This Duron I got is pretty damn incredible....

[RoboTECH]

It seems that it is handling 1 Gigahertz easier and easier as I burn it in. What's VERY VERY weird, and has me curious, is that it can loop a Q3 demoloop and run RC5 for hours, and not go up by more than 5 degrees celsius.

My ambient case temp is low 30s, my CPU idles @ 40, and it runs Q3 + RC5 @ 44-45 @ 1.75v

I'm missing something here, right? It starts off warm, but doesn't get hot when RC5 and Q3 are cooking it? WTF??

 

[Eug]

<< I'm missing something here, right? >>

Yeah, your temps are unreliable. I wouldn't worry about it though if you've got good cooling.

 

[Viperoni]

My suggestion for the ULTIMATE stability test: Turn off all your system fans, except the CPU fan.
Then run a Quake3 loop at the lowest possible res, with the lowest quality settings, with RC5 and seti running in the background. That'll load up your processor HARD.
If it's stable for more than 3-4 hours, you got a hell of a chip there 🙂

 

[glen]

Those are all ok tests.
For the REAL test of stability, I turn off all fans, and put a hair dryer on the CPU while I run SETI.

 

[Rastus]

I remove all fans, put the mobo in the oven and play freecell.

EDIT
Sorry guys: disregard this post. Sometimes I cant help myself.

 

[shiznut123]

the real test. take off all cooling hardware and run seti. put the case in the oven and set it to 500degrees c 🙂

 

[flyfish]

Pour some lighter fluid in the case and throw in a match while playing UT.

 

[Assimilator1]

RoboTECH

I take it your not using some kind of software cooler? ,with 1 you would see much greater cpu temps differences between idle &amp; full load.

 

[SolTek]

take the heatsink off, place a some uranium 236 on top, then place C-4 around the uranium and your processor in an evenly dispersed soccerball-like formation, then crush the uranium by detonating the C-4 in a shaped charge - creating a nuclear reaction... 😉

 

[Jugernot]

Ummm, am I the only one who removes their CPU from their case and chews on it until its stable? Those damn pins are harder than hell to get out of the teeth. They go back into the CPU rather easily using crazy glue, so that isn't a problem.

 

[DaddyG]

The PRO2A does not appear to have a thermistor in the Socket A (according to Dr TOM) so I'm not sure where the readings are coming from.

 

[lenjack]

SolTek.....I like your idea, but I don't think U 236 exists. You must mean U235 or 238.

 

[Nih]

No ,No , not uranium,,, all you have to do is remove the h/s compound and replace it with good old canned heat. Jeez hasn't anyone been camping before?