidia cards. I'm sure if you guys threw in an 8800gts(g92) or an ultra in your system instead of the x2, it would get higher scores than is listed there. Hell my gts scores a hell of a lot higher than those scores(not as high as tatty's though, damn impressive, thats what a xeon and fast ram will do for ya), but I certainly don't assume my card is more powerful than an ultra or x2. Come on.

 

[mandelore]

ok, your saying an ultra with 4xaa and 16af will get higher?

show me, out of curiosity

 

[wolf]

ok well take this for instance guys, proof an Ultra should get more.

thats using a DUO not a quad, with a quad i could be looking at around 12k, and a GT shouldn't best an ultra with AA + AF, especially not by almost 3000 points at the same settings, and almost half the Cpu power.

imo those results are either bogus or done by an idiot.

 

[mandelore]

we needs some1 with an ultra to test who also has a quad, and not with any lod bias set incase any1 got a temptation

we can at least trust what we bench. any volunteers?

 

[mandelore]

imo those results are either bogus or done by an idiot.

or it really could just be borked drivers? is nv due an update anytime soon?

 

[wolf]

so you'll believe the random review but not a TPU member's score?, every result i ever post here is guaranteed genuine by me, who could i possibly want to fool by posting a altered result?

i think my results speak for themselves in discrediting at least the GTX/Ultra scores.

 

[Tatty_Two]

I am not saying it's false, it just seems a bit low to me, is it a coincidence that there are no ATi cards there to compare?

 

[mandelore]

so you'll believe the random review but not a TPU member's score?, every result i ever post here is guaranteed genuine by me, who could i possibly want to fool by posting a altered result?

i think my results speak for themselves in discrediting at least the GTX/Ultra scores.

huh? you dont have an ultra, i was stating we needs some1 with one to post a genuine score for comparison. and i certainly do not think you would have altered anything. hence it must be down to driver issues, but simply out of curiosity, i would like to see an ultras score at those settings

edit: also, is than card oc'd at all?

 

[wolf]

yes, yes it is, when have i NOT oc'ed a card , but even at my clockspeeds, on paper at least, it does not best an ultra.

i too want to see ultra scores, personally i just think that test was wrong, maybe they just wrote the res down wrong? it seems more like a score of 1600x1200+

 

[candle_86]

Got your numbers a bit mixed up mate, you need to divide the GX2's by the ultra's, works out at about a 22% boost. When the X2 is 50fps faster than the ultra in some benchmarks, shows that it's got a lot of potential. At the moment it seems my X2 is cpu limited(q6600 @ 3.9ghz) so it's plenty fast for me. Thing is though is that we aren't seeing the usual generational performance leap, it's usually about 2x last gen's high end, and we're only seeing 20% at the moment, i'm sure some of you are a bit shocked. That's why i reckon nv have been caught off guard i think they weren't expecting the X2 to surpass their ultra by much.

we have just been spoiled is all. The Geforce 6 over the FX really did it to us, the 7800GTX over 6800Ultra was about 1.5x and the 8800GTX over 7900GTX was about 1.7x

 

[das müffin mann]

i didn't mind being spoiled at all, but i to would like to see how the ultra would score any takers?

 

[wolf]

so which TPU memebrs actually have a Quad they can clock at 3ghz and an Ultra, or even a GTX for that matter.

 

[reviewhunter]

AA Quality Level

ACTUALLY I wud like to know that myself , looking at their 3dmark06 screenie i tried to match what i saw , but i dont see any level 7 options when i place 4aa in selection , anyone who knows anythign about this , please let me know ill run another benchmark if there is a settign change , thx for pointing that out.

It does make a difference to the score.

 

[reviewhunter]

Please ignore the score in my screenie.

As I can't test it out with my system because my old 3DMark06 v1.0.2 does not support 1280x768.

 

[zOaib]

i dont get that option , maybe a diif b/w professional and advanced .............. am at work will fiddle with it once i get home.

 

[wolf]

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

maybe they just wrote the res down wrong?

 

[zOaib]

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 .............. =)