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 : All About Sun Microsystems -- Ignore unavailable to you. Want to Upgrade?

To: Prognosticator who wrote (42751)	4/5/2001 8:50:56 PM
From: Brian Sullivan	Read Replies (2) \| Respond to of 64865

Well look who is giving the explaination, a SUNW spokesman, whose job it is to put the best case forward. You can bet that one or more of the SPECfp tests degraded by 5%. I'd also be willing to bet that one can also easily construct a test case that would lose a lot more that 5%. And regarding your comment: Like as if a 950MHz chip was clocked at only 902MHz. Like SUNW even has a processor that can run at 950MHz, or even at 902MHz, Hah!

To: Prognosticator who wrote (42751)	4/5/2001 10:34:27 PM
From: Tony Viola	Read Replies (2) \| Respond to of 64865

Prognosticator, Re: Not five whole percent. Damn! That's terrible. FIVE. Wow! A whole 0.95 of the full speed. Nooooo. Five Percent, I can't believe it. No. No really I can't. Five Percent. That machine is no good for anything other than heating a room. System architects and developers at any server company that sells performance would kill for an additional 5%. Your trashing of the significance of it is a bit irresponsible, I would say. Tony

To: Prognosticator who wrote (42751)	4/6/2001 12:46:24 AM
From: rudedog	Read Replies (3) \| Respond to of 64865

Prog - that 5% is an amazingly optimistic projection. Disabling prefetch on another RISC chip I worked with dropped performance by 30% in database applications and by 25% in a transaction mix. 5% implies a very lucky instruction sequence and infrequent processor stalls on that sequence, or an I/O bound process where the processor speed is immaterial. I'm sure someone will run some real-world tests and that will tell the tale.