Worried about VCore - need help

[KMaN21]

Hello folks,

Noobie here seeking the help of the experts on this forum. I just overclocked my i5 6600K to 4.2GHz (first ever overclock for me & want no more than 4.2) and changed my CPU VCore in BIOS to 1.25. However when I open HWMonitor it shows my VCore as being 1.6 (!) as u see below.



In CPU-Z I got this ...



And in HWinfo64 I got this ...



Which one is real? I am worried that 1.6 is the one thats real even though BIOS shows vcore at 1.25

I appreciate any help guys.

 

[MNMadman]

Note all of the voltages listed in the group with VCORE in HWMonitor -- they are ALL wrong. Your system wouldn't be working at all if those were correct.

FYI: The VIDs you have in the screenshot of HWinfo64 aren't the actual voltage supplied to the CPU. The actual voltage is a single number listed lower with all of the rest of the system voltages. That's the one you should pay attention to.

HWinfo64 is what I use, as it has proven to be the most accurate of the monitoring programs for me.

Also, I suggest trying a lower voltage than 1.25v for that 4.2GHz overclock. Try it at 1.2v and see if it's stable. My 6600K is an average overclocker, and it can do 4.2GHz at 1.2v.

 

[KMaN21]

Note all of the voltages listed in the group with VCORE in HWMonitor -- they are ALL wrong. Your system wouldn't be working at all if those were correct.

FYI: The VIDs you have in the screenshot of HWinfo64 aren't the actual voltage supplied to the CPU. The actual voltage is a single number listed lower with all of the rest of the system voltages. That's the one you should pay attention to.

HWinfo64 is what I use, as it has proven to be the most accurate of the monitoring programs for me.

Also, I suggest trying a lower voltage than 1.25v for that 4.2GHz overclock. Try it at 1.2v and see if it's stable. My 6600K is an average overclocker, and it can do 4.2GHz at 1.2v.

Thank you Mad, that 1.6 V got me worried. I did what u just advised and lowered the cpu Vcore to 1.2 in BIOS and did a quick 10 min stresstest with AIDA64. These are the temps I got, are they normal for 4.2 GHz @ 1.2 V ?



How long should I stress test my current with the levels I just mentioned (10 min is very little) & is AIDA64 sufficient for that test?

Btw your right Mad, HWInfo64 showed me correct I think



Is that the number I missed in my first post that I am supposed to watch out for? Because if so, it shows 1.2 which is exactly what I put in BIOS.

 

[MNMadman]

That's the correct number to be looking at in HWinfo64 for CPU voltage, yes.

And those are good temps, though I'd get rid of HWMonitor completely and use HWinfo64 for all of it. It has all of the temps, voltages, speeds, and usage numbers for the CPU, mainboard, RAM, GPU, and drives.

I'm a stickler for system stability, so I use several programs in addition to normal usage to determine stability.

Using AIDA64...
CPU only: CPU, FPU, and cache checked. Minimum eight hours to test a final overclock that you plan to run every day.
Complete system: Everything except Local Disks checked (yes, GPU should be checked too) just to make sure everything is working well together. Minimum eight hours.

Yes, I do both for eight hours each. I set them to run while I'm at work. Plus I test with Prime95, x264, RealBench, and OCCT for certain lengths of time before declaring something stable. But I'm an extremist. Most people run a test for one to four hours and call it good.

 

[KMaN21]

That's the correct number to be looking at in HWinfo64 for CPU voltage, yes.

And those are good temps, though I'd get rid of HWMonitor completely and use HWinfo64 for all of it. It has all of the temps, voltages, speeds, and usage numbers for the CPU, mainboard, RAM, GPU, and drives.

I'm a stickler for system stability, so I use several programs in addition to normal usage to determine stability.

Using AIDA64...
CPU only: CPU, FPU, and cache checked. Minimum eight hours to test a final overclock that you plan to run every day.
Complete system: Everything except Local Disks checked (yes, GPU should be checked too) just to make sure everything is working well together. Minimum eight hours.

Yes, I do both for eight hours each. I set them to run while I'm at work. Plus I test with Prime95, x264, RealBench, and OCCT for certain lengths of time before declaring something stable. But I'm an extremist. Most people run a test for one to four hours and call it good.

You have been very helpful Mad, appreciate it very much. Thank you.