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: Hans de Vries who wrote (90983)	2/1/2000 11:23:00 PM
From: Bill Jackson	Respond to of 1575772

Hans, I wonder what max speed would be on an Athlon for very short periods? By that I mean find the max speed it will run at for 1/10 second. This would allow the thermal inertia of the chip etc to absorb the heat for that short period and you could then clock it faster and faster(with cool down times) until you found the maximum speed at which it could operate. Even .1 seconds is many operations and so some test program could be run. The chip could even be cooled down and run for 2/10 th second. At some point it would make errors or whatever and that would be the maximum. With 25-50 watts input this time would be quite short, but it would indicate the basic ability of the chip tp operate at that speed. You get some indication of this when they overclock and start to hit unstable/shuts down after 5 minutes/fails to boot etc as they try assorted parts under faster and faster clock speeds. Now you say the maximum overclock point with colling is probably ~40% with some expectation for a long life for the chip? How can you tell if you have a 4 week lifetime in such a short test at 40% overclocked? If it went into an area of high heat that caused increased diffusion so it failed in 4 weeks how could you tell this? I am sure if you tested it each day and plotted some leakage or other aspect you would find someway to measure the approach of failure in 4 weeks. I suspect that AMD/Intel engineers do perform some accelerated life tests with some variation of over coltage/overclocked/undercooled to aid them in determing the operational envelope of their parts? Has any of this assessment work ever made it into public view from AMD/Intel? Have any third parties ever done this as a systematic exercise( a costly one, I am sure) Bill

To: Hans de Vries who wrote (90983)	2/2/2000 3:13:00 PM
From: tejek	Read Replies (1) \| Respond to of 1575772

If anyone get's an Athlon running at 1.4 GHz then it must be a 1000 MHz or 1050 MHz one. Hans, so what you are saying is that only thru special methods (like cooling)are chips overclocked and is not an a real indication of potential Mhz? In other words if the the best yields for the AThlon on Al is 1000 Mhz, then the only way to get to a higher Mhz aside from overclocking is to use say Cu or go to .18um process from .25 process. Is that correct? Thanks again for your help. ted