Motherboard's "TMPIN3/CPU (PECI)" temperature always at 40°C? a broken sensor ?

wen87n

Prominent
Sep 24, 2017
10
0
510
9HxAWPT.jpg


Hi guys, I ran a stress test and realised one of the sensor name  "TMPIN3" or "CPU (PECI)" always remain at 40°C.

As you can see in the picture, its Min temp and Max temp is remain at the same 40°C never change after stress test. unlike other sensors will getting hot when stress test and cool down when idle.

Am I having a broken sensor? is this normal?

My motherboard is MSI B350m Mortar Arctic.
CPU is Ryzen 5 1600 OC@3.7Ghz
16gb ram @2666mhz
 
Solution

I wouldn't mind I just don't know how LOL

Suffice it to say that my reading for "TMPIN3" and "CPU (PECI)" in HWMonitor and HWInfo64 both read the same as yours: a constant 40C.

As CountMike says: the most reliable and accurate CPU temp is HWInfo64's, specifically the "CPU (Tctl/Tdie)" reading. I think that's the same as the "Package (Node 0)" temperature reading in HWMonitor as they certainly seem nearly identical even as loads change.


It's not often that all sensor lines of the monitoring chips on motherboards actually have a sensor connected. When that's the case the readout in monitoring utilities will be random: sometimes 'free-wheeling' and all over the place, sometimes fixed in an illogical reading. I think that's the case here and I have seen the same readings on my B350M Mortar.

But I also know HWinfo has a new beta release out that supposedly fixes some MSI sensor readings so that may be the case. My desktop system is down RMA'g the processor so I can't check it.
 

wen87n

Prominent
Sep 24, 2017
10
0
510


Would you mind to pose a screenshot of yours for compare purpose?
 

I wouldn't mind I just don't know how LOL

Suffice it to say that my reading for "TMPIN3" and "CPU (PECI)" in HWMonitor and HWInfo64 both read the same as yours: a constant 40C.

As CountMike says: the most reliable and accurate CPU temp is HWInfo64's, specifically the "CPU (Tctl/Tdie)" reading. I think that's the same as the "Package (Node 0)" temperature reading in HWMonitor as they certainly seem nearly identical even as loads change.
 
Solution