AMD Boards with Overheating Shutdown protection

From the previous posting "AMD BURNUP", I am curious as to how well the overheating protection that is built into some motherboards handle. An extremely valid point was made also, that if your CPU crashes(due to a fan failure for example), then the Motherboard Monitor will crash with it.

My question is, can your hardware monitor chipset still function when your CPU crashes, & if so, does the BIOS overheat shutdown use it's own hardware system to shutdown the PC, or does it too, like Motherboard Monitor, rely on the CPU to still be processing.

Another Question I'm curious about is Athlon/Athlon XP average temperatures, I can't seem to find any figures to compare my system on.

Athlon XP 1600 w WBK38 Cooler @ 1.425Volts
Soltek SL75-DRV2 KT266A Mainboard
256MB Century Micron Memory(slightly OC, CL2.5 => 2, & 1T Command)
Voodoo 3 2000 PCI(Thought I'd add that in to point out the large bottleneck of my system, still, it overclocks well 175/175 )

I shaved off approx 8-10 degrees dropping the voltage down that low, however it's perfectly stable at 1.4V, though, for extra stability, I added .025V

Night:
34 degrees internet usage/idle(add 2-3 degrees full load)

Day:
38 degrees internet usage/idle(add 2-3 degrees full load)

<font color=blue>"An extremely valid point was made also, that if your CPU crashes(due to a fan failure for example), then the Motherboard Monitor will crash with it."</font color=blue>

Not a valid point at all. If your fan fails, the heatsink will continue to absorb heat, allowing plenty of time for Motherboard Monitor to shutdown the system.

Now, if you remove the HSF while the system is running, as demonstrated by THG, the results are a fried chip. But who would remove a HSF during operation, and how would a HSF fall off by itself?

The temperatures you list are fine. Anything below 60c is OK, and according to AMD, anything below 90c is OK. Some people claim that operating above 60c will shorten the life of the chip. At this time I have never seen any facts to back that up. My T-bird operates around 48c under full load.

<font color=blue>This is a Forum, not a playground. Treat it with Respect.</font color=blue>

You have to set your shutoff temp BELOW the crash temp. Thats the point of a cutoff, if it crashes, there is a chance mobo monitor will lock up as well, simply set the temp for a high end safe temp, if it shuts down, but you were fragging on quake, increase the temp 1 degree....toast would be great for this, run toast, then set the temp 1 degree above peak, this is below lockup temp, but hotter than it ever should get.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

Amd's thermal shutdown logic does run independant of the cpu, and when the new motherboards begin using it properly amds solution will be just as effective as intels.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

I forgot to mention that CPU StabilityTest 6.0 is able to drive my temps. from 37°C to 61.5°C after 10 minutes of "CPU Warming", but I've also just discovered that the stock temp. controlled fan seems to stay put at 2400RPM no matter how hot the CPU gets. I have to talk with Abit about this. Also under such extreme load conditions the two 80mm low noise 19.4CFM Pabst 8412NGL case exhaust fans isn't sufficient. The reason for the high CPU temp. is due to the case temp climbing steadily throughout the testperiod to about 50°C, so the CPU fan is essentially dragging 50°C hot air through the fins of the heatsink.


/Copenhagen - P4 Willamette 1700MHz@2109 MHz, Vcore 1.75V@2.20V on Abit TH7II-RAID.