Help Overclocking 4770k

[SabrToothPanda]

Hi,

 

I'v been struggling to find any stability when overclocking my 4770k. I'v read many online tutorials and watched videos but cannot come up with a solution.

My Specs:

4770k

Asus Sabertooth z87 motherboard

Corsair dominator 1866 DDR3

Asus 1080

Corsair H850I PSU

 

Some Things to Note:

My system posts and enters the operating system at 4.6 at 1.2volts, but will crash with stability testing. Some weird things that I've noticed is that as soon as I go past 4.1Ghertz, I'll get rounding errors in prime95 within the first minute (no increase in voltage has fixed it,) and as soon as I raise Vcore past 1.275volts my computer will bsod when there's any load on the cpu.

 

My most stable settings so far where, 4.5 core @ 1.25 volts and Cache 4.4 @ 1.245volts with ram on xmp (1866) and was able to get 24 hours on Intel's Etu stability test, but two days later it crashed during a virus scan, and continued to crash ever since.

 

Any help is appreciated, I'v tried everything I can think of.

 

 

[GDRRiley]

try 4.7 at 1.3v

[aisle9]

5 minutes ago, SabrToothPanda said:

Hi,

 

I'v been struggling to find any stability when overclocking my 4770k. I'v read many online tutorials and watched videos but cannot come up with a solution.

My Specs:

4770k

Asus Sabertooth z87 motherboard

Corsair dominator 1866 DDR3

Asus 1080

Corsair H850I PSU

 

Some Things to Note:

My system posts and enters the operating system at 4.6 at 1.2volts, but will crash with stability testing. Some weird things that I've noticed is that as soon as I go past 4.1Ghertz, I'll get rounding errors in prime95 within the first minute (no increase in voltage has fixed it,) and as soon as I raise Vcore past 1.75volts my computer will bsod when there's any load on the cpu.

 

My most stable settings so far where, 4.5 core @ 1.25 volts and Cache 4.4 @ 1.245volts with ram on xmp (1866) and was able to get 24 hours on Intel's Etu stability test, but two days later it crashed during a virus scan, and continued to crash ever since.

 

Any help is appreciated, I'v tried everything I can think of.

 

 

You're not really pumping 1.75v through your CPU, right? That was a typo?

 

Try 1.27v at 4.6GHz. That should be more than enough--far more than enough--to stabilize that clock speed. From there, you can either push forward to 4.7 (1.3v is the safe max for Haswell) or dial down to see how low you can go on 4.6.

[SabrToothPanda]

Oops, yes I meant to say 1.275V. Ill try 4.6 at 1.27volts.

[Daveeede]

I had similar issues on my 4790k which is why I usually run stock on my rig. Haswell is not the best for overclocking. I remember being stable-ish at 1.27V @4.6 GHz. It would still crash about once a month.

[SabrToothPanda]

Ya, I'm starting to think the same. Anyways, I tried 4.6 at 1.270Volts and it bsod on the first run of cinebench. Then I pushed it back to 4.5 at the same voltage and it can get through cinebench. But when trying to stress on p95, it gives me the error (Fatal Error: rounding was 0.5, expected less than 0.4.) within the first couple seconds. I did some experimenting and figured out that anything past 4.1Ghertz will give me this error, even when I'm stable in other stress test like Aida and Etu.

[SabrToothPanda]

For some odd reason setting each core sperately rather than syncing all cores seems to allow me to go above 4.1 without rounding errors. Currently I'm stable at 4.2 this way, and I'm slowly increasing the clock of each core at a time.