should i trust my temp readings?

[the duck man]

So i put together my new pc, did some reading, decided to dabble in a bit of overclocking, and got hooked

I'm running an e6750, MSI p35 neo MB, 2x1gb corsair pc6400 ram and an x1950xt, air cooled with an arctic freezer 7.

So far i've bumped the FSB up to 433, increased the ram base speed to match, and tightened the timings to 4-4-4-12 from 5-5-5-15, with a .05 increase to ram voltage and none to Vcore (i think the MB default Vcore is quite high to begin with). Everything's running smoothly, no stability problems, but i'm a bit wary of the temperature readings i'm getting.

Core Temp, Speedfan and TAT all show the cores at 13C / 17C idle (speedstep enabled, so that's with the multiplier dropped to x6) and 40C / 42C under 100% load generated by TAT, which seems stupidly low given that i havent done anything special cooling-wise and the ambient temp is probably around 20C. Speedfan shows Temp 1, which i'm assuming is Tcase, as 33C idle and 55C under load, which seems more likely.

The question is should i believe the core temperature readings and keep pushing the FSB, or quit while i'm ahead? Are those temperatures reasonable or is it more likely they are offset, and if so is there any way i can find out by how much?

cheers,

duck

 

[Mousemonkey]

So i put together my new pc, did some reading, decided to dabble in a bit of overclocking, and got hooked

Welcome to the Dark side young padawan, now Orthos for 3 hours you must run.

 

[bosshoss]

What I found to work for me was to pull up all my temp readings, i.e. Asus probe, TAT, RMclock, Core Temp, etc... and watch my idle temps. I would quickly reboot into my BIOS and check my BIOS temps to see which program was reading temps closer to the BIOS temps. For me, Core Temp was the closer of 'em all. And personally, when I began to OC my chip (E6300 @ 2.8), I turned off Speedstep and and a few other power management functions in the BIOS so I knew I would get a high idle temp as the CPU was never throttled down.

 

[Mugz]

OCing is a very rewarding hobby. Don't place too much faith in the monitoring programs, I've been getting reports of -115ºC in Speedfan on my Northbridge lately. And, so far, only my GPU and CPU are water cooled... I'd hate to live in a place where a component can really hit -115ºC on air.

 

[gomerpile]

So i put together my new pc, did some reading, decided to dabble in a bit of overclocking, and got hooked

I'm running an e6750, MSI p35 neo MB, 2x1gb corsair pc6400 ram and an x1950xt, air cooled with an arctic freezer 7.

So far i've bumped the FSB up to 433, increased the ram base speed to match, and tightened the timings to 4-4-4-12 from 5-5-5-15, with a .05 increase to ram voltage and none to Vcore (i think the MB default Vcore is quite high to begin with). Everything's running smoothly, no stability problems, but i'm a bit wary of the temperature readings i'm getting.

Core Temp, Speedfan and TAT all show the cores at 13C / 17C idle (speedstep enabled, so that's with the multiplier dropped to x6) and 40C / 42C under 100% load generated by TAT, which seems stupidly low given that i havent done anything special cooling-wise and the ambient temp is probably around 20C. Speedfan shows Temp 1, which i'm assuming is Tcase, as 33C idle and 55C under load, which seems more likely.

The question is should i believe the core temperature readings and keep pushing the FSB, or quit while i'm ahead? Are those temperatures reasonable or is it more likely they are offset, and if so is there any way i can find out by how much?

cheers,

duck

If you get a few extra buck and don't know what to buy for you computer, try a digital monitor or display.More on temps there are 2 available types of controllers a sensor that measures heat by a coded color chart ANSI temp controller built into a front panel display like the aerogate. The sensor that is attached to a point of interest and feeds a digital number to the controller than matches the number to a coded ANSI chart than displayed to the LCD
2 is a micro controller that most computers now have with this kind of controller it measures voltage usages to measure heat. example the core cell chip in msi boards tc620,tc621,tc622,tc623 these are micro controllers that measure heat by voltage usages and it to is ansi coded these kind of controllers usually comes with software to change settings of the controller. This protects a computer when overheating occurs these controllers will send a shutdown to the psu when overheating occurs, in the bios we have the option to disable this warning.