lloyd@must-have-coffee.gen.nz (Lloyd Parkes) writes:
> The 25 degree gradient is between the thermocouple and the thermal
> pad. I expect the thermal gradient between the heatsink and the
> thermocouple is quite sharp.

That was a possibility I wondered about too.  Mark-I fingertips
confirm that the heatsink is barely warm.

I think the Asus A7V's CPU sensor just needs a lookup table to convert
it to real temperature units.

cjs@cynic.net (Curt Sampson) writes:
> I don't have access to my Athlon machine right now (it's running memory
> tests rather than NetBSD) but there are two things you might want to
> look at. First some of the older BIOSes caused Windows to report wrong
> (too high) temperatures, so this might affect this program as well. Use
> at least version 1007 of the BIOS on a A7V333.

Although netbsd probably doesn't go through the bios to read the temp.

A few of the authors of m$-based temperature monitoring programs
mentioned that they have to add/subtract a fixed number of degrees
from the CPU sensor to get the right answer.  This doesn't give me a
"warm fuzzy" feeling about the care used in designing this hardware.

> Second, I can't see whether the second number above is the
> motherboard monitor for the temperature of the CPU or the CPU
> monitor. (There are two sensors for CPU temperature, one external
> and one internal.) You may be looking at the wrong one.

I'm pretty sure the first is the mobo temp and the second is the CPU
temp.

-wolfgang
-- 
Wolfgang S. Rupprecht 		     http://www.wsrcc.com/wolfgang/