AMD Zambezi news, info, fans !

[[XC] Oj101]

If it's fake it's the most accurate one to date

[Apokalipse]

Pretty nifty stuff... Is it me or, maybe with the exception of HyperThreading, Intel not do much interesting designing stuff like that, instead concentrating on making stuff that "has more torque" (if you will)?

I'd say Intel's design works in a "brute force" way, just pouring tons of resources into it.
Whereas AMD relies on being clever and more efficient with the resources they have.

But I agree on what you have said that comparing an 8 core AMD vs. 8 core Intel isn't "fair" not cause AMD took the module route. I never compare products like that, I compare it they way you should which provides the best bang for my buck?

Having the modules design allows more cores per die area.
Didn't someone say 8 core BD was about 1% larger than 4 core SB?
Which would mean Intel is forced to compete on a smaller node.

[Smartidiot89]

I'd say Intel's design works in a "brute force" way, just pouring tons of resources into it.
Whereas AMD relies on being clever and more efficient with the resources they have.
Having the modules design allows more cores per die area.
Didn't someone say 8 core BD was about 1% larger than 4 core SB?
Which would mean Intel is forced to compete on a smaller node.

1 BD module(2 cores) is about 19.4mm2 while 1 SB core is 18.4mm2, both excluding cache. So as long as Bulldozer performs I would say AMD has a lead over Intel so that they have to compete with a smaller node as you say

[2good4you]

AMD is REALLY holding it all back. No leaks at all.

[Apokalipse]

Check out this totally legitimate screenshot:

Chuck Norris was nice enough to design it himself

[FlanK3r]

hehe, n1

[hieuhef]

Check out this totally legitimate screenshot:

Chuck Norris was nice enough to design it himself

Hilarious, put a smile on my face this morning.

[Manicdan]

1 BD module(2 cores) is about 19.4mm2 while 1 SB core is 18.4mm2, both excluding cache. So as long as Bulldozer performs I would say AMD has a lead over Intel so that they have to compete with a smaller node as you say

on the contrary thought, all that space between modules does greatly factor into $/mm2 since the total size of the chip is larger. not sure what that extra space is doing though, so i have no idea why its important, needed, or wasted.

[Smartidiot89]

on the contrary thought, all that space between modules does greatly factor into $/mm2 since the total size of the chip is larger. not sure what that extra space is doing though, so i have no idea why its important, needed, or wasted.

An 8-core BD shouldn't be much bigger then 20% compared to SB 4-core, and is most likely smaller than Intel's coming 6-core.

Just a very good hunch I have looking at this un-photoshoped die

[Manicdan]

is that the one from back in january-ish?

if i remember that right, if we just add cores + cache it came to like 250-275mm2, but its really going to be like 325mm2+

[Smartidiot89]

is that the one from back in january-ish?

if i remember that right, if we just add cores + cache it came to like 250-275mm2, but its really going to be like 325mm2+

No this is an almost full dieshot, and it should be approx. 280-290mm2.

[Formula350]

mabye only one reason, could not be fake is: it was CPU-Z 1.56 or 1.57 and sample was in AM3 board (not AM3+). But from my side, I thinking, its fake . We can only waiitng for relase date

We discussed that on the last page, that it's fake. :\

So I was digging into making that photoshopped CPU-Z image to show 2good, when I noticed something about the ORIGINAL... Wouldn't the "Package" line be blue like the rest of the labeled of filled boxes? Yet, it shows Gray, as if it was originally empty...

That image is fake, the word would not be gray if the line is filled.
Its a really nice photoshop, but then again I could change the frequency to 3400 on my last photoshopped image and then proclaim its true.

(still in response to Flanker) Like I mentioned, creating that isn't hard, I'll go ahead and finish mine just to show that point. That's probably why they blocked over the L3 and version part, was more an amature job. Not that it was bad (or that I'm not an amature), but what sells it is how they went about presenting it!

An 8-core BD shouldn't be much bigger then 20% compared to SB 4-core, and is most likely smaller than Intel's coming 6-core.

AMD did it before, with the K7 and then again with the K8, beating Intel with more design thought vs the 'brute force' or 'pure horse power' method. I think P4 is a great example of that. Anyways, I think it is definitely time for AMD to shine again, and will be more impressive (and a good "gotcha!") if they achieve more performance with the smaller die size (OK, not 'die size', but per-core size)

I think Intel's biggest 'oops' with designing was actually ditching the development of their improved graphics core Their current GPUs are obviously a lot better than their past iterations, but it's quite lacking compared to the AMD approach. The E350 might only have an HD5450 (if not completely, damn near), which performance-wise fits amazingly well with the E350, so I'm expecting Llano to really put the hurt on the i3; maybe even i5!

[FlanK3r]

yes, but it was repost again as image....So we talking again about it

[AKM]

AMD did it before, with the K7 and then again with the K8, beating Intel with more design thought vs the 'brute force' or 'pure horse power' method. I think P4 is a great example of that. Anyways, I think it is definitely time for AMD to shine again, and will be more impressive (and a good "gotcha!") if they achieve more performance with the smaller die size (OK, not 'die size', but per-core size)

Yeah, there are many signs that Bulldozer will be an impressive chip and they've been working on it for what, half a decade?

[ReKcOlNu]

I was also under the impression current Bobcats were K10.5, but now I'm seeing it be referred to as K14. CPU-Z is in line with that, though strangely Windows7 is not, showing actually a family number of 20 for the E350 >_> lol That might be because of the Stars and in that time it's gotten corrected *shrug* It's hard when the manufacture isn't giving any info heh Not that I don't blame them though, just sucks for us lol

Those numbers are in Hex , 14 hex = 20 dec.
i.e. Phenom Family 16 (decimal 16 = hex 10)

Windows registry, Identifier string...
HKEY_LOCAL_MACHINE\HARDWARE\DESCRIPTION\System\Cen tralProcessor\0

[Formula350]

Yeah, there are many signs that Bulldozer will be an impressive chip and they've been working on it for what, half a decade?

haha Yea, it's been awhile that's for sure. I'm honestly curious as to how much is original and how much has essentially been scrapped in favor of new architecture (though not necessarily from the ground up)... Would be quite interesting.

Those numbers are in Hex , 14 hex = 20 dec.
i.e. Phenom Family 16 (decimal 16 = hex 10)

Well, ya learn something new everyday lol

Anyone know why Llano is being designated "12", when 'Dozer and Bobcat are 15 and 14, respectfully?

[FlanK3r]

why not? Llano is not K10 and not is K15...

[hieuhef]

I've wanted to contribute to this thread since I started reading, but had nothing to add. Hope this is as intriguing/comforting to others at it was to me:

AMD release schedule

[Darakian]

9 series chipset is due out on June first then. About time

[Formula350]

why not? Llano is not K10 and not is K15...

... Because it's Bobcat with up to a HD5670 integrated, and an E350 (Bobcat) is K14? >_> That's why I put that info in the previous post lol

[Halk]

There's a Fudzilla article today, . I don't think I'm going out too far on a limb by saying...

990FXA-GD80 and 990FXA-GD65 are the same chipset, the 990FX and have the same number of PCIe lanes available on the chipset, so the 4 x x16 slots on the first board will be x8/x8/x8/x8 or some combo, and the second board will be x16/x16. That means little or no difference from the config of motherboards that we have now. Although perhaps they'll try to distance the x16 slots a bit more to accomodate triple width cards which are becoming more popular.

I don't think we'll learn much from distributors listings of the boards etc until reviewers get the boards with CPUs as it looks like much of the chipset features will be power management related?

[hieuhef]

I'd imagine it would be very similar to the current AM3 GD70 and GD65, so just a few options swapped out, but largely the similar.

[FlanK3r]

... Because it's Bobcat with up to a HD5670 integrated, and an E350 (Bobcat) is K14? >_> That's why I put that info in the previous post lol

I dont think so....Bobcat is still new architecture as new is Bulldozer. K12 Llano is redesign of Propus die (so K10.5).

[hieuhef]

Leaked slides via Techreport:

http://techreport.com/discussions.x/20886

Original site translated via Google:

http://translate.google.com/translat...-sonuclari.htm

Well, some things definitely stick out, like the 2nd image showing under BD/FX: DX11 and Eyefinity. Thoughts? I mean, unless they're packaging a 6670 with every Zambezi chip sold, do the 3dmark scores hold? Bizarre to see.

[BeepBeep2]

Leaked slides via Techreport:

http://techreport.com/discussions.x/20886

Original site translated via Google:

http://translate.google.com/translat...-sonuclari.htm

Well, some things definitely stick out, like the 2nd image showing under BD/FX: DX11 and Eyefinity. Thoughts? I mean, unless they're packaging a 6670 with every Zambezi chip sold, do the 3dmark scores hold? Bizarre to see.

Creative ways to make engineering look better by the marketing team, thats all. It seems the hope is that you dont scour the chart, just glance at it.