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: Ali Chen who wrote (148898)	11/20/2001 12:32:56 AM
From: wanna_bmw	Respond to of 186894

Ali, Re: "And again, do you think that a workstation CPU has a right to have 50% leakage?" I think a better question (assuming that there is a difference between Intel's .13u server and workstation chip and their .13u mobile chip), is what kind of trade-offs are there if the transistors are made such that there is less leakage current? In other words, does Intel lack the ability to lower leakage, or does lowering the leakage at this point in time result in other tradeoffs that could affect performance or reliability? I think the big question here, though, is what does it mean to have higher levels of leakage? It seems to me that the most visible means of judging the chip involve simply looking at the levels of power dissipation from the data sheet. Whether leakage composes a small or a large part of the total power dissipation, is it not all the same to the end user? If the end user is still getting a lower power chip, does it matter how much of the power that exists is due to leakage? Keep in mind that over the course of a process generation, the transistors do get improved. Witness the evolution of Intel's .18u process, which eventually allowed the development of ultra low power Pentium III chips running at 1.1V. Likewise, AMD's improvements in their .18u process eventually allowed the development of low voltage, high speed .18u Mobile Athlon 4 CPUs. Judging the process this early in the game may not yield conclusions that hold up later on. Unless, for example, it can be shown that these leakage levels are detrimental to the reliability of the current CPUs. So far, however, that doesn't seem to be the case. I think the more immediate concern is whether or not these leakage levels make it more difficult to achieve higher frequencies through barriers in power dissipation. If Intel can improve their process enough to reach the levels in frequency that they currently have on their roadmap, then they won't have much to worry about. On the other hand, if their process becomes a handicap, it would give the competition a larger advantage. Speaking of which, who's to say that AMD won't encounter similar difficulties with their .13u generation. While they may be using smaller gate lengths now, shrinking the entire transistor may introduce new problems, not to mention problems with their Silicon on Insulator technologies. I'm not a process expert, but I certainly don't take it for granted. There is certainly a lot involved, and I'm not sure whether dwelling on leakage levels is going to paint the full picture. wbmw