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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Mani1 who wrote (98213)	3/13/2000 1:02:00 AM
From: Scumbria	Read Replies (2) \| Respond to of 1572560

Mani, The 60 deg C max die temp is the lowest I have seen for a PC CPU, it is hard to do and I am sure there is good reason for it. It's actually quite simple. Apparently PIII doesn't run reliably at 1GHz and normal operating temperatures, so Intel lowered the maximum specification temperature. This is probably the sorriest excuse for a processor launch I have ever seen. Scumbria

To: Mani1 who wrote (98213)	3/13/2000 9:27:00 PM
From: THE WATSONYOUTH	Read Replies (1) \| Respond to of 1572560

Re: "The 60 deg C max die temp is the lowest I have seen for a PC CPU, it is hard to do and I am sure there is good reason for it." The most likely reason is performance. It is very hard to believe that the part is not reliable at 85C. For Al BEOL, you see approximately .37% decrease in resistance per degree C. So for 25C lower temperature, you would see about 9% lower BEOL resistance. This probably translates to about 3% overall performance increase. Remember, when Intel decided to stay with Al, they had to carefully redo the layout of all the critical paths in the BEOL in order to not be BEOL limited. (They admitted as much) In addition, the 25C decrease in junction temperature probably adds about 2.5% performance from a device point of view. Also, it probably buys them .05V to .1V in gate oxide reliability which allowed them to increase Vcc to 1.7V for 20A gate oxide. Overall, they probably got about 7% performance increase or about 70MHz. The trade off is the the more expensive/complicated heat sink which looks like they will need some kind of plenum to draw air thru to achieve this temperature. Wonder what the box makers think?? The intriguing question is: Did Intel have to do this just to squeeze out 1 GHz??? (IF) so, I don't know what else they could do to go higher with this design other than additional improvements in device design. THE WATSONYOUTH