A7V8X temperature sensor

12th January 2003, 12:27

I have an A7V8X KT400 motherboard, AMD XP2000 (palomino) cooled by a Volcano 9 coolmod @ 4000 RPM and PSU fan (no other casefans at the moment). With motherboard monitor 5.2 I get these readings (verified with the maker of MBM):

(Full load)
temp MB: 43
temp CPU diode: 40 (diode temp lower than MB is possible if the MB sensor is located in a spot with little airflow according to the developer of MBM)
temp CPU socket: 62 (socket is cooled less than the actual core so it will always be higher than the core temp according to the maker of MBM)

Now what temperature should I worry about. If you look at the diode temp I have some some room for overclocking (max core temp is 85-90 but I wouldn't dare go anywhere near that high), but the socket temp also rises when I overclock. Is there a max socket temp that I shouldn't exceed?

jmke · 12th January 2003, 12:47

only the CPU diode temp is important afaik

12th January 2003, 14:39

isn't 40C a bit low for a XP 2000 palomino at full load, without any additional cooling besides a volcano 9? I would have thought that 62C was more like it.

jmke · 12th January 2003, 15:08

I think your MBM is switching the 2 values of diode <> socket

12th January 2003, 15:12

nope I asked the developer of MBM himself, and he was sure that it's right. The socket gets less cooling than the core itself and therefore shows higher temps.

Another reason why I'm pretty sure the temps are correct, is because the CPU diode temp reacts much quicker to change in CPU load than the CPU socket temp. What is possible is that one of the sensors isn't calibrated properly, but that's something I can't test.

jmke · 12th January 2003, 15:43

the CPU temp in the diode will be higher then the temp readings from the socket, always

28th January 2003, 11:40

My sucky coolermaster @43° idle, 45° load.

look sig for pc.

12th January 2003, 12:27	#1
Manny Posts: n/a	A7V8X temperature sensor I have an A7V8X KT400 motherboard, AMD XP2000 (palomino) cooled by a Volcano 9 coolmod @ 4000 RPM and PSU fan (no other casefans at the moment). With motherboard monitor 5.2 I get these readings (verified with the maker of MBM): (Full load) temp MB: 43 temp CPU diode: 40 (diode temp lower than MB is possible if the MB sensor is located in a spot with little airflow according to the developer of MBM) temp CPU socket: 62 (socket is cooled less than the actual core so it will always be higher than the core temp according to the maker of MBM) Now what temperature should I worry about. If you look at the diode temp I have some some room for overclocking (max core temp is 85-90 but I wouldn't dare go anywhere near that high), but the socket temp also rises when I overclock. Is there a max socket temp that I shouldn't exceed?

12th January 2003, 12:47	#2
jmke Madshrimp Join Date: May 2002 Location: 7090/Belgium Posts: 78,420	only the CPU diode temp is important afaik __________________

12th January 2003, 15:08	#4
jmke Madshrimp Join Date: May 2002 Location: 7090/Belgium Posts: 78,420	I think your MBM is switching the 2 values of diode <> socket __________________

12th January 2003, 15:43	#6
jmke Madshrimp Join Date: May 2002 Location: 7090/Belgium Posts: 78,420	the CPU temp in the diode will be higher then the temp readings from the socket, always __________________

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
"Duh" news of today: Voltage Has Great Influence on Temperature & Power Consumption	jmke	WebNews	1	11th February 2009 22:03
Fancy images prove HSF Lapping Drops CPU Temperature	jmke	WebNews	0	25th June 2008 15:12
Real Temp new DTS sensor program using simple Ia32 driver	Liquid3D	WebNews	10	9th March 2008 16:15
Maximum CPU Temperature	Sidney	WebNews	0	11th October 2007 14:25
NZXT Sentry Temperature Controller	jmke	WebNews	0	19th June 2007 09:49
Thermaltake Water Level Indicator and Water Temperature Indicator	Sidney	WebNews	0	3rd July 2005 05:48
Accuracy of the sensor?	goingpostale1	Air - Water - Peltier - Phase Change - LN2 Cooling Projects	3	24th March 2005 17:59
Myth of CPU temperature	Sidney	General Madness - System Building Advice	2	25th December 2004 16:38
Ultra Low Temperature Experiment: LN2 & AXP	jmke	WebNews	0	28th September 2004 12:28
Where is the mobo temp. sensor located?	GiantKiller	Hardware/Software Problems, Bugs	4	15th June 2003 13:52

12th January 2003, 14:39	#3
Manny Posts: n/a	isn't 40C a bit low for a XP 2000 palomino at full load, without any additional cooling besides a volcano 9? I would have thought that 62C was more like it.

12th January 2003, 15:12	#5
Manny Posts: n/a	nope I asked the developer of MBM himself, and he was sure that it's right. The socket gets less cooling than the core itself and therefore shows higher temps. Another reason why I'm pretty sure the temps are correct, is because the CPU diode temp reacts much quicker to change in CPU load than the CPU socket temp. What is possible is that one of the sensors isn't calibrated properly, but that's something I can't test.

28th January 2003, 11:40	#7
Str0ntj3 Posts: n/a	My sucky coolermaster @43° idle, 45° load. look sig for pc.