The first published 2P Barcelona 3dmark06 test result - here, on XS

[BitpowerPM]

Just ran this on my 2.33GHz dual Clovertown, totally unoptimised, all I did was turn BOINC off for the run. OS is Vista X64 btw.

Wow, that's really unoptimized.

My dual clover E5320@2,33 GHz (like yours) score over 5.000 easily while running a virtual machine during the benchmark.

The AMD results @ 2,0 GHz are nothing special - no chance against the good old clovertown clock per clock in 3D Mark 06.

cu, BitpowerPM

[kiwi]

I probably shouldn't post this but here is what Kinc did:



http://www.nordichardware.com/news,6863.html

If AMD could do this on air or close to that we would all benefit

[mAJORD]

BTW guys, don't ask me why but CPU scores are not comparable with different vid cards.

check out 2900Xt vs 8800 systems with identical CPU configs, the cpu scores are different between systems, and one of the startup tips confirms this

[NH|Delph1]

BTW guys, don't ask me why but CPU scores are not comparable with different vid cards.

check out 2900Xt vs 8800 systems with identical CPU configs, the cpu scores are different between systems, and one of the startup tips confirms this

NVIDIA simply has better drivers.

//Andreas

[MAS]

BTW guys, don't ask me why but CPU scores are not comparable with different vid cards.

check out 2900Xt vs 8800 systems with identical CPU configs, the cpu scores are different between systems, and one of the startup tips confirms this

+1
but is it so difficult for somebody who has 2P Xeon system to get 7950GT and run Mark? it would be interesting to compare AMD and Intel systems with the same video

[hollo]

Dumb statement as usual...
2900 Crossfire is also behind X1950XT Crossfire with overall game fps. So the point is what?

the point is that 3d'06 CPU and graphics benchmarks are biased towards AMD and ATI when compared to other (real world) benchmarks, for some reason - i was comparing between AMD and Intel/ATI and nVidia, rather than comparing ATI with ATI - the x1950xt with gddr4 beating the HD2900xt with gddr3 in overall games fps shows some heavy memory bandwidth bias, but that's a side point.

Using VGA charts totally drivers dépendant to conclude on CPU is totally pointless.

i didn't do that, i was just pointing out the stupidly good scores for ATI cards in 3d'06 gfx benchmarks. it helps explain why the 3d'06 world record is held by HD2900xt's in crossfire when they're obviously slower than 8800 GTX's SLI in most games. yes, it was a little offtopic.

[BeardyMan]

So Penryn on X38 and SkullTrial boards gets:

3DM 06 (CPU)
1x QC 3GHz (1333FSB) : 4569 (unreleased part)
2x QC 3.4GHz (1600FSB) : 6359 (unreleased part)
1x QC 4.14GHz (1840FSB) : 7000 (Kinc)
1x QC 5.2GHz (1600FSB) : 8092 (Kinc)

2x K10 2GHz : 4242

According to Intel at IDF. Not bad, not bad at all.

Thanks s7. Must be a good workout

taken from this qoute
2x QC 3.4GHz (1600FSB-->LOL) : 6359 (unreleased part)
1.8points per Mhz

2x K10 2GHz : 4242-->200FSB LOL
2.21points per Mhz

3.4ghz with 200fsb would result in 7500 points.
It's fair to say that Phenom will be teh real deal

[Shintai]

+1
but is it so difficult for somebody who has 2P Xeon system to get 7950GT and run Mark? it would be interesting to compare AMD and Intel systems with the same video

Just look at mine, its run with a 7900GT, dual 2.66Ghz clovertowns and scoring 5772.

Or 4330 if you do a raw linaer math down to 2.0Ghz.

So..not really impressive by K10...

[VulgarHandle]

wonder what the scores will be when the hardware, much less the O/S and 3dmark, recognize the cpu....

[BeardyMan]

Just look at mine, its run with a 7900GT, dual 2.66Ghz clovertowns and scoring 5772.

Or 4330 if you do a raw linaer math down to 2.0Ghz.

So..not really impressive by K10...

wrong.

[informal]

s7e9h3n has a possible knock out solution with working bios which supports K10s on the ASUS L1N64!

Also a note: K10s are B1s running registered 677 cas 5 memory in single channel(lol)!

[Shintai]

Oh I forgot, its another wait and see game. Then around xmas time its gonna be wait and see on 45nm and then wait and see on bulldozer...ye ok. Sorry my mistake. All hail AMD.

Wake me up when its actually faster...

[Tuvok-LuR-]

that's not bad, neither I'm too enthusiast beacause the real deal is how far it clocks

[informal]

lmao Shintai,same story all over again...

[Shintai]

lmao Shintai,same story all over again...

Ye, and we are still waiting for your magical numbers to appear.

[PhilDoc]

s7e9h3n has a possible knock out solution with working bios which supports K10s on the ASUS L1N64!

Also a note: K10s are B1s running registered 677 cas 5 memory in single channel(lol)!

Strange this didn't get mentioned earlier. The K10's are the crippled versions, running 667 fsb and single channel. It's a start, and I applaud S7 for his effort, but there's a ways to go.

[Movieman]

C.mon, Play nice guys or it's the same flame war over and over and over ad nauseum..
Please..For my sake..My fingers can't take the typing!

[tictac]

Take note that score by s7e9h3n...
done on everything at stock

2 x Barcelona @ 2GHz
Stepping B1
Single Channel Memory
at 5-5-7-15-21-1T

[AbelJemka]

the x1950xt with gddr4 beating the HD2900xt with gddr3 in overall games fps shows some heavy memory bandwidth bias, but that's a side point.

Even with GDDR3 2900XT has much more memory bandwidth than X1950XTX (100 Go/s vs 60 Go/s) so you plainly wrong

Telling 3D06 is biased because ATI cards hold world record is easy. The way this benchmark is coded favoured ATI that's all.

[cky2k6]

Strange this didn't get mentioned earlier. The K10's are the crippled versions, running 667 fsb and single channel. It's a start, and I applaud S7 for his effort, but there's a ways to go.

these results are perfectly fine for the clockspeed. you can't compare a desktop quad to server chips... 06 does not scale to 8 cores as well as it does from 2 to 4, and the ram is much slower on a server board. its comparable to the clovertown score, and it will score the same or higher most likely, when the bugs in the chipset are worked out (if you've actually read the thread this came from, you would've seen the problems i'm talking about)

just nobody listened... intel fans see what they want to see.

[BeardyMan]

Take note that score by s7e9h3n...
done on everything at stock

2 x Barcelona @ 2GHz
Stepping B1
Single Channel Memory
at 5-5-7-15-21-1T

Yeah but most people are selective readers, you beat them with numbers still they wil say a banana is a pine apple

[tictac]

Yeah but most people are selective readers, you beat them with numbers still they wil say a banana is a pine apple

LOL... yeah... anyway he got new bios from asus to test it on Asus L1n64-SLI board ... give him some times

[informal]

LOL... yeah... anyway he got new bios from asus to test it on Asus L1n64-SLI board ... give him some times

Yep,the "teaser" thread in AMD section became even more teasing now .Waiting for some news from Steven,if anyone can get it to work right,it's him

[EvlUndrWareNome]

a step, lets see some desktop platform Phenoms now! The wait is killing me!

[hardrail]

a step, lets see some desktop platform Phenoms now! The wait is killing me!

Agreed. I need a new system bad, and I want to wait for Phenom, but dang!!!