Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Dan3 who wrote (119164)	11/25/2000 2:41:28 PM
From: Ali Chen	Read Replies (2) \| Respond to of 186894

Dan, <the L2's must maintain coherency>and<slide 15> I have no knowledge about SQL Server behavior and don't care how PPro-1995 16k cache behaves on this workload. The question was about SPECfp95 (and SPECfp2000 to the same extent). Alphas, HPs, and SGIs all have caches that must maintain integrity too. Yet they all show significant improvement in results when the SPECfp source code gets compiled with proper compiler and executed on SMP machine. No such things were reported for Intel x86 architecture while Elmer was bragging on several occasions about advantages of Intel SMP workstation and lack of AMD's thereof. I asked Elmer this same question several times, but he is chicken or apparently has no clue, although he pretends to be in the know about Intel's SPEC activity. The task to run N apps on N-SMP is a no-brainer. To split an application algorithm at source code level is a task of much higher complexity. Until now there were no x68 compiler that would do this automatically. The only s/w company who claims ability to do such transformation at source-code level was Kuck $ Associates. They were recently acquired by Intel, kai.com so in some future we may be able to see SPECfp on Intel multiprocessor machines. Eventually. Implications for future AMD SMP machines are obvious. Regards, - Ali