6600k and z170 pro gaming, stable overclock settings?

[djsolidsnake86]

hi, actually i'm running this cpu at 4.4hz and 1.25v, but sometimes happen that during games there are freeze, and i must press the power button for 5-10 seconds for power off totally the pc (the restart button don't do nothing when there is this freeze)
i suspect that the problem could be some wrong settings with the overclock, what do you suggest me to do?

 

[EarthDog]

Welcome back.

Well, first, please list ALL of your hardware so we know what you're working with... this includes RAM, power supply, and video card...

Second, if you think it's the overclock, the most obvious thing to do is disable anything that is overclocked (select optimized defaults in the BIOS) and see if it still happens. If it happens again, it's clearly not the overclock but something else.

Let's start there.....

 

[mackerel]

1.25v is low for 4.4GHz for sure. I can't remember exactly what sort of voltage I had to use back then but typically 4.7 GHz is possible on decent air cooling.

 

[djsolidsnake86]

1.25v is low for 4.4GHz for sure. I can't remember exactly what sort of voltage I had to use back then but typically 4.7 GHz is possible on decent air cooling.

i can try 1.30 for 4.4?

 

[mackerel]

i can try 1.30 for 4.4?

Certainly try, that's part of the overclocking process. If chosen clock is not stable, increase voltage as long as you have thermal headroom.

Looking back at my hwbot 6700k submissions (didn't do any on 6600k it seems), it looks like I might have topped out around 1.40v and 4.7 GHz might be limited in stability depending on workload, with bench clocks in the range 4.5 to 4.7 GHz. I may or may not have been cooling limited also and those submissions didn't include a photo.

 

[Nebulous]

Hmm, I remember on my 7700K/Z270., under water I was able to max the voltage to 1.42 to get 5.0Ghz. So yeah 1.3v for 4.4 is low. I would start at 1.45v for that clock speed and start lowering the voltage and testing until it becomes unstable.

Under nominal operation (browsing/emails/etc), it ain't needing voltage, but once you start gaming it will start sucking power down and 1.3v isn't enough.

 

[djsolidsnake86]

Hmm, I remember on my 7700K/Z270., under water I was able to max the voltage to 1.42 to get 5.0Ghz. So yeah 1.3v for 4.4 is low. I would start at 1.45v for that clock speed and start lowering the voltage and testing until it becomes unstable.

Under nominal operation (browsing/emails/etc), it ain't needing voltage, but once you start gaming it will start sucking power down and 1.3v isn't enough.

1.4 is high i think, for now i stay 1.30
if i will have freeze again i will go 1.31

 

[mackerel]

Hmm, I remember on my 7700K/Z270., under water I was able to max the voltage to 1.42 to get 5.0Ghz. So yeah 1.3v for 4.4 is low. I would start at 1.45v for that clock speed and start lowering the voltage and testing until it becomes unstable.

From memory on the top end Kaby Lake was about +100 MHz over Skylake due to process changes. I also vaguely remember 1.40v being stated as a long term maximum, but don't recall where that came from.

Oh, there's also load line calibration if available. That might help also by reducing voltages drops under load, without having to increase the maximum.

 

[Nebulous]

1.4 is high i think, for now i stay 1.30
if i will have freeze again i will go 1.31

It's easier to work down on the voltage than going up. You're testing steps. It's not a long term thing. The idea is to reach the speed you want with the lowest voltage stable.

You're going to work harder using raising the voltage, but if that's the way you want to do it, all up to you.

From memory on the top end Kaby Lake was about +100 MHz over Skylake due to process changes. I also vaguely remember 1.40v being stated as a long term maximum, but don't recall where that came from.

Oh, there's also load line calibration if available. That might help also by reducing voltages drops under load, without having to increase the maximum.

Yeah I remember LLC. From top going down; 1 being overvolting and 2 being the steady, I always used #2