AMD Zambezi news, info, fans !

[charged3800z24]

damn, nehalem at 4ghz does 31TPF :/ It looks like perf scales very badly with freq increase :/

I doubt The Folding client is ready for BD. It isn't going to do well.

[El Gappo]

Ln2 testing tomorrow, cascade tonight.

If anyone has the p-state checker I would be grateful...

[Oese]

any information if 4100/6100 can be unlocked to 8100 as of yet?

[Allubz]

AMD stated they can't be. Probably hardware locked somehow.

[Oese]

aha, ty

[muziqaz]

I doubt The Folding client is ready for BD. It isn't going to do well.

it won't ever be. Avx,fma4,xop are useless for current simulation methods

[doompc]

You mean power consumption when overclocked, right?

125W TDP models have 110W TDP for cores, 15W for the NB & external logic, while the readings for 95W TDP models are 80W for the cores and the same 15W for NB & external logic. So it ain´t that hot chip at stock.

Yeah, I know...
We are not at StockSystems.org thou.

L3 cache count for Core TDP on Sandy Bridge, and for NB logic on AMDs, is that correct?

[dess]

If anyone has the p-state checker I would be grateful...

Do you mean PSCheck? I've downloaded earlier 3.4.1.

[Mechanical Man]

Is there any results on CPU-NB scaling yet?

[muziqaz]

Is there any results on CPU-NB scaling yet?

There was/is in this forum. I seen someone done very good job with NB scaling and memory scaling. NB scaling did not give any increase of perf. Though everything was pretty much flat as far as I remember. Sorry no link, but I am sure it is somewhere in AMD section

[xVeinx]

any information if 4100/6100 can be unlocked to 8100 as of yet?

No, unlocks aren't possible (at least on current bios). I don't think they left that option open (especially with the modular arch, it wouldn't make sense anyway).

[Leeghoofd]

There was/is in this forum. I seen someone done very good job with NB scaling and memory scaling. NB scaling did not give any increase of perf. Though everything was pretty much flat as far as I remember. Sorry no link, but I am sure it is somewhere in AMD section

This is the weird part amongst reviews : I tested NB and RAM scaling and I had both giving decent consistent nrs. Though maybe my bad was that I had tested NB scaling with just 1600mhz C8 rams, Sin tested with 1866C9 and had less scaling then me, maybe at some point it all stops as the ram bandwith is far better then previous AMD generations. Issue with my NB tests was the unability to go over 2800 on air... I needed to go cold to pass it and to test 3200Mhz.

Ram wise up till 1866 from there it was still sort of scaling but not so much noticeable. I even had some slower results at 2133 then at 1866Mhz... but the average out of three gave a better overal result. Now it would be nice to test a MSI or GB board to compare efficiency vs the Asus board. Something feels not right....

[Mechanical Man]

This is the weird part amongst reviews : I tested NB and RAM scaling and I had both giving decent consistent nrs. Though maybe my bad was that I had tested NB scaling with just 1600mhz C8 rams, Sin tested with 1866C9 and had less scaling then me, maybe at some point it all stops as the ram bandwith is far better then previous AMD generations. Issue with my NB tests was the unability to go over 2800 on air... I needed to go cold to pass it and to test 3200Mhz.

Ram wise up till 1866 from there it was still sort of scaling but not so much noticeable. I even had some slower results at 2133 then at 1866Mhz... but the average out of three gave a better overal result. Now it would be nice to test a MSI or GB board to compare efficiency vs the Asus board. Something feels not right....

With P2 I think it was that nb stopped scaling after its 3*mem speed (base, eg. ddr3 2133 would need ~3,2GHz NB).

[bubik]

Hi Guys,
Not sure whether this was posted here before, in reference to the Linux patch for the IC aliasing, I found this bit about the the performance:

Borislav Petkov | 27 Jul 18:42
Re: [PATCH] x86, AMD: Correct F15h IC aliasing issue

On Wed, Jul 27, 2011 at 11:57:45AM -0400, Avi Kivity wrote:
> Out of curiosity, what's the performance impact if the workaround is
> not enabled?

Up to 3% for a CPU-intensive style benchmark, and it can vary highly in
a microbenchmark depending on workload and compiler.

http://permalink.gmane.org/gmane.linux.kernel/1172226

Does anyone know if the patch made it and it may be interesting running some tests on linux before and after?

[jazkat]

what you guys think about this? more BS or maybe its true? its bubiks news i just done some digging and found a link.

AMD Bulldozer Registry Fix?

AMD FX-8150 Passmark CPU Score has jumped from 8500 to 13000 using REG Patch. But it is not stable yet.

Do not listen to the critics and hardware sites that do a few generic benchmarks and come to a conclusion.

AMD FX-8150 Passmark CPU Score has jumped from 8500 to 13000 using REG Patch. But it is not stable yet.

The boys over at Kubuntu has been working on a AMD Bulldozer fix that sort-of applies across several platforms. The ARM Cortex-A9 8 core processor had a similar problem that was overcome with a software patch. The problem right now is that the patch has to be re-registered with each Windows start. Look for a 40% performance boost if this works…. more to come.

There is most definitely a Windows 7 AMD FX – software patch in the works. By most estimates the AMD Bulldozer FX is underperforming by 40-70% in most Windows 7 benchmarks.

LINKY: http://quinetiam.com/?p=2356

[zhadoom]

The news about this "miraculous" patch are spreading around many forums.

Waiting for some XS user to confirm .... if the patch exists and if it works.

[Dr. Vodka]

Well if that's true then it'd be the 40% speed it's desperately in need of to be a real bulldozer!

[PatRaceTin]

? magical patch?

[EniGmA1987]

Seems kinda doubtful to me, maybe in a few specific situations it would work but I dont think it is suddenly going to make BD perform a lot better on average

[B NEGATIVE]

The news about this "miraculous" patch are spreading around many forums.

Waiting for some XS user to confirm .... if the patch exists and if it works.

Lies i think.....
http://kubuntuforums.net/forums/index.php?topic=3118749

Which is a shame....because

[Chrono Detector]

So there is hope for Bulldozer after all? It needs a miracle patch to show its real performance and optimization? Hope so.

[Apokalipse]

The news about this "miraculous" patch are spreading around many forums.

"miraculous", "magic" - silly words thrown around by people.
It's more like removal of crippling performance penalties, to make it perform like it's supposed to.

[Leeghoofd]

There is most definitely a Windows 7 AMD FX – software patch in the works. By most estimates the AMD Bulldozer FX is underperforming by 40-70% in most Windows 7 benchmarks.

LINKY: http://quinetiam.com/?p=2356

So it works far far far better in Windows 8 ? If you can get performance up by 10-20% that would be excellent, 40-70% is utopia... then it would be close as being the fastest CPU out there.... IF AMD knew this then they rather had waited a few more weeks for this "patch" Same for the CH V is too slow discussion, if I compare my stock nrs with sites that have used other than asus boards I see not much discrepancies...

[muziqaz]

I also think that AMD would have waited for that magical patch or even pushed for it to be released sooner. And regarding CH V, i suppose AMD had a deal with ASUS that AMD provides cpus for reviewers and ASUS provides motherboards. They couldn't change the motherboards because of obligations. And most of review sites just took that system and tested just that system, while some others tried other motherboards.
I myself will be getting MSI. Its my 3rd MSI in a row (790FX, 890FXA) even though I was planning on getting asus.

40-70%
Man imagine that? Suddenly a wild bulldozer appears, Sandy bridges not very effective.
We can dream, right?

[Apokalipse]

So it works far far far better in Windows 8 ? If you can get performance up by 10-20% that would be excellent, 40-70% is utopia... then it would be close as being the fastest CPU out there.... IF AMD knew this then they rather had waited a few more weeks for this "patch" Same for the CH V is too slow discussion, if I compare my stock nrs with sites that have used other than asus boards I see not much discrepancies...

Personally, I wouldn't expect 40-70%, though I would like to be wrong about that.
10-20% would be nice though, and I could see that happening.