Bulldozer Preview - faked results (warning)

[Rofas]

Donanimhaber published a new Preview based on FX-8130P B1.







Source with Benchmarks: http://www.donanimhaber.com/islemci/...-sonuclari.htm

[PhotonPass]

Hmm, aren't these scores still almost needless because these are off ESs and not the final retail product? Or am I totally wrong in saying that?

[Rofas]

Originally Posted by PhotonPass

Hmm, aren't these scores still almost needless because these are off ESs and not the final retail product? Or am I totally wrong in saying that?

Maybe, maybe not. It's only a Preview. Not to forget that Bulldozer is coming with FX8150 instead FX8130P.

[muziqaz]

Here is my post from xs regarding this 'preview':

Memory subsystem is all over the place. Look at the latency of L2. double of that of my thuban.

And superpi again
At least it is kicking my thuban @3.9ghz by 3k points in cinebench 10
and loosing in PC mark to my 3.9ghz thuban by 100points
and about the same as 4ghz thuban at chess benchmark
and wins by 2fps in physics score on 3dmark11

[PorscheRacer14]

Well the memory scores on the mainboard memory, and L3 cache are good (as compared to my slightly OC'd 1055T) but the L1 and L2 cache are garbage.... That also could be why it needed a fix. Could be the SRAM structures needed a respin to get the performance up. Other than that, not much to see here.

[muziqaz]

3.6ghz 2600K gets 7100points in physics 3dmark 11.

[Evildead666]

The CPU-Z shot has 3.2GHz, and the AIDA64 has 4.2GHz.

That's a 1GHz Turbo.

One of the slides we saw a while back had something looking like the highest end FX processor would get 1GHz Turbo Boost.

[MTd2]

Bad news for you people. These benchmarks are probably fake:

"CPUz doesn't support BD in version 1.57.1 only in 1.58. So these so called BD benchmarks are probably fake."

http://www.xtremesystems.org/forums/...=1#post4900374

[PhotonPass]

Originally Posted by MTd2

Bad news for you people. These benchmarks are probably fake:

"CPUz doesn't support BD in version 1.57.1 only in 1.58. So these so called BD benchmarks are probably fake."

http://www.xtremesystems.org/forums/...=1#post4900374

Mmm, I am inclined to agree. Many of the benches that I have seen to me, are fakes and/or irrelevant (but fun to speculate about) because they can be 'shopped or that they are from ES's; neither of which represent the final retail boxed version of the product.

AMD, once again, are really masters at clamping down on information. We've seen it before with Northern Islands, we're seeing it now with Bulldozer, and when may most likely see it again next go around. I think someone mentioned that AMD should get some sort of reward or something for their secrecy.

PS: I kinda see AMD with the Internet's Troll Face, not in a negative light though. xD

[inf64]

Benchmarks are real,performance is fake . Pcmark 7 score is dead giveaway,something is borked in the bios or chip itself.It is scoring 3000pts while Llano 3850 gets 3340pts running at 2.9Ghz and no Turbo <=This is not going to happen.

edit: not to mention the dreaded super pi 1M score is 19+s while the amd-hater guy posted the score of 14.5s on his blog long time ago(3.2Ghz +Turbo). That is 31% difference in result.