5.22.2007

Unreleased Barcelona will be DOA

From ZDNET:
The company ran a demo comparing the performance of two four-socket servers, one using the quad-core Barcelona chip and one using a dual-core Opteron chip. The demo measured the performance of the chips on an imaging benchmark called POV-Ray, ...
The quad-core chip processed about 4,000 pixels per second in rendering the image, while the dual-core chip could only hit around 2,000 pixels per second. Allen said Barcelona was not running at the fastest clock speed that will be available at launch, although he declined to specify the speeds that will be available... AMD did not test Barcelona's performance against one of Intel's quad-core Clovertown Xeon processors, but Allen said Barcelona "will be the highest-performing x86 chip out there. It will blow away Clovertown."

Remember the number: POV-Ray at 4000 for a 16-core 4S Barcelona config (4Sx4core).

To compare that number, check this Intel V8 Review: http://www.xbitlabs.com/articles/cpu/display/intel-v8.html
Intel's 8-core Clovertown scores 4677 p/s in POV-Ray.

UPDATE: From Uberpulse: "Why do you need 16 cores, when you can do better with 8. Our 8 core system is 30% faster than the 16 core machine AMD showed to the press yesterday. I just don't understand how they can claim to be 40% faster", said Francois Piednoel, an Intel engineer present at the show.

It takes Intel half the number of cores to beat Barcelona! If this doesn't spell disaster for AMD, I'm not sure what. Forget about competing with Intel. I just hope that AMD releases Barcelona soon enough to at earn at least something to pay for the bancruptcy fees.

6 comments:

Anonymous said...

I find it hard to imagine that AMD would embarass itself by showcasing a benchmark where it is currently beaten by Intel.

You'd think that AMD would want to be perceived as being credible, no? Why they did not compare Barcelona against against Clovertown is obvious.

Yes, Barcelona "will be the highest-performing x86 chip out there. It will blow away Clovertown."

Randy Allen failed to give a timeframe for when and what variant of the Barcelona core would blow Clovertown away.

Does anyone want to take a guess? On POVRay? Or on SPEC2006_Rate, Or on whatever it is this week.

AMD BK Q2'08 :)

Anonymous said...

One thing to be careful of is that there was no mention of speed - whle this won't double the performance again, I find it hard to believe this was a high clocked part. The reason why I believe this is AMD stated that the 4 core 4P solution was about 2X better than the 2core K8 2P solution.

As you should get almost a doubling from the cores to begin with either AMD is way off about the delta between K8 and K10 performance or the K10 chip was lower clocked (which is likely as they also indicated power was the same).

So for the higher end parts (>65W) it should be better.

Also I know very little about POV - are the results between the sbit review and AMD numbers comparable (at least as a rough estimate)? Not sure how much the setup, or the actual image matters...

That said, the amazing lack of information doe s lead one to wonder, WTF? Forget the stupid Osbourne effect and don't want Intel to copy arguments - if K10 was clearly superior, AMD would be releasing benchmarks by now or they would be "accidently" leaked by partners.

My guess (just a guess) is K10 will be superior in FP but INT will be mixed - AMD will tout the FP applications so they can say all of their claims were right (much like they will claim the mid-2007 schedule has been met?), but the press will look at the INT #'s and give it a mixed review - thus by delaying the benchmarks, and dribbling them out, AMD can delay any potential mix/bad reviews.

Or maybe they're waiting for the Penryn reviews to come out first? :)

Anonymous said...

Randy Allen failed to give a timeframe for when and what variant of the Barcelona core would blow Clovertown away.

amd will, in 2009 when intel has QC nehalem out and AMDs high-end Barcelona i competing against intel's low-end 1P clovertown.

Unknown said...

I suspect that K10's true potential of being a native quad core has yet to be shown. As I already explained in another post, it is unusual for AMD to show a benchmark of a product that is slower than its competitor's current offering. Randy Allen's commend of "Barcelona will blow Clovertown away" is not supported by this benchmark, at all.
Something really doesn't add up.

Anonymous said...

Randy Allen failed to give a timeframe for when and what variant of the Barcelona core would blow Clovertown away.

This guy also said that at AMD they did not believe they needed to go and build more FABs

They're probably right on that one because at the rate they are going they're not going to need 'em...

Anonymous said...

'nuff said

http://www.uberpulse.com/us/2007/05/amds_barcelona_lack_of_performance_still_unknown.php

http://www.uberpulse.com/us/2007/05/intel_responds_to_amd_barcelona_has_a_problem_you_better_fix_it_before_you_ship.php