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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: milo_morai who wrote (74112)	3/9/2002 3:44:59 AM
From: Ali Chen	Read Replies (2) \| Respond to of 275872

Milo, "..the effect is likely even more pronounced than you infer". The guy constructed his example using three assumptions, about cacheline fill time, about cache miss rate, and about cache latency. He got the "total latency domination" and 10% gain per 50% speedup solely by this arbitrary construction. His assumption could be valid for some single task, but it is not what the industry benchmarks are today. According to me, in SPEC2000int, the P4 gains 32% per 50% speedup in clock frequency, from 2000 to 3000MHz, and for SPEC2000fp the gain is 20% for the same speedup. The "througput" calculations have limited use, same as IPC: who cares if IPC is slightly higher, but the clock is way below competition. - Ali