X850XT PE is slower than X1600XT !!

[Gamer X]

INQ Link

 

[Corporate Thug]

hmmm, gotta see some real world benchies before jumping on the x1600 wagon but thats a nice start

 

[yhelothar]

Who cares about what 3dmark says..
Who even takes 3Dmark for more than a grain of salt?

 

[Frostwake]

YOU'D EXPECT wouldn't you, that the Radeon X850 XT PE - equipped as it is with a full sixteen pipelines - would be much faster than the new eight-pipeline X1600 XT card.

Am i missing something here? 8 pipeline x1600? hmm 2400 is still nice though isnt it? i mean.. in sm2.0b it gets 883 vs 1153 on the x850, thats great for the price

 

[Budarow]

If 3dmark06 was my favorite game, I'd be sure to replace my X800XTPE with a X1600XT:Q

 

[CKXP]

we all know the x850xt pwnz the x1600xt, but when you look at the SM3.0 score for the 850xt it's N/A....the x1600xt has a SM3.0 score of 896. 3Dmark06 shouldn't be used to judge the performance difference between the x850xt and x1600xt, take a look at the real world benches instead(hl2,farcry..etc)

 

[v8envy]

3dmark has the problem of being 'damned if they do, damned if they don't'.

Here's what I mean. In 3dmark05, if a card doesn't have capabilities to run a test, the test isn't counted. So a card with no SM3.0 functionality may get more 3dmarks than a card with otherwise identical performance, but with poor SM3.0 functionality. This is why the X850XT pulls in 3dmarks closer to a 7800 than it should.

Looks like that's not the case in 06. A card gets docked severely for not being able to do a test. Which is why a 4 pipe X1600XT 'outperforms' an X850XT.

For current games, an X850XT is still >>>> 1600. And we all know you can't upgrade for future games. Which is why I find the 1600 amusing. It's slower than the competition for current/past games, and will probably not be ballsy enough for future games where in theory it will outperform it's current competition.

 

[Gamer X]

People !! the purpose of the Inquirer article is to say that there is something wrong with
3Dmark06 not that X850XT PE is slower than X1600XT .

 

[Cooler]

Its the SM3.0 thats all.

 

[MegaWorks]

Originally posted by: Cooler
Its the SM3.0 thats all.

 

[NoStateofMind]

Originally posted by: CKXP
we all know the x850xt pwnz the x1600xt, but when you look at the SM3.0 score for the 850xt it's N/A....the x1600xt has a SM3.0 score of 896. 3Dmark06 shouldn't be used to judge the performance difference between the x850xt and x1600xt, take a look at the real world benches instead(hl2,farcry..etc)

Nail on the head.

 

[dguy6789]

All 3dmarks score by using an algorithm similar to this:

test 1 fps x test 2 fps x test 3 fps x test 4 fps ect... = score

If one video card lacks the hardware to do a test, then that test is not counted.

97*87*34*42 = 12,050,892

200*200*200 = 8,000,000

As you can see, a card that clearly is faster but less feature capable will receive a lower score than a slower card that can run all the tests.

 

[Budarow]

I never look at results from "bench marking" software. I only look at actual game performance.

 

[apoppin]

If you want co "compare" cards capabilities in current games . . . use 3DMark05 . . . in which case the 1600xt "loses"
[badly]

otoh, i AM looking forward to the x1700xt to see if it is faster . . . and theinq's title is:

3Dmark06 quirks confuse us all

While we can allow that Shader model 3.0 is becoming important, as at least a few titles are shipping with HDR enabled, it seems that the weighting in the tests is skewed. And besides, every single one of these titles has an option to use Shader model 2.0 instead. So, except some of fancy visual experience, an X850 card can run all these games just fine.

This leads us to conclusion that 3Dmark06 doesn?t indicate the real performance of the card. It gives us some bogus number that might be used only for Shader model 3.0 cards. It is a very relative thing rather than just based on pure logic.