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: deibutfeif who wrote (135994)	5/24/2001 1:28:20 AM
From: pgerassi	Respond to of 186894

Dear Debutif: You need some training before shooting off your mouth. Tdelta = Pexpended * Cheatsink. Cheatsink is expressed in degree C rise per thermal watt flow. A good heatsink coefficient is 0.30C/W (mine is better than this, an Alpha 6035MFC). Thus, for every watt, the temperature rises so much. This is typically very linear and has always been used to determine thermal watts flow (it actually underestimates flow by some factor as some is conducted to the CPU pins). Tdelta is the difference between ambient and the average temperature of the die surface. You can reform the equation to enter Tdelta and the heatsink coefficient to get power expended: Pexpended = Tdelta / Cheatsink. Thus, a Tdelta 4 times higher yields a Pexpended 4 times higher. Thus, his point that base load of the OS appears to be 1/4th of the UD client load. This is against Intel spec as a typical load is supposed to be only 1.3 to 1.4 times base load. So one or the other of Intel's guidences must be incorrect. On my 1.425G Tbird, UD agent runs about a 12C delta and when base load (Win98SE only), about a 5C delta measure from top surface of die to ambient. BTW, did the tester verify the chip temperature to be accurate? Thus, by the coefficient of my heatsink 0.25C/W, I run 20W at base and 48W at UD load on my AXIA Tbird 1.425G @ 1.75V. If we give a typical coefficient of 0.30C/W, the idle power would be 30W and the UD load would be 107W, far higher than Intel claims. Assuming it was just typical, the Cheatsink would be 0.60C/W, a really poor number. Even taking into account that the active bottom side to top side of die thermal resistance, its bad. The typical one would either blame the on die readings (probably too low) or blame Intel's thermal wattage specs (which are known not to cover max usage). Your pick! Just think of what would happen in an environment of 100F (33C). The on die would exceed the 70C spec'd. Oops, out of spec., may get incorrect results per Intel. Pete