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: pgerassi who wrote (142528)	8/30/2001 5:46:46 PM
From: Windsock	Read Replies (3) \| Respond to of 186894

Pete You suggest that IPC is the best performance measure. The first question is what instructions are you going to use to determine. The rate will vary widely for different instructions. So the question becomes which "average" set of instructions are best to use. In reality this means what software should be used. Now you are back to the benchmark arena and the performance numbers will vary depending on the benchmark you select. You can prove about anything if you find the right benchmark. Dan3 suggests that you use SuperPi but how many people want sit around calculating the vale of Pi? The relevant test for a user is how the system performs running the specific software on the data of the target use. The only performance that you can accurately measure on a processor -- and currently people agree on the standard -- is the clock frequency. This has proved the most accurate indicator both in the past and currently. AMD thought frequency was important and when they had the lead or were competitive with Intel. Now that AMD is 600 MHz behind and increasing, you can't change the scoring system.