Em...Increased Voltage and temps have gone down...

beany_bot · 15 Dec 2015 at 21:32

So I went from 1.21v to 1.22v after a BSOD.
My temps have dropped 3 degrees.

?

Borealis · 15 Dec 2015 at 21:36

???

What BSOD did you get? Are your voltages set to offset or manual?

beany_bot · 16 Dec 2015 at 07:27

Borealis said:
???

What BSOD did you get? Are your voltages set to offset or manual?

They are manual. Not sure what the BSOD was. Happened a while ago. Basically I was just tinkering and went from 1.21 to 1.22. Seems my temps in game are ever so slightly lower?

beany_bot · 16 Dec 2015 at 08:26

OK just ran some tests. In stress testing 1.22v (as you would expect) actually occasionally spikes slightly higher temps than 1.21v. But last night, in game for hours, 1.22v was defo 3 degrees cooler than 1.21v. Meh. Not a problem just odd.

Borealis · 16 Dec 2015 at 08:44

As long as the temps are safe, accept it as an oddity. 1.22V is safe and isn't going to damage your chip.

Frank Jeagar · 16 Dec 2015 at 09:14

beany_bot said:
OK just ran some tests. In stress testing 1.22v (as you would expect) actually occasionally spikes slightly higher temps than 1.21v. But last night, in game for hours, 1.22v was defo 3 degrees cooler than 1.21v. Meh. Not a problem just odd.

Yeah, as long as they are safe I wouldn't worry about it.

The difference could be due to anything or a combination of things you wouldn't have noticed, like a small drop in room temperature.

Do you know what your load line calibration is set to? It might be that at 1.21v, to meet the demands of the CPU, LLC was pushing the voltage higher than that, but at 1.22 it's stable and LLC isn't increasing it as much?

tyler_jrb · 16 Dec 2015 at 09:18

Probably the difference in room temperature? 1.22v isn't much anyway and 1.21-22 wouldn't be noticed temp wise.

Frank Jeagar · 16 Dec 2015 at 09:23

tyler_jrb said:
1.22v isn't much anyway and 1.21-22 wouldn't be noticed temp wise.

That's what I was thinking, I doubt that voltage increase would impact much at all.

Disco_P · 17 Dec 2015 at 15:57

0.01v is such a minor increase in voltage and 3C is within margin of error for CPU temps.

I've always found cpu temps to be all over the place with a range of 10C between cores and temps changing a couple of degrees every second. Plus the lower the temp the less accurate the measurement as the temp probes only know when the CPU is to hot (around 100C) and the software uses an algorithm to work out the current temp.