thermal sensor adjustments

would it be reasonable to assume a constant 3.6C difference between the temp probe and bios all the time. bios at idle after 15 min, reads 3.6C higher than temp probe. so 15 minutes into winxp, would that still be a reasonable assumption? temp probe + 3.6C=diode temp....or are there just too many variables?

When running in the BIOS...that is like having the CPU running at 100%...so there will always be a slight temp difference from the BIOS to windows...

i was curious about that cause i read that windows would actually raise the temp...but would my reasoning still apply? 3.6C difference, both sensors at steady temps/reading....

Windows does not raise the temps after it is fully loaded...it might raise it slightly when loading...but after it fully loads the CPU will be at idle...unless you have a program that is causing the CPU to run at full load (folding, seti, or a virus for examples)

And as far as I know...the asus probe is not reading from under the CPU...which would give you that 3.6c difference...so yes...that would be a safe assumption...

if you have a temp probe, do u place it in-between the heatsink and processor? im just curious coz my case came with it.

NO...don't place it between the processor and the heatsink!!!! That will cause higher temps. Place it just above or below or somewhere around where the heatsink will touch the processor.

Best place to place a sensor is center of the bottom of the CPU...

it's not asus probe. i bought a 3.25 display but i put the temp probe next to the cpu...like i said, idle in bios 15 minutes stable and the temp difference is 3.6C constant. so i would hope that theory is true...3.6C difference period. answers? thanks glock19 and all...

A CPU does not run 100% in the BIOS. The temp difference can be caused by a few things namely placement of the either thermistor(MoBo sensor or your independent probe) & how they are calibrated. The relatively cheap products that use ribbon thermistors are by no means considered accurate, they do have a large margin for error.

The best place for your thermistor would be to lay it next to the core, worst place would be underneath the CPU were it is the farthest from the core, with part of its plastic cover removed like so;



SECONDARY HEATPATHS affects you temp reading & will cause you to be in error.

gonna try that...i'll remove part of the plastic cover....but i still don't have an answer to my question. can i use 3.6C as a constant difference? thanks again...

No, I dont believe you can. The readings are not even accurate enough for it to remain a constant. You can try your best to minimize the errors by calibrating it with a known good lab tool & making sure it is mounted properly,make sure to use thermal compound on the tip.

Remember that when you mod the product you are voiding your warranty. If you break it you alone are resposible for any damage.

yeah, i kinda figured that but you can always hope, right?...not worried about warranties so that's not an issue. thanks for your help...

It's always good to hope, if you get a stable reading then it might be a good product to recomend to other members who are not capable of reading the internal temp diode. Good Luck

Sonny has brought up some good points. Here's a little more on the subject of obtaining accurate CPU temps.

Any diode inside the chip will be far more accurate than a thermal probe placed outside the chip. Think about it, how accurately can you tell the temperature of an oven or a refrigerator with a thermometer mounted outside on the door? For more info read Why Many Thermal Measurements Are Not Valid. As most of us already know, in-socket thermistors will not give you anything resembling an accurate temperature. “Oh but I put the probe next to the CPU core, it must be accurate!” Think again. Here's an interesting heatsink round up that illustrates the point.

Somehow or other, but the main conclusion is this: for hardcore overclocking, the readings of external thermal sensor are absolutely useless since they don't reflect the factual processor core temperatures.

Click to expand...

You may find Temperature Sensing Technologies a good read too.

Now that we have figured out the best place to read the temperature is inside the CPU die, another problem pops up. With PII and PIII CPUs, the sensor was in the hottest part of the chip. Here's what Intel has to say about PIII diode placement. With P4s and Athlons the diode is no longer in the hottest part of the chip. A little more reading from Intel on the subject. We have the same problems measuring temperatures inside the core when the diode is not in the hot spot as those outlined in Why Many Thermal Measurements Are Not Valid.

The only accurate CPU temperature measurements are from PII or PIII internal diodes provided you have a motherboard that can read the internal diode. There is no way for the end user to get an accurate temperature reading from an AMD CPU or a P4.

Furthermore for those of you with P4s, if you consider the last link from Intel, your CPUs are most likely throttling more often than you may realize.

This is information I was looking for. LOL It seems that I have some temp dofference issues. My motherboard uses the LM-90 remote sensor and it gives me 46°C while my hardware temp sensor gives me 52°C. If the temp sensor is still reading low, then I'm very unhappy.

So far the closest reading we can get that is with the Internal Diode even though its still not completely accurate. It has the least amount of distance from the hottest part of the CPU which in my mind makes it more reliable compared to other ways of measuring the device. Having said that we shouldnt be to convinced of Internal Diode readings, its a guide unlike before with thermistors on the back of a CPU which was like a broken lighthouse in a storm now we have a semi working lighthouse.

Given the inherent improbability (perhaps impossibility even) to get an accurate CPU temperature reading how then should we intrepret such statments as an AMD XP chip has an upper temperature limit of 90ºC (or whatever the figure is). Or the generally quoted rule of not running your processor over 50ºC.

Now granted I'm sure AMD is gonna have a better idea of how to get an accurate temperature rating, and ultimately the important factor really is stability. However the point I'm trying to raise is that if we are all inherently inaccurate, and from what I gather the inaccuracy is generally to read cooler than the actual (valid assumption?), at what point point can we say "close enough". In another words if everyone who is saying 50ºC is a good rule of thumb are actually running they're processors at 60ºC when they are reading 50ºC then how useful is to get really accurate temps you can as you would then be 10º out of step with everyone else.

Granted I'm making gross generalizations here to beg a question, something along the lines of: how accurate should we strive to be?