Voltage and Heat and Time - Oh My!

[FZ1]

I ran across a great thread where CPU voltage/heat effects are discussed with some very knowledgeable folks (from Intel even!) giving some great info. I just had to share...
http://forums.anandtech.com/messageview ... erthread=y

Disclaimer - Please note that none of these folks speaks for AMD or Intel. So if your chip craps out from one or the other, look in the mirror for the culprit.

[kenc51]

:bookmarked:

Super find Vstrom.......

I done chemistry in school........now to Google "NMOS hot-electron"

[pointreyes]

Disclaimer - Please note that none of these folks speaks for AMD or Intel. So if your chip craps out from one or the other, look in the mirror for the culprit.

Well, not necessarily in the mirror. I had a P4 2.53 die on me - it was not oc'ed and it was running within temp specs (or so I thought). Back in that day, I used a screen saver - the typical MS one, not some fancy 100% CPU load type - and all of sudden one day my system was locked up on the screen saver. This was a Asus p4pe board and it did not have a bios setting to shutdown the system if it overheats. Intel fanboys back then would claim that was not needed since the P4 had throttling. I countered that I would like to know if any proc is overheating that I want the system to shutdown because that helps to lead me to what to look for. It for that very reason that I did not trace the problem till after the proc killed the motherboard and one stick of RAM or visa versa. I had MBM on the system but with a screen saver you would have no idea what the temp was. A few days later the problem occurred again but on a reboot - a blank screen. Opened up the case and there was huge dust bunnies all over the procs' HSF. The rest of the system was dust free! I normally keep my systems relatively clean so this did not make sense to me. And then I realized that maybe the proc or the mobo went haywire and somehow maybe caused a powerful electromagnetic suction to the proc. I moved the proc to another system I had and the temp of the proc was 70C+ at idle. Now I'm not sure what caused the failure but it sure seemed to me that it was electronic and heat.

Normally, I will blame voltage before heat in a well built system.

[FZ1]

Maybe I should say "if you exceed the manufacturer's recommendations"

[killswitch83]

holy crap is all I have to say......I consider myself an intelligent person, and much about the analytical side of concepts like electromigration and the rest of them has flown slightly above my head.....they do well in explaining it, but I guess it takes a great knowledge of the components used in certain layers of the processor to understand it better, and how it all works together (not to mention their separate purposes)

The Intel Design people there are all our daddies, roflmao