AMD vs Intel

[druga runda]

Whooo????

at Extremetech... in R9700 review.

The guy there seems to have sold his sould to Intel , but the benchmarks give you totally different story that he himself in the comments below makes.

First of all the guy uses the latest and meanest Intel setup.
PIV 2.53, 1066RDram. vs lowly 2000+AMD xp with 333DDR.

And guess what suprised me!?! AMd BEATS PIV in many benchmarks with FSAA on. Yes correct BEATS the PIV system, comparing GF4 or R9700.

The biggest beating is in Nature test

R300 4xFSAA 1024*768
AMD = 74.5 FPS
Intel = 53.2 FPS

Other than that the scores are pretty much the same in that test despite Intels huge mhz/FSB etc overall advantage.

I was thinking maybe this is just misprinted graph, but than go to first page and looked at 3D Mark test once again

R300 AMD 1024*768 4xAA
AMD 11315
Intel 10748

add 8*AF
AMD 9236
Intel 8932

do 1600 x 1200 with 4xAA & 8*AF
AMD 5454
Intel 4959

And in genereal when you crank up the settings AMD gets either close or beats PIV.

I for one did not expect that. It seems to me that PIV has the most trouble with AA, as if it prediction unit cannot handle the data that well.

Interesting benchies indeed. (apart form Extremetech dully, saying how Intel in faster, pointing out to lowest resolutions/games that favour Intel despite Intels hughe mhz/fsb advantage and higher cost, and the system not even being the best from AMD)
This fact that AMD managed to squeeze more out when it mattered most is indeed suprising.

[Kruno]

Bad driver optimisations for Intel

Originally posted by druga runda
I for one did not expect that. It seems to me that PIV has the most trouble with AA, as if it prediction unit cannot handle the data that well.

err, FSAA is done in hardware on the video card

[Matthyahuw]

Originally posted by K.I.L.E.R
Bad driver optimisations for Intel

Kind of the opposite of what happend with the R8500, when it 1st came out, it barely even ran on AMD systems...

[Kruno]

ROTFLMAO!
Ati better have their drivers optimised for both cpu's to make sure they get a great market share.

I might jsut add I have seen some benchmarks of the R300 showing the opposite as well, running better on Intel P4's than AMD Athlon XP's.
All I have to say is: Bulldust. Either the reviewers are doing something or Ati are playing around with drivers again.

[druga runda]

Originally posted by StealthHawk


err, FSAA is done in hardware on the video card

That is what all they say, but there seems to be consistency in his review, who knows what data goes trough the CPU - as something definitley slows PIV.

Plus looking at Hardocp, 1800+ Atholon is either the same, or beats PIV 2.53 a little when the resolution and settings go up.

What was the review that went the other way around?

[Kruno]

Dunno now. Me thinks somewhere on Inquirer.

[SavagePaladin]

Comparison benchmarks can be kind of silly, since theres no way to get a p4 and AMD system with the same parts except the processor+RAM setup...
and AMD definately has better floating point, from what I know