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: Robert Salasidis who wrote (135742)	5/21/2001 1:52:04 PM
From: dougSF30	Respond to of 186894

Robert, Re: To add my .02 worth to the debate, anyone that could conceivably claim a crash is better than a throttle back, doesn't do any useful work on a computer. As someone who has done useful work on a computer, let me make a point or two in response: 1) Actually, there are often times when a fatal error is preferred to a 'subtle failure' (read: bug) that doesn't manifest until far downstream, making diagnosis and correction much more difficult. In this case, imagine you're running a busy server cluster for an ecommerce site (okay, pretend it was a year ago or so :) ). Would you rather discover hard failure during initial stress testing, or, on that day you finally get 5000 simultaneous visitors, have everything become slow as molasses for periods of time? One could argue that the initial stress testing might not trigger the crash, so this argument is so-so. 2) I think we're claiming we prefer a throttleless design accompanied by real thermal specs (e.g. the Athlon), which OEMs can design into systems that a) don't throttle back, and b) have no thermal-related crashes, as they have adequate thermal solutions. Intel is 1) making it difficult for OEMs to know what an adequate thermal solution is, since they publish 'real world' thermal specs, instead of 'real' thermal specs, and 2) actually encouraging them to get by with cheap thermal solutions, relying on the throttling in hot situations. 3) apparently unable to tell us that the throttle will not activate at times even with a great thermal solution, with some (granted not the best) evidence to the contrary floating around. Doug p.s. I agree that the duty-cycle register would be nice.