looks sexy! :)Quote:
Originally Posted by tictac
Printable View
looks sexy! :)Quote:
Originally Posted by tictac
There is still something, I don't understand... What is Budapest?
Is it a specific version for cray? Or is it the desktop version of the barcelona?
Quote:
Barcelona is a strategically important chip for AMD, as it tries to regain market share lost to Intel's newer Core 2-based Xeon server processors, the dual-core 5100 and the quad-core 5300.
However, Budapest shipments to Cray slipped. "Specific to Cray, we did have a change in our schedule, as far as when Cray gets Budapest parts," Hughes said, declining to share further specifics.
Barcelona is designed for systems with two or more processor sockets, but Budapest is for single-socket systems.
Budapest is single socket with HT3.
it's opteron K10 for AM2+ with HT3. Budapest is used by cray ;)Quote:
Originally Posted by nemrod
I still find it wierd that Cray want to buy single socket Opterons and not MP sockets.
maybe budapest is mutli P able, and cray want HT3 :confused:Quote:
Originally Posted by Shintai
And the barcelona hasn't HT3?
Why MP are ready and single socket not? I would have think that MP processors are more difficult to realize?
They are same ;)Quote:
Originally Posted by nemrod
Hint as of when NDA on Barcelona will be lifted!
SourceQuote:
At the CTO Summit, AMD laid out a fully finished 300mm wafer to touch and photograph, and we were shown specifics on the fab rollout schedule for AMD’s 45nm process. We got an advance look at new AMD/ATI chip-set technology as well, along with a completely new and radical AMD CPU, the nondisclosures on which lift this month. AMD is not running behind Intel. It is simply not practicing reactive engineering, and if you pay attention, you’ll see that AMD’s take on 45nm process, 300mm wafers, desktop chip sets, and dual-core mobile architecture are more than mere snapshots of the marketed leading edge, which is a coat of gloss on the present. AMD, through its partnership with IBM, defines the leading edge. Watch.
nice :D
This is "May 16, 2007"... "the nondisclosures on which lift this month"...
It's a joke... There is no "hint as of when NDA on Barcelona will be lifted!"
Quote:
Originally Posted by Lightman
...
Did you look at the date of the article? May 16, 2007.
So far, all AMD showed was how pathetic K10's FP performance improved over K8.
This is kind of a continued discussion from this thread.
http://www.xtremesystems.org/forums/...d.php?t=147675
But man, if that article is true, then we (desktop users) won't see K10 until next year. Hopefully AMD can survive this with the revenue from the ATi chipset and video card sales.
Perkam closes this thread:
http://www.xtremesystems.org/forums/...d.php?t=147675
But if the first post was biased, the new article from Anandtech is interesting:Quote:
Originally Posted by perkam
http://www.anandtech.com/tradeshows/showdoc.aspx?i=3006
Obviously, there is still same kind of information about AMD not fully ready but there is too informations like
And this could a little bit invalidate the "quick and dirty benchmarks" result on cinebench.Quote:
We know that Barcelona works and runs benchmarks, as we saw back at AMD in May. But the demos that AMD ran were on its own motherboards, not on motherboards from its partners. AMD's partners just recently received their first Barcelona samples, and as expected, the boards require some heavy BIOS work before the new chips will even work.
The motherboard we tested on had minimal HT functionality and wouldn't run at memory speeds faster than DDR2-667; most video cards wouldn't even work in the motherboard. Memory performance was just atrocious on the system, but the motherboard manufacturer we were working with attributed this to a BIOS issue that it expects to be fixed in the very near future.
On an other hand, this give some questions about the "simply plug in AM2..."
Yeah but if it's just BIOS problems, that is no surprise right? I guess I assumed that AM2+ boards would still need BIOS update to run the new CPUs.Quote:
Originally Posted by nemrod
Big surprise a New CPU class requires new Bios..
The surprise is that apparently nobody was able to do a correct bios for those buggy devices... Memory issue? Is it really simply a bios problem? Where is the controler in the bios perhaps? Where are the bench on the AMD itself motherboard? A yes 4000 for the barcelona with 16 core when 8 K8 do 2000 and 8 core 2 do 5000... Play with that...Quote:
Originally Posted by nn_step
If it's a backward compatibility issue, why doesn't AMD just skip AM2+ and go to AM3. Then the motherboards can be designed to meet the cpu's needs.
http://www.anandtech.com/tradeshows/...spx?i=3006&p=5
Just in case anyone missed it, Anandtech..
Quote:
Oh yeah, after scouring several computer shops in the greater Taipei area we happened to find a pair of Barcelona chips available. Okay, maybe we broke into an engineering lab and ran off in the middle of the night with them but hey, nobody caught us. (Ed: and you thought all that time playing Ghost Recon was useless....) Once we get proper motherboard/BIOS support, we'll be ready to actually discuss performance.
I think AMD's very wise not to switch to DDR3 that soon.Quote:
Originally Posted by awdrifter
Wanna have AM2+ and a pair of DDR2 Rams!
Did you read about the new B2 Stepping? Think I heard that whole BS already about B0, but even fudzilla seems to become more pessimistic (or realistic?).
Because the "drop in upgrade" thing is a big marketing point for their server clients... and this is where they make the big bucks ;)Quote:
Originally Posted by awdrifter
A secret revision
SourceQuote:
Our sources have confirmed that a few Barcelona chips that the company demonstrated at Computex behind the closed doors are B0 revision.
We also learned that B0 can go faster than 1.6 GHz but that it was a very buggy revision. Now our sources are talking about B2 revision that is suppose to save the day and this one is expected any day now.
The same people also indicated that Barcelona will either be the best thing that even happened to AMD or it will be doomed to be the next Netburst. Unfortunately it looks more like the later one, at least from this perspective.
But that just meant it has to be compatible with socket F right? Not necessarily AM2.Quote:
Originally Posted by Sumanji
I think, the problem is the branch prediction, which didn't improve that much compared to other parts of the CPU. I measured 15% mispredicted branches for the povray benchmark on my X2. Standard code usually has ~5%. Then there is a branch about every 8 instructions. That's normal for a raytracer, which has to do a few calculations to find out, what to do next.Quote:
Originally Posted by nn_step
MSVC 8 also wasn't able to vectorize any calculations. So there are practically no 128 bit SSE instructions, which could make use of K10's improved FPU.
According to Prime95 benchmark, Willamette was much faster than Athlon and Northwood faster by 30% or more than Athlon 64.
All K10 chips will be compatible with current desktop/server platforms.Quote:
Originally Posted by awdrifter