Question Help with suspected CPU temperature problem

[BaronVonChickenpants]

I am running a AMD Bulldozer FX-6100 AM3+ 3.3 GHZ 6core CPU in a MSI 760GM-P23 FX Motherboard, with a ASUS ROG StriX GTX960 graphics card running Win 10 64 bit. I used a Cooler Master Hyper 212 EVO CPU Cooling System for the CPU heatsink.

It’s been a long time since I rebuilt my PC, the build I made about 9 years ago has been great, but now getting a bit slow so I went for a cheap upgrade. I bought the above collection of parts from another working PC as a set and have rebuilt using them. The system is working, and I can run /some/ games without issue however I have been experiencing lock ups that can only be resolved by pulling the power on the box. ( the screen goes black, tower fans stay on and the system is unresponsive to even the tower power button) Anecdotally I was only getting this while playing Fortnite so assumed there was a problem in just that game but now other games are affected so I have investigated further.

I used the AMD Overdrive tool to capture these details while the PC was running. That Thermal Margin figure /never/ leaves the large negative number, and I understand that means my CPU thinks it is running too hot and is throttling down:-

https://imgur.com/XSAHnsA

View: https://i.imgur.com/XSAHnsA.png



The only time I see a non negative figure is after rebooting the system:-

https://imgur.com/w1Xicgr

View: https://i.imgur.com/w1Xicgr.png


Even then, that seems to be running pretty hot when idle 45/50 degrees if I am reading the stats on the left correctly .

I’d appreciate some steps to follow to troubleshoot this.

I am currently planning on opening the case and verifying that all the fans are running, specifically the CPU cooling fan , although I’d expect temps to be much higher if that fan wasn’t running at all.

I was next going to take off the heatsink and reapply it with some different thermal paste. ( How do I clean up the old stuff?)

What else should I look at or check for please?

 

[Obakasama]

As you'll probably know by now, AMD Overdrive is the most accurate for those old AMD CPUs with respect to thermals. It is strange the sensor seems to work, but then goes to such an extreme reading - which if true I'd expect the system to shut down.

As for cleaning off the old thermal paste, use isopropyl alcohol and a lint-free cloth (or similar) before applying new thermal paste. I would suggest doing this first to see if it resolves the thermal reading.

 

[drivinfast247]

Did you reinstall Windows after the upgrade?

 

[BaronVonChickenpants]

Yes, it was a full reinstall onto a new SSD

 

[BaronVonChickenpants]

As you'll probably know by now, AMD Overdrive is the most accurate for those old AMD CPUs with respect to thermals. It is strange the sensor seems to work, but then goes to such an extreme reading - which if true I'd expect the system to shut down.

As for cleaning off the old thermal paste, use isopropyl alcohol and a lint-free cloth (or similar) before applying new thermal paste. I would suggest doing this first to see if it resolves the thermal reading.

My understanding from the little research I did is that this V high negative figure it show when it is throttling it's speed down to try to cool off.

The weird thing is once it gets like that I can never clear it, other than a reboot.