SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Advanced Micro Devices - Moderated (AMD)
AMD 237.62-7.2%Nov 6 3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Sarmad Y. Hermiz who wrote (229938)4/15/2007 1:15:18 AM
From: PetzRead Replies (2) of 275872
 
Do you consider it a "nit pick point" that the highest SPECfp2006_rate score for 2 chip systems is held by a lowly 2.1 GHz dual core IBM Power5+ system, not by a pair of quad-core Xeons running at a much higher clock speed?

System Clock # Chips CPU Model Cores/Chip Result
Bull Escala PL450R+ 2100 MHz 2 IBM Power5+ 2 62.5
PRIMERGY RX300 S3 2666 MHz 2 Intel Xeon X5355 4 60.2
Tyan n4250QE(S4985) 3000 MHz 2 Opteron 2222SE 2 52.1
Dell PowerEdge 2970 2400 MHz 2 Opteron 2216 2 40.7*

* - base score
So despite having twice as many cores, Intel loses outright to IBM and only beats a dual core AMD system by 16%.

So 6 months from now, the tables will be more than reversed. Not just by AMD having twice as many cores as now, but each one of them will have twice the FPU's, twice the cache and twice the cache bandwidth of the current Opteron. I expect a 2-chip Barcelona system at 2.3 GHz will have a SPECfp2006_rate of over 120, making the Xeons look like a joke.

Even in SPECint2006_rate, Intel has only a 1.5:1 lead. Doubling cores, cache and bandwidth will obliterate that even with lower clock speed.

My prediction is that Intel will start losing server share in Q4.

Petz
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext