Spajky tried to express:
On 30 Aug 2004 15:09:17 -0500, wrote:

the driver software shows a default
core slowdown temp of 140 celcius and this seems WAY HIGH!

sensor is bad! or
a farenheit value maybe?

This is the value that was entered as a default as to when the GPU should
fall back to a slower speed. Yes, my guess is that the programmer put
a farenheit value in where they should have used a celcius value...This is
what I'm trying to determine.


What would be a reasonable operating temperature for the GPU and at what
temp should I flag a temperature alert?

70° would be Ok IMHO

I agree. I think 70C should be top end for safe semiconductor operation.


The Xeon CPUS automatically throttle back at 71 celcius. This seems like
a much more reasonable value.

IMHO they start throttling after 100°C from the onDie diode ...

I think you are mistaken. According to the Intel product documentation the
maximum operating temperature for the Xeon is listed as 71C.

Overall, the system doesn't seem to overheat. Ambient temperature in the
case is in the mid 40sC. The CPUS are in that range, and the NVidia GPU
usually runs in the mid 50sC. I am not doing any real 3D stuff (YET):
just Linux X-windows and win-xp, with an occasional simple OpenGL wireframe
of a cave map or some other engineering drawing.

The power supply fan output is pretty hot, but the internal case temp seems
well within acceptible limits.

Thanks