InvestorsHub Logo
icon url

chipguy

09/10/07 10:27 AM

#47744 RE: alan81 #47741

So what does it mean...
When they ship a 2.0Ghz product, but publish benchmarks on a 2.5Ghz product... and compare them to the competitors 2.33Ghz product... when the competitor has been shipping a 3.0Ghz product for a while already?


It means AMD is a desperate corned animal with nothing to
lose by such sleazy tactics.

BTW, this is an interesting read:

http://www.realworldtech.com/forums/index.cfm?action=detail&id=82684&threadid=82680&room....

Don't agree with anandtech's specjbb2005 results

Two flaws IMHO
1) They used an old version of the Sun JVM, from before Sun adopted Intel HW. That old version contains optimizations for AMD but not for Intel. A more recent Sun JVM (1.6.0_02) or a recent version of JRockit would have resulted in much better Xeon scores. While they had a valid reason for going with 1.5.0_08 (be able to compare with old results) they should have mentioned this fact in the article.

2) They worked around the main AMD weakness (poor performance if accessing non-local memory) by running with several JVM instances, each affinitized to a NUMA node. Xeon does benefit from multiple instances too, but not nearly as much as AMD does. For completeness, they should have run in a single-JVM config too, or without using numactl.

-- Henrik


Somewhere else I read that this test was run with prefetch
turned off. IIRC a while back AMD submitted SPECJBB2005
results for Conroe with prefetching turned off.

I wonder if AMD provided some "technical assistance" to
Anandtech's Barc review?



icon url

wbmw

09/10/07 11:06 AM

#47765 RE: alan81 #47741

Re: So what does it mean...
When they ship a 2.0Ghz product, but publish benchmarks on a 2.5Ghz product... and compare them to the competitors 2.33Ghz product... when the competitor has been shipping a 3.0Ghz product for a while already?


It means they signed an NDA that allows AMD to review and alter their report before they publish.
icon url

Windsock

09/10/07 11:53 AM

#47786 RE: alan81 #47741

So what does it mean... When they ship a 2.0Ghz product, but publish benchmarks on a 2.5Ghz product...

A little bird, who many of us remember, privately predicted the AMD con job with the 2.5 unobtainiums.

"I predict that AMD is sending special complete systems to their suckups with 2.5 to 3 GHz Barforamas - as well as the "standard" 2 GHz slugs.

<snip>

Basic cheap magician stuff - look at what we are waving in front of your eyes (and don't look at the crap we are actually manufacturing and shipping)."




icon url

Snowrider2

09/10/07 12:41 PM

#47797 RE: alan81 #47741

To me, it means the glory days of Opteron/Athlon were an anomaly in the 30+ year history of competition between INTC and AMD. Looks like AMD is returning to the pre Opteron/Athlon days. Just my $.02.

Snow