• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Help me OC my I5 6600k

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
4.5 on 1.300: Windows didn't even loaded, blue screen
4.4 on 1.300: Windows loged, OCCT Core 0 error
4.4 1.3 till 1.35: As above, only sometimes Core 1 not 0
4.5 on 1.35: Windows didn't even loaded.

Now trying to go from the bottom:
4.2 on 1.25
 
Is that on XMP settings then only changing the multi and vCore?
 
I used XMP only for ram and manualy change multi and vCore. All those auto OC options were giving me sick voltages, so I don't use them. 42 multi on 1.25 is stable, temps 59-63 for 20min of OCCT. Will try 43 on same voltage and if not stable, add 0.01 to it, right?

- - - Updated - - -

BTW, does it matter, and what is the couse, that core temps sometimes are aside from each other 8 degrees?

EDIT1:

Another BTW, should I turn off turbo mode?

EDIT2:

Update:

I have a lame *** CPU.. 1.300v and 43 multi, seems stable. Is there anything else I can do to lower voltage?

EDIT3:

Nah, after ~20 minutes freezed, now at 1.32 with 43 muli

I've printed Steve's Skylake Overclocking Flow Chart and there he wrote about changing FCLK and setting uncore/cashe/ring ratio, should I start doing this and how?
 
Last edited:
I used XMP only for ram and manualy change multi and vCore. All those auto OC options were giving me sick voltages, so I don't use them. 42 multi on 1.25 is stable, temps 59-63 for 20min of OCCT. Will try 43 on same voltage and if not stable, add 0.01 to it, right?

- - - Updated - - -

BTW, does it matter, and what is the couse, that core temps sometimes are aside from each other 8 degrees?

EDIT1:

Another BTW, should I turn off turbo mode?

EDIT2:

Update:

I have a lame *** CPU.. 1.300v and 43 multi, seems stable. Is there anything else I can do to lower voltage?

EDIT3:

Nah, after ~20 minutes freezed, now at 1.32 with 43 muli

I've printed Steve's Skylake Overclocking Flow Chart and there he wrote about changing FCLK and setting uncore/cashe/ring ratio, should I start doing this and how?

Sounds like a good starting point.

I wouldn't worry about the temperature spread.

No, you need turbo on to overclock.

Is your board letting the vCore droop under load?
If so, raise the LLC setting by one.

Also, thank you for going back and editing the multiple posts into one post :)
 
You can try the latest BIOS https://pl.msi.com/Motherboard/support/Z170A-GAMING-PRO.html#down-bios
I would also set power limits to the max as MSI boards are sometimes throttling under full load when standard current/power limits are set. It should be in CPU settings tab on the bottom of the list. On Z170I it was causing stability issues while overclocking.
Generally cache clock @4.2GHz should be always stable at ~1.15-1.20V. Since CPU and cache voltage is linked then it will be always higher than required for cache. Check everything at 4.2GHz cpu + 4.2GHz cache ... like full stability at 1.20V or something near and next change only CPU ratio and CPU voltage. Everything else should keep full stability.
 
On Real Bench and Cinebench there are no errors at 44 multi with 1.3 voltage, only on OCCT, I will post my logs, would be great if someone could take a look 2016-08-24-00h25-CpuUsage-CPU Usage.png 2016-08-24-00h25-Frequency-Bus.png 2016-08-24-00h25-Frequency-CPU #0.png 2016-08-24-00h25-Frequency-GPU #0 Mem.png 2016-08-24-00h25-Frequency-GPU #0.png 2016-08-24-00h25-Memory Usage-Memory Used.png 2016-08-24-00h25-Temperature-Core #0.png 2016-08-24-00h25-Temperature-Core #1.png 2016-08-24-00h25-Temperature-Core #2.png 2016-08-24-00h25-Temperature-Core #3.png 2016-08-24-00h25-Voltage-GT Offset.png 2016-08-24-00h25-Voltage-IA Offset.png 2016-08-24-00h25-Voltage-LLC-Ring Offset.png 2016-08-24-00h25-Voltage-System Agent Offset.png 2016-08-24-00h25-Voltage-VID.png 2016-08-24-00h25-Voltage-VIN0.png
 
And on the other hand. OCCT:Linpack starts it's 4th minute, and I've read it's more error friendly than OCCT:CPU
 
Maybe check other soft, Prime95 or something else. I remember that in the past some OCCT versions were throwing errors regardless if PC was stable or not. I'm not saying it's a bad soft as I'm not really using it lately but maybe errors are related to software, not hardware.
 
Tried AIDA64 becouse I though maybe it's OCCT problem, like you said, but ~6 minute freezed my computer. When temperatures in my room will get more life friendly I will try to make stable 4.4 again with different software, becouse it maybe be the point. On the other hand, real bench and cinebench don't stress so much (temperature are a lower than on OCCT).

Are OCCT and AIDA64 a good way to check if in everyday work I will met errors? I mean, I will be playing some games on highest settings and may those be equal on OCCT/AIDA stress test?
 
Back