Originally Posted by bluej511
I think hotsocks meant does pll actually change the reading since thats how it seems to be worded to me, ie does changing PLL actually change temperature reading that makes it go thru the roof and thats far from fake. Considering we cant physically measure cpu temps this is how its measured thru SOFTWARE. So yea changing PLL makes the temps go thru the roof but its the READING that does not the actual temperature, so the reading isn't fake the actual temperature however is.
Its like saying yea the soc vcore is fake because its reading in software.
I heard just to leave it at 1.80 or your temps go thru the roof, real or fake?
Hotstocks has asked do temps go throw the roof, is it real or fake, the temperature increase. It really doesn't matter at this point if he asked is the reading vs actual temperature is real or not. You'll see in a moment why I say that.
It´s fake in my opinion. If your CPU under full load and you raise 1.8V PLL by 0.01V you will notice an immediate jump in the temperature by 1-2 degrees, this can´t be real.
Ramad states the reported reading must be fake, as he believes actual temperature did not increase. So he has covered the "bases", reading / actual temperature.
I have a R7 1800X on order.
When my mobo is within case it's difficult to mount a temperature sensor on plate of HS for me. I will do this when swapping CPU. I may see if I can get one on the side of processor as don't wish to modify HS base to allow temp sensor between HS base / IHS. The other thing that is possibly viable is mounting a sensor on the back of CPU, then feed I'll the wire through a hole which is present in the socket and then get a measurement.
Another angle which all can do, is use the CPU socket sensor, this has no skewing AFAIK, so people adjusting PLL could use that as reference point for before and after.