I7 2700K BSOD after 30 minutes of Prime95?

Toggle sidebar Toggle sidebar
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530
Hi

I just replaced my old board after it died on me but now I can't replicate the overclock I once had. I was stable at 4.3 ghz using 1.28v but now I can't get a 4.2 OC to run in prime for over 30 minutes even at 1.3v. Does the board make that much of a difference or could it be something else?

Also, I was using a Corsair H70 liquid cooler (I believe) with my old board but now I'm using a Hyper Evo 212 push and pull set up because the back plate doesn't fit on my new board (mini itx to micro atx) so could that play a part in it? Idle temps are high 20s - low 30s and it goes up to 60-70c during prime

I haven't experienced any crashes during normal usage but Prime crashes so I'm concerned.

Old board: EVGA Stinger Mini ITX
New Board: ASUS P8Z77-M
CPU: I7 2700K

Thanks.
 
Solution
larkspur


Personally I wouldn't go over 1.35v. That board might have problems even before that. Many people would consider 1.4v too high but you can read about plenty of people that have exceeded that - personally I think that's a bit much but I am a conservative OCer. You'll probably be limited by temperature based on your cooling solutions.

Here's what Tom's did with an i7-2600k (and of course those specific settings are just for reference, your results will vary. Your goal is to get a good clock with the minimal voltage needed to keep the chip 100% stable at that clock...
Sort by date Sort by votes
nukemaster

nukemaster

Titan
Nov 23, 2004
19,626
6
74,965
Maybe the voltage system is just not as clean.

Your temperatures do not seem to be an issue. Memory compatibility may also play a role, but I always figured it would be less of an issue with the memory controller in the cpu. It may still be worth ensuring your memory settings are exactly the same. Maybe running looser timings when testing just to ensure it is not a memory issue.

Some users report different cpu sockets are better as well. If this is true maybe the current one does not make as good of contact with the pads on the cpu. You could try to remove the cpu clean the pads with an eraser(if they look like they have become dirty or have some oxidization on them) and give it another go.

It is also possible that running higher clocks since the chip was new have caused it to slight degrade and now it does not pass testing like it once did(not that I think your voltage or temperatures are high for a 2700k).

I am surprised an mATX board would limit cooler installation.
 
Upvote 0 Downvote
larkspur

larkspur

Distinguished
Mar 10, 2011
2,252
111
20,640
Based on a quick look at both those boards, the evga board is definitely the better OCing board. The Asus isn't a good OCing board. Strange that the backplate would fit the mini-ITX board but not the mATX board. Anyway, with that Asus board if you want to OC with that board pay attention to the temperature of the VRM (these are often reported as some variation of 'system' temp). It simply doesn't have robust voltage regulation vs the evga board and it sounds like you are experiencing that first-hand.
 
Upvote 0 Downvote
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530


I'm running XMP so I assumed it would put my memory in an optimal place right? I am, however running mismatched RAM (2 x 4gb Corsair Ballistix Tactical and 1 x 8GB Corsair Value) so maybe that is the issue but the BSOD states that "A clock interrupt was not received on a secondary processor" which is a sign of an unstable OC right? So i'm unsure why my OC is unstable now when it was stable at a higher clock with my old board.

I'm thinking of going above 1.3v but I've never had to that before so I'm not sure about it. I could clean and reseat the CPU but isn't that something you usually do when your temps are abnormal?

And regarding the liquid cooler bracket, I'm assuming it did come with a micro ATX bracket when it was bought but I got it from an old friend and it only had the Mini ITX back plate with it so i'm probably gonna buy a new bracket and back plate for it.
 
Upvote 0 Downvote
nukemaster

nukemaster

Titan
Nov 23, 2004
19,626
6
74,965
I was talking about cleaning the bottom of the cpu, not the heat spreader.

The brackets should be the same for sure. I have a mITX board and the cooler bracket works on all boards that it has socket support for. I had to cut part of it because of backside components on my mITX board.
 
Upvote 0 Downvote
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530


I can definitely believe that. The Stinger was great for me and I would have gotten another but there are simply none ANYWHERE. Z77-Z97 1155 MATX and Mini ITX boards are prettty damn rare anyway and end up being sold for stupid prices. I got this board because it was the last 1155 Z77 available. So i'm guessing I'm gonna have to bump up the voltage past 1.3v aren't I?

The plate fits but it doesn't. It lines up perfectly and I can insert it but it doesn't stick and I don't have anything to fasten it with, if that's what it needs. With the Mini ITX board, I just put the plate on and it sticks, no fastening required.
 
Upvote 0 Downvote
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530


I thought we were never to touch the bottom of a cpu because its really easy to damage?
 
Upvote 0 Downvote
larkspur

larkspur

Distinguished
Mar 10, 2011
2,252
111
20,640

Sounds like it needs it. If you aren't already, use load-line calibration (LLC) to see if that helps.


Hmmm, the backplate should be attached to the block with some kind of bolts or screws or whatever. As I recall they usually include double-sided tape to get the backplate to stick to the back of the board so you can then attach the block without having to hold the backplate in place.
 
Upvote 0 Downvote
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530


At what point should I stop with the voltage? What's the max I should push it (if it comes to it)

And I don't see any tape on the back plate at all. Mind you, I did get the cooler second hand, so I cant say what the previous owner did to it. Just gonna get a new back plate and bracket (especially because the bracket is supposed to have screws to hold it in place and mine doesn't =/)
 
Upvote 0 Downvote
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530


Correct, my back plate is just a plastic plate with metal inserts but it doesn't stick to my board, implying that I need something to fasten it but I don't have anything like that.

I'd prefer to leave cleaning the bottom of my CPU to a last resort as I'd don't really wanna damage it and have to buy a new CPU as I just bought a new motherboard (which was quite pricey) Now I just need to find out how much voltage I should use, now that I'm out of my 1.2v comfort zone.


 
Upvote 0 Downvote
larkspur

larkspur

Distinguished
Mar 10, 2011
2,252
111
20,640


Personally I wouldn't go over 1.35v. That board might have problems even before that. Many people would consider 1.4v too high but you can read about plenty of people that have exceeded that - personally I think that's a bit much but I am a conservative OCer. You'll probably be limited by temperature based on your cooling solutions.

Here's what Tom's did with an i7-2600k (and of course those specific settings are just for reference, your results will vary. Your goal is to get a good clock with the minimal voltage needed to keep the chip 100% stable at that clock: http://www.tomshardware.co.uk/sandy-bridge-overclocking-efficiency,review-32107-4.html
 
Upvote 0 Downvote
Solution
W

Way2much

Honorable
Dec 4, 2012
35
0
10,530


Just gonna go with it. 1.35v for a 4.2 OC would be HORRIBLE in my opinion but I'll just have to test and see. My temps are already kinda high so I don't wanna go too high but I'll just have to wait and see.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom