It's impossible to tell what the real 3D clock speed of the 3870 X2 was, it still may be fanboy slant - then again DUH, a forum for fanboys, what do you expect?

The 9800 has a 20%+ overclocked speed advantage over stock. I doubt the 3870 was clocked much. I was hoping to see a apples to apples comparisson.... 

A fairer dollar-for dollar would be X2 + 3870 for three way CF which could turn the tables.
Admittedly it does look like the 3870X2 has been underclocked however its hard to be sure as once in 2D mode the card should underclock itself to about that level.

Regardless its hard to be sure and the results aren't really valid anyway since the 9800GX2 is overclocked.

I want to see some solid trustworthy results. Maybe The INQ could whip some up for us?
I think their new standard of allowing only one free test is bullshite and I'm not afraid to say it. The Onion sold out, face it. They used to be about the gamers and now they're only about themselves.
The screenshots are 2d mode.

The Radeon downclocks itself in 2d mode to conserve power.

Morons.

You honestly thing it could score 8xxx when only half clocked??? lol
Other than to people that spend more time "testing" (and i use the word loosely) their rig than actually playing with it? I mean, really folks, who gives a crap what a great 3DMark score a card has, if it still runs games only mediocre? Who cares that some idjit overclocked and nitrogen cooled their rig to get 1 bazillion marks, since it doesn't translate into real-world performance? Who cares since both companies hack the s**t out of their drivers in order to "win" using methods not applicable to any real game?

Personally, Kyle Bennett's approach is the proper one, take the cards, and run them flat out adjusting the settings to get the best experience per game. 3DMark should die an ignominious death...
Nvidia hit it out of the park. Poor ATI, maybe their 770 will make up the difference. Finally nice to see what performance is like in a pure DX10 enviroment.
I've downloaded it and given it a go, it took about three tries just to get into the application because they're running everything through their servers which are being hammered at the moment. 

I don't have a score yet either as it crashes about halfway through the tests claiming it's tried to make an invalid DirectX call. I'm not sure if it's the Nvidia drivers being rubbish (which wouldn't surprise me) or if Futuremark needed to spend a bit more time bug testing.
Well, if you really took the time to check the screenshots, you would see radeon being downclocked(438 from 850) and Nvidia overclocked (from 600 to 732). So the results are not really valid. Anyway 45min till it comes. (new Vantage benchmark that is)
Who's getting card-spanked? Is the splurg really working? General dismahem (that's what this is)!
It's impossible to tell what the real 3D clock speed of the 3870 X2 was, it still may be fanboy slant - then again DUH, a forum for fanboys, what do you expect?

The 9800 has a 20%+ overclocked speed advantage over stock. I doubt the 3870 was clocked much. I was hoping to see a apples to apples comparisson.... 

A fairer dollar-for dollar would be X2 + 3870 for three way CF which could turn the tables.
"which appears to text throughput"
Admittedly it does look like the 3870X2 has been underclocked however its hard to be sure as once in 2D mode the card should underclock itself to about that level.

Regardless its hard to be sure and the results aren't really valid anyway since the 9800GX2 is overclocked.

I want to see some solid trustworthy results. Maybe The INQ could whip some up for us?
I think their new standard of allowing only one free test is bullshite and I'm not afraid to say it. The Onion sold out, face it. They used to be about the gamers and now they're only about themselves.
The screenshots are 2d mode.

The Radeon downclocks itself in 2d mode to conserve power.

Morons.

You honestly thing it could score 8xxx when only half clocked??? lol
Other than to people that spend more time "testing" (and i use the word loosely) their rig than actually playing with it? I mean, really folks, who gives a crap what a great 3DMark score a card has, if it still runs games only mediocre? Who cares that some idjit overclocked and nitrogen cooled their rig to get 1 bazillion marks, since it doesn't translate into real-world performance? Who cares since both companies hack the s**t out of their drivers in order to "win" using methods not applicable to any real game?

Personally, Kyle Bennett's approach is the proper one, take the cards, and run them flat out adjusting the settings to get the best experience per game. 3DMark should die an ignominious death...
Nvidia hit it out of the park. Poor ATI, maybe their 770 will make up the difference. Finally nice to see what performance is like in a pure DX10 enviroment.
I've downloaded it and given it a go, it took about three tries just to get into the application because they're running everything through their servers which are being hammered at the moment. 

I don't have a score yet either as it crashes about halfway through the tests claiming it's tried to make an invalid DirectX call. I'm not sure if it's the Nvidia drivers being rubbish (which wouldn't surprise me) or if Futuremark needed to spend a bit more time bug testing.
Well, if you really took the time to check the screenshots, you would see radeon being downclocked(438 from 850) and Nvidia overclocked (from 600 to 732). So the results are not really valid. Anyway 45min till it comes. (new Vantage benchmark that is)
seem like fake nvidia fanboy potatoshop.
3870 X2 is half-clocked and 9800 X2 is overclocked...