Computer would shut down when my i7-2600k was overclocked beyond 5ghz

[m8x776]

Seems pretty simple at the surface. However, I would get no BSOD at all. I was able to get the PC to pass ten passes of IBT at max and I did 10 rounds of it. I was running at 5.1ghz (50multi X 103) It would only shut down during Prime95 or randomly when I was watching YouTube videos. Everything is fine just as long as I don't exceed 4.9ghz, and I have to have the multi at 48x.

It didn't matter if I had the multiplier at 48x or 50x when I was overclocked to 5ghz or above. The setup would BSOD if I had the core voltage set to anything below 1.43v when I was overclocked to 5ghz.

Sadly, I had to give the motherboard and processor to my daughter so no OCing this setup again.

My specs were different than what's in my profile.
Core i7-2600k
Asus P8P67 WS Revolution
8gb G.Skill Ripjaws X DDR3-2133 CL11
EVGA GTX 570 ( also had a PNY 1gb GTX 460 as my primary card while doing this)
Corsair HX1000
XSPC Rasa 750 RS360
Corsair Performance 3 128gb

I did have a fan aimed at the VRM

I did have the VCCIO set to 1.2v, spread spectrum off, etc. I also disabled the Marvell storage controller. I tried with the C1E and the other enabled and disabled but I would still have the same problem.

Not looking for exact answers but guesses since I can't really go back and OC my kids PC. Do you think it could be an issue with the power supply?

 

[HugoStiglitz]

a total system shutdown is usually only caused by HEAT.

the more you overclock the more power u need (hence needing 1.43 above 5ghz) the more power there is the more heat is produced.

the 1000w psu should handle it with no problems.

 

[m8x776]

Well I always had coretemp going and I never broke past 71c, even in IBT. I wondered if the VRM were still getting too hot. Despite the fact that I had a fan blowing air on them.

 

[Energy96]

I'm not sure I see what the issue is here. Not all chips are the same, even from the same wafer. Just because one person gets to 5ghz stable doesn't mean everyone else chip will also. Anything over 4.5 and truly stable is lucky really. 5ghz is pretty rare and probably not a good idea to run at even if it was mostly stable as the life would be significantly reduced. Lots of people can do 4.8 and claim stable, and might run fine most of the time but still suffer from a random failure.

I run at 4.3 even though I was able to do 4.8 just fine. It didn't really make much of a difference in anything outside of benchmarks so not much point in putting unneeded stress on the chip for no reason. 4.3 is a nice sweet spot that runs flawless even at stock voltage.

 

[m8x776]

Yeah....Only 10% of them could hit 4.8ghz or so i had read. This was long before the cherry-picked 2700k came out. I thought I was special because I was stable at 4.8ghz at 1.37v. Well, at First I could OC at 4.8ghz at 1.34v with the old BIOS. After I updated I needed .03v to stay stable at 4.8ghz. I was running with an H70 at 4.8ghz and the highest it ever got was 71c. IDK, I was pleased as punch and I though I could hit 5ghz easy with some good ol W/C, so I bought the XSPC kit.

Not bragging or anything, I just got lucky. I had other processors that must have barely made the cut and were poorer than average overclockers.

 

[aqe040466]

Icing 4.7-5.0 ghz don't make much difference in performance. Too much vcore voltage will shorten the life of the CPU. So if you are stable at 4.7 ghz I think you are good. Going at 5.0 or higher for me is just bragging not practical even if you are a heavy gamer. Gaming performance will be attain with a high end video card. High end video cards will result in higher fraps(frames per seconds).