Finalizing 4690K overclock

[nims0c]

I'm new to overclocking so I followed LTT's Devil's Canyon overclocking video guide. With CPU Core Ratio set to Sync All Cores and CPU Core Voltage set to Manual Mode, I found 4.5 GHz with 1.25V to be stable throughout an 8 hour Aida64 stress test. HWMonitor showed max temps of 77 on 2 of the 4 cores, so I don't know if I want to push the voltage any higher. 4.6 GHz was unstable at 1.25V, and 4.7 GHz even more so even at 1.275V so I kinda gave up there.

 

Should I stress test more, for 24 hours like Linus mentions?

 

Last step would be to switch CPU Core Voltage to Adaptive Mode, correct?

[runit3]

I'm new to overclocking so I followed LTT's Devil's Canyon overclocking video guide. With CPU Core Ratio set to Sync All Cores and CPU Core Voltage set to Manual Mode, I found 4.5 GHz with 1.25V to be stable throughout an 8 hour Aida64 stress test. HWMonitor showed max temps of 77 on 2 of the 4 cores, so I don't know if I want to push the voltage any higher. 4.6 GHz was unstable at 1.25V, and 4.7 GHz even more so even at 1.275V so I kinda gave up there.

Without a better cooling solution you'll hit a thermal limit fairly quickly. That being said nothing will stress your system more than AIDA or XTU, except maybe x264 encoding. You could give 1.3-1.35v a go, just keep temps under ~85C. Some Haswell/DC chips just aren't going to be able to hit those voltages without a delid, regardless of cooling solution.

 

24hr stress tests are unnecessary IMO. It would be better to run real-world benchmarks like ASUS Real Bench, play some games for a few hours, watch YouTube/Hulu/Netflix, etc. Just use your computer as you would normally and note any problems you come across.

 

If you feel the need you can swap to adaptive. No real point as voltage does not equal wattage (power use). If you were running 1.35v+ it would make more sense as voltage degradation is a thing (although arguably not even at that low of a voltage). I run all my systems at static.