i5 6600K / i7 6700K OC results

[AmatuerOC]

Hello OCers, this is my first post on these forums. I recently built a custom system revolving around an I7 6700K + GA-Z170X gaming 7, paired with a corsair H115i CPU water cooler and WOW I am so happy with the system. Having upgraded from a very old E8600 that gave me many good years OC'd too 4.0Ghz but was definitely ancient architecture. Anyway I am here of course to share my results for my 6700K, pics will be forth coming if requested.

I7 6700K @ 4.6Ghz - 1.26 Vcore fully stable. I can push it too 4.8Ghz @ 1.28Vcore. Did I win the silicon lottery? I bet this is a 5Ghz chip, but I don't want to fool around and push my luck. Temps are topping out around 70C, probably because of the ambient temperature here (80F). I will answer any questions regarding my BIOS settings if you have any.

Oh ya, I forgot to mention RAM because I know that can be important (it's what limited my E8600 OC). I am using G.Skill Ripjaws V 16GB dual channel DDR4 3000.

 

[wingman99]

Was 1.26v your peak voltage at 4.6GHz, also what stress test did you use to measure your peak voltage?

 

[AmatuerOC]

Yes peak voltage is 1.260v, it bounces from 1.248v too 1.260v max. I am Using Intel extreme tuning utility stress test. CPUID for voltage monitor.

 

[Kenrou]

Intel Extreme tuning is garbage, i will not use it even for browsing as it will crash as soon as you put a little more heavy load on it.

Try Realbench : http://rog.asus.com/19932016/news/realbench-v2-43-new-version-available-now/

 

[Woomack]

I've never had issues with XTU stability on stable PC. It's using Prime95 in older version to perform stability tests ( or at least you can see it on process list ). I had more issues with realbench as it requires additional applications to run stable.
Anyway XTU isn't the best tool to check stability. What is weird is that benchmark is showing errors faster than the stability test itself.

 

[Kenrou]

I used it twice, once in Win8.1 and other in Win10 and never again unless its fixed, ran the test and it marked the system as stable then got repeated crashes every time i played a game or used something CPU heavy like Seti@Home (which runs perfectly fine on my P95 non-AVX setting). Even crashed once halfway through a youtube music video, had to hard reset

 

[EarthDog]

I won't use a stress test for browsing either...

 

[Kenrou]

 

[AmatuerOC]

Well XTU works for me, I heard it recommended several times against other torture tests. As they go above and beyond real world performance. My system is stable fully on all the games I have tested it on so far. I don't do any more extreme types of processing on this PC so maybe with encoding or something this would be an issue but that's moot for me personally. I was kind of hoping for some feedback on my results, not on the stress test anyway...

4.6Ghz @ 1.260v max is extremely good isn't it? I still have a lot of headway I could try and push for 5Ghz when the ambient temperature drops here right?

 

[Kenrou]

Your results seem too good to be true if you take my meaning, [email protected] and [email protected] ? the voltage difference between the 2 is too small ?

 

[AmatuerOC]

I can't confirm the 4.8Ghz Vcore max, it's very likely it could top out higher than 1.28. I only ran very short stress test @ 4.8ghz and did no real-world performance testing. So the 4.6Ghz results are all I can fully confirm. When my ambient temperatures come down I will do some more testing at 4.8 and above and will post results, might be a few months until I can do that.

 

[Kenrou]

With it topping at 70c you have 20c headroom still, no need to wait for the temps to go down ?

 

[wingman99]

4.6Ghz @ 1.260v max is extremely good isn't it? I still have a lot of headway I could try and push for 5Ghz when the ambient temperature drops here right?

Depends on the stress test, some motherboard increase voltage with a higher stress test like prime95.

 

[Woomack]

Maybe not i5/i7 but I thought I will post anyway ...

Pentium G4400 with 2x8GB TridentZ 3200C15 @ 4333 on MSI Z170I Gaming Pro AC




and little update at higher CPU clock ... I just had to add FX8150 as reference

 

[Leegit]

Here are my Prime 95 Small FFT results from the i7 6700k thread for cross-reference: http://www.overclockers.com/forums/showthread.php/772650-OC-i7-6700K?p=7951976#post7951976

Next, I'm going to turn off HT and repeat the testing for a comparison in voltage/temperature/power

 

[SalemSaberhagen]

I'm only half joking here but would some of you 6600K owners PLEASE buy a 1070 based card and benchmark it EVERYWHERE!

I swear to cows I am the loneliest benchmarker on the planet.

 

[SalemSaberhagen]

Oh and please forgive my shoddy documentation but last week I worked 6 full days, next I work 7.

The other day I got bullish and went for it, the paste settled, the flow was established, a new hotter videocard was there to push.

Current overclock without a single solitary crash in any bench or ... I've only had time for some GTA5 but the game I'm playing lol...



It's marginally wonky, sometimes my Ethernet wont start, otherwise titanium stable!

Edit: AIDA is showing true Voltage

 

[Kenrou]

I'm only half joking here but would some of you 6600K owners PLEASE buy a 1070 based card and benchmark it EVERYWHERE!

I swear to cows I am the loneliest benchmarker on the planet.

But i like my Strix 980Ti

4.8ghz at 1.231v ? that must be a hell of a golden chip to get that speed at near stock voltages, tried going higher ?

 

[EarthDog]

1.4V... check the AIDA voltage.

 

[Leegit]

What other micro-tweaks can I make to get a little more stability?

I can reach 4.7 GHz @ 1.33V with HT on and 4.7 GHz @ 1.30V with HT off but I can not reach stability long term at 4.8 GHz.... Even if I raise my voltage to 1.39-1.40V with HT turned off it will BSOD after awhile?? My temps are around 80C at 4.8GHz so I still have some thermal headroom... I know all chips are different but I figured if I can do 4.7 GHz @ 1.30V then I would be able to make 4.8 GHz.