AMD did another sealed demo of Barcelona where the press couldn't inspect the systems. They ran POV-Ray and Barcelona was able to do 4000 pixels.
An anonymous reader sent in this link (thanks) showing how existing Intel Xeon's can best that.
Link to Yahoo message board
Barcelona can't even compete with Intel's existing products. It will be history when Intel introduces Penryn. AMD will bleed money like water this year. If Barcelona can't give Penryn a fight, AMD could lose another billion $s through the rest of the year. I'm still waiting for proper Barcelona benchmarks so that I can make my forecast.
Subscribe to:
Post Comments (Atom)
22 comments:
Man have you gone mad. Times like these call for you to rise above Sharikou and Scientia!
Did you read the link - or have any unbiased data that's different.
...and why should only they have all the fun
-:)
You lose credibility when you mimic those fools. Come on I expect better.
please don't turn into a literal opposite of Sharibob. your blog is pretty good.. well written, lots of facts. but if you start writing over the top, "pervasive journal" sounding articles like this one your credibility is going to go downhill.
Anonymous said...
please don't turn into a literal opposite of Sharibob. your blog is pretty good.. well written, lots of facts. but if you start writing over the top, "pervasive journal" sounding articles like this one your credibility is going to go downhill.
actually there is nothing wrong with 180's content in this post, but the title:)
anyway, personally i think Barcelona would have its unique advantage in MP judging base on its architecture details; but just like what other said, it's too little, too late.
it has to compete with intel's coming 45 nm cpu, which promised higher clock, lower power and some architecture enhancement.
Wasn't the point of this to allow Sharikou to show both sides of his personality? Is this blog being written by some other person?
Hi, can you explain why you were upset about fish a couple of articles ago?
Actually, times like these call for all of us to buy whatever matches our budget instead of worrying about speed, cause there's plenty of speed even in a $200 WalMart sweatshop PC. But we like to type at 15 hunt-and-peck-words-a-minute on a PC capable of more freaky frames a second.
lol whats up with all that anonymous.
180 is not pulling this numbers out of the air, various sources confirm this.
But imho i think thers something wrong with the amd system... it would be insane of amd to call this a new product.
btw. you have a typo on your article, its says 400 p/s. ;)
it should be 4000.
I just want to point out that we've got 10 comments on this article simply because it's contentious as opposed to many of the other well researched and probably more intelligent posts I've written in some time.
Check this out
http://www.uberpulse.com/us/2007/05/intel_responds_to_amd_barcelona_has_a_problem_you_better_fix_it_before_you_ship.php
Its all over the news. The buzz is out!
Quote from the link
"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.
I guess we were not the only ones questioning this move by AMD
http://www.uberpulse.com/us/2007/05/intel_responds_to_amd_barcelona_has_a_problem_you_better_fix_it_before_you_ship.php
I agree with the anonymous who claims 180 and 0 are the same person. This is nothing but a Jekyll and Hyde deal. Admit you're the doctor, man.
180 Sharikou said...
I just want to point out that we've got 10 comments on this article simply because it's contentious as opposed to many of the other well researched and probably more intelligent posts I've written in some time.
BINGO!... you're on to something here... Scientia and Sharikou gets plenty of posts... why is that? hmmm ;)
Uberpulse has an updated article:
http://www.uberpulse.com/us/2007/05/amds_barcelona_lack_of_performance_still_unknown.php
Roborat, Ph. D. said...
180 Sharikou said...
I just want to point out that we've got 10 comments on this article simply because it's contentious as opposed to many of the other well researched and probably more intelligent posts I've written in some time.
BINGO!... you're on to something here... Scientia and Sharikou gets plenty of posts... why is that? hmmm ;)
actually no matter how you put it, for the hits, first and foremost, you need to attract the AMD fan(bois) to this blog. Then only you can start to have stupids comments. Then the Intel fans would start to laugh at it and tease them with facts, or make fun of them using parody comments. Then the AMD fan(bois) would fight back with even stupider comments, and the cycle goes on :).
So, ability to attract AMD fanbois is key. Or you can act as Intel/AMD fanbois to post stupid post, which i do not think that is what you want to do :)
btw, just to point out the facts (posted by enumae in the other blog), current C2d actually is already beating the K8 by big margin in the POV-Ray test by 40%.
http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070402-00723.html
http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070319-00685.html
that might explain the result. But again, seems like Barcelona makes no big improvement in this bench, and thus i'm wondering on their floating point gain claim.
Another thing is that it shows how bad their marketing people is, choosing a very wrong FIRST bench demo :)
Naturally, everyone jumps on this unrelated demo to claim that Intel is still faster. The Pov-ray demo isn't using SSE. The only reason this demo is being run is to show fairly good scaling for a drop-in replacement. Keep wringing though and maybe you can get some water out of that stone.
I don't think I get a lot of fanboy comments on my blog or if I do then they tend to get corrected pretty quickly. I would say the main reason I have traffic is that there are often very involved discussions.
Honestly, I don't get the Intel BK in Q2 posts like Sharikou gets or the other nonsense posts that claim that Intel cpu's are junk or overpriced. However, if it makes you feel good to pretend that I do, knock yourself out.
The main difference between my blog and sharikou's, roborat's, and this blog is that I don't cherry pick news items either for or against AMD. To me, cherry picking is a waste of time.
Scientia - your cherry picking comment is mildly amusing in light of the fact your own latest post completely ignores the fact that AMD lost 5 points of market share to Intel in Q1.
If there were some good news to report out of AMD, I'd call it. But there ain't. I predicted a reversal of fortune way back in Q306. Yup - that's what I do here. Analyze and make predictions. So far...doing well on that front.
Until I see Barcelona independant benchmarks and a concrete schedule, I have nothing good to say about AMD. Even their cost cutting is too little. Hector needs to stop looking at the world with 30% market share stars in his eyes and consolidate his position so he can live to fight another day. Barcelona is not going to be the same knockout punch that Opteron was. He needs to re-strategize.
And the part that gets me the most. For goodness sakes, stop whinging like a baby. When Hector was winning market share through 2005-2006, he was out there talking it up like a king among lions. Now all he seems to do is complain about how Intel picks on him in the schoolyard. Same for Dirk and Henri. Get over it...and get on with the job dude. There are people who are losing their livelihoods because of your decisions. I said it for Otellini and I say it for Hector - get competent or get out of the way.
Otellini seems to have got his stuff together. Now Hector needs to do the same.
180
"Scientia - your cherry picking comment is mildly amusing in light of the fact your own latest post completely ignores the fact that AMD lost 5 points of market share to Intel in Q1."
I was ignoring it when I mentioned a 30% drop in volume for AMD in Q1 and mentioned that Intel had no similar drop? Curious perception.
1. The main difference between my blog and sharikou's, roborat's, and this blog is that I don't cherry pick news items either for or against AMD. To me, cherry picking is a waste of time.
LOL
2. How is 82% good scaling with twice the cores, and a new one at that?
Scientia - first, AMD suffered a 38% drop in processor revenue. Which does not necessarily equal a 30% drop in processor shipments.
Second - if you wanted the answer to this anomaly, you should have just asked. AMD stuffed their distribution system with too much product end Q406 - probably to try and meet their revenue numbers. Until those parts work their way out of distribution/channel inventory, their volume will be lower than normal. Couple that with huge share loss and you have your answer. Read my post below - hopefully that should answer your questions:
http://sharikou180.blogspot.com/2007/04/amd-q1.html
Post a Comment