Wednesday, January 24th 2007
Details on AMD’s Barcelona
Although AMD's next generation of server processing hasn't been a complete secret, here are some of the latest details about it. Codenamed "Barcelona", the processor is likely to go head-to-head with Intel's Xeon 5300, both being quad-core processors. However, one key feature of the Barcelona core is that (unlike the 5300) it will natively be a quad-core CPU, not two dual-cores built into one unit. The chip will continue the advance into 64-bit computing, and like the rest of AMD's 64-bit processors it has a memory controller build into it, another difference between this and the Xeon 5300. Despite much higher performance than older models, the processor will still only use 95 watts or 68 watts of power depending on the different designs, which apparently gives a 60% performance increase per-watt over the older generation Opterons. It can also support up to 256TB of memory, now that the memory controller supports full 48-bit addressing. AMD's benchmarks suggest that it performs much better than the Xeon 5300 due to fewer bottlenecks, but this could obviously be biased against Intel.
Source:
internetnews.com
24 Comments on Details on AMD’s Barcelona
Barcelona is scheduled for Q2 around june.
but as others have already stated, intel isn't just gonna sit there.
P.S. Didnt AMD already have a codenamed proc called "Barcelona"?
The die (~290mm^2) should be about a quarter larger than Windsor Rev.F and almost 1.5 times the size of Toledo Rev.E DCs. Unfortunately the manufacturing cost will be more than just a quarter higher compared to Rev.F DCs...
Time say good bye to cheap AMD chips.
It maybe expen$ive, but it sure is pretty:
That's 2.4x fold the area of a 65nm Brisbane DC.
and i want one :D
HTT provides the I/O necessary to scale much more effectivly.
Perhaps a dual quad core Xeon will be very very competative, but when it comes to actual individual CPU's, the xeon is toast at this point in time.
Core != CPU count.
This is a very blurry line these days.. and everyone has to be careful in statements :o
2.4x die size... if thats true, the K8L is gonna be a beast. It is important to note the xeons get a nifty dual dual fsb trick with FBDIMMS I believe, which greatly helps intel.
Its no longer a normal classic FSB. Whereas an A64 is still utilizing a the same dual channel FSB base.(Although each CPU does get its own)
This is where intels scaling fails... if you have an overwhelming amount of processors, CPU->memory is able to be handled, but what if processor #320 needs to talk to processor #102, get 400 of the CPU's in a cluster doing this, Intel's fsb is going to fall flat on its face.
then again, this is completely useless from a consumers standpoint. Unless ofcourse, you have 300 keyboards to transfer keystrokes.
In addition, he's somehow affiliated with AMD because he has access to ES CPUs and other inside material which as you know, are quite rarely seen outside AMD's official companions.
And I'm willing to wager a cookie he will be one of the first on the net to ever benchmark a K8L...
I'll let the pictures speak before you get bored ;) with my rambling:
But, theres no reason to see huge price hikes, because the clawhammer die was gigantic too.
Intel loves charging out the wazoo (see. EE)
AMD has relied on being a price competitor, and if thats a dual core, thats got.......... a lot of cache.
Otherwise, quad cores having a die size similar that is TBE, and is not news.
Yeah, working and all I dont remember the codenames atm :P
edit: lol helps to RTFA.
Its quad core. Big news? did you expect it to get smaller or something? 2.4x seems to be accurate, because of the inclusion of extra xbar circuitry and memory controllers logic to accomodate the extra core(s).