temp Problems

[Silent_Bob]

I'm kinda new at overclocking so this might be a stupid question. I have an IBM netvista with an Intel CA810E board and a Celeron coppermine CPU @ 767Mhz
I tryed to use speed fan to monitor temps while I try to OC it but I'm getting some messed up readings. There are three readings, #1 is between 28-34c #2 is 70-72c and the last is -46 t0 -48c. I'm no expert but that can't be right and there is no way my CPU is at 70c cause the inside of the case is never warm at all. I tried Hmonitor and it gave the same readings. I still OC it cause I'm ready to toss it due to software/hardware issues and I didn't care if I cooked It. The temp readings never increasded more than 2 degrees. Dunno what the problem is but I did get my comp to a stable 866MHz. Just wondering if this is normal for this board or if I missed something with the SMBus or the sensor type(there are two choices, Thermistor Diode and PLL Diode. they both give completly different readings.) Any advice would be great.

IBM Netvista 6276 34U
Intel CA810E
Celeron coppermine 767Mhz OCed to 866MHz
kingston pc133 256mb SDRAM
20g maxtor harddrive
Benq 52-32-53 CD-RW
Windows Xp Home SP2

[lptech]

Silent_Bob-

The systemboards on these IBM NetVistas are not really designed to be overclocked. That is why the Intel Pentium-IIIs were not really popular with overclockers. The Thermistor Diode might give you a more reliable reading since it is directly touching the bottom of the CPU itself. You can only tweak the FSB to overclock these. Good luck!

LPTECH

[Silent_Bob]

Thanks tried setting to the thermistor diode but now it says the temp is between 5-7c so if thats true sweeeeet but i doubt it cause the only cooling mods are a 80mm fan in the side of the case(no room in the back) and I replaced the stock heat sink with a Microfins cooler from a local comp shop. I think the sensor might be off or the programs are reading the SMBus wrong. I can't check the CMOS cause its an IBM system and the CMOS is pretty locked up.

[lptech]

Silent_Bob-

The way that some of those sensors are positioned, it is very hard to get an accurate reading anyway. Just go ahead and experiment as much as you want since you said that this is on its way out anyways. Good luck!

LPTECH