Possible wrong temperature readings?

[bigbali141]

I can't decide which one is wrong, because something obviously is.
Picture 1
Picture 2
My PC is sometimes shutting down, I thought it's a PSU related issue. Now I realize it might be an overheating one.
How is it possible the temperature readings differ so much?
I have done intense CPU rendering with Blender before, but it did not shut down. (rendering time was hours, 100% cpu usage)
I even have replaced the cooler (well, the factory 50mm with an aftermarket cheap 80mm, with the paste it came with).

 

[krzyimprt]

I can't decide which one is wrong, because something obviously is.
Picture 1
Picture 2
My PC is sometimes shutting down, I thought it's a PSU related issue. Now I realize it might be an overheating one.
How is it possible the temperature readings differ so much?

Seems it may be an incorrect reading, whats the temp in the bios say? Try realtemp and speccy, see what the temps are. Also turn on high performance mode if you want better performance in windows, cores should be running at 3.9GHz.

 

[bigbali141]

I can't decide which one is wrong, because something obviously is.
Picture 1
Picture 2
My PC is sometimes shutting down, I thought it's a PSU related issue. Now I realize it might be an overheating one.
How is it possible the temperature readings differ so much?

Seems it bought be an incorrect reading, whats the temp in the bios say? Try realtemp and speccy, see what the temps are. Also turn on high performance mode if you want better performance in windows, cores should be running at 3.9GHz.

My BIOS does not allow me to set it to performance, so I left it on balanced. Base clocks are 3.9 GHz, but it reaches up to 4.2 with AMD's automatic profiles.
BIOS readings are between 20-50 degrees. I forgot to put in AMD OverDrive readings, so when the encoding finishes I will post it here.
(have to restart)
+ Is it possible the PC shuts down not because the PSU (which I will replace anyway), but the false readings?
If so, can I fix it in any way?

 

[krzyimprt]

I can't decide which one is wrong, because something obviously is.
Picture 1
Picture 2
My PC is sometimes shutting down, I thought it's a PSU related issue. Now I realize it might be an overheating one.
How is it possible the temperature readings differ so much?

Seems it bought be an incorrect reading, whats the temp in the bios say? Try realtemp and speccy, see what the temps are. Also turn on high performance mode if you want better performance in windows, cores should be running at 3.9GHz.

My BIOS does not allow me to set it to performance, so I left it on balanced. Base clocks are 3.9 GHz, but it reaches up to 4.2 with AMD's automatic profiles.
BIOS readings are between 20-50 degrees. I forgot to put in AMD OverDrive readings, so when the encoding finishes I will post it here.
(have to restart)
+ Is it possible the PC shuts down not because the PSU (which I will replace anyway), but the false readings?
If so, can I fix it in any way?

Whats the make and model of your power supply?

 

[bigbali141]

http://www.tomshardware.com/answers/id-3603252/computer-randomly-shuts.html
Here you go.

 

[krzyimprt]

http://www.tomshardware.com/answers/id-3603252/computer-randomly-shuts.html
Here you go.

Yeah... That power supply has to go regardless if it fixes your issue or not.

 

[bigbali141]

So far, I have run a stability test with AIDA64, and the clocks were down to 1.9 GHz, which seems strange.
I restarted the machine, run the test again, and saw the clocks jumpint from 1.9 to 4.1 GHz (usually ran at 3.9-4.1).

In the meanwhile HWMonitor was showing well above 100+ degress in Celsius. So that's the faulty one.
Picture
Thinking, I just realized there were moments in games while playing, suddenly my framerate got 1/3 of what it used to be.
Usually a restart fixed it.
What makes the clocks drop so dramatically?

 

[krzyimprt]

So far, I have run a stability test with AIDA64, and the clocks were down to 1.9 GHz, which seems strange.
I restarted the machine, run the test again, and saw the clocks jumpint from 1.9 to 4.1 GHz (usually ran at 3.9-4.1).

In the meanwhile HWMonitor was showing well above 100+ degress in Celsius. So that's the faulty one.
Picture
Thinking, I just realized there were moments in games while playing, suddenly my framerate got 1/3 of what it used to be.
Usually a restart fixed it.
What makes the clocks drop so dramatically?

Enable high performance mode under power options in windows. Then it should run at a constant 3.9GHz.

 

[bigbali141]

It is enabled ever since I have the machine.