Random Freezes

[SeaFoam]

So my 920 has been running at 4.2 since I got it, I don't use it that often but temps never get high and it has been trouble free.

Well the other day I noticed it would start getting blue screens and doing crash dumps then restarting. I backed the OC to 4.1 and haven't had a crash yet. Do you think it's something going on with the board/chip or maybe a stick of ram going bad?

 

[bigbadgreen]

I could be a lot of things. If you're sure it's not the overclock, especially after backing it down, there are a few other things to check. First back the oc back to stock speeds, see if that helps. It may not be the cpu but the north or southbridge not liking things. Makes sure all of your cooling is working for the chipsets as well. If everything checks out then it's time to start checking the RAM. There are several benchmarking programs that should be able to provide you with the info if a stick is going bad. Next thing to check is the power supply. the easiest thing if you can is to replace it with another power supply that you know is working and can handle the load. If you don't have another one then you can get some voltage monitoring software and put it under load and see if anything drops or spikes. You can also use a voltage meter to test each connection although it's tough to load down a power supply and also test it with a votlage meter.

If all still appears good it's time to look over all the connections and wires inside the case to make sure none have been pinched and are grounding out or are loose. If all still good it's time to start doing the hardware dance and only powering up essential conponents until you find one causing an issue. Also if you have updated any drivers it may be worth it roll back and see if that helps.

It could be simple and an easy fix or you could be pulling your hair out.

 

[Gorilla]

Probably not the answer, but in my case I discovered that one of my ram slots was malfunctioning. I moved the stick of ram to a different slot and bang, the blue screens went away. Interestingly, this wasn't a problem unless I overclocked. The only downside is that now I am running my memory in single channel. Someday I will have to RMA my motherboard, but not today.

 

[SeaFoam]

Well, problem is if it's a slot then the board would have to go back - I am running 12gb. It's just so strange that it happened out of nowhere, nothing has changed - it's in a clean environment, no dust inside.

I need to put the new bios on the board anyway, I know the chip will go higher than 4.2 so I guess I will just dick around with it tomorrow and see what I come up with.

 

[kidgonesmall]

Run memtest, see if anything goes wrong.

 

[BillParrish]

Go into the settings of your operating sysem and uncheck/tell it to NOT restart on critical/systemerrors you will then get a BSOD and the system event logs will give you a clue as to what is causing the problem. At least you should be able to determine if it is a software issue with something in the event logs or a hardware issue with only a general error msg or no error at all.

And/or back your OC off a fair amount and if the problem goes away you can be sure its hardware/heat/stress related and start troubleshooting your bios settings and hardware.

 

[neybis]

if backing your oc down to 4.1 fixes the issue you can be certain that its your overclock and not anything else. if the issue is software related, running your rig completely stock will not fix the issue. if the issue is hardware related, it is only defective at your higher clock speeds which according to the manufacture will probably not be labeled as defective. I would set your entire rig at stock settings and prime95 for 24hours see what happens. may not be a bad idea to memtest 2 passes before continuing with windows stability testing in fear of os corruption

 

[troisanh]

Whats your settings (volts) OP, running 12gs ie all mem slots filled required extra volts. 6g was much easier on memory controller.

 

[SeaFoam]

if backing your oc down to 4.1 fixes the issue you can be certain that its your overclock and not anything else. if the issue is software related, running your rig completely stock will not fix the issue. if the issue is hardware related, it is only defective at your higher clock speeds which according to the manufacture will probably not be labeled as defective. I would set your entire rig at stock settings and prime95 for 24hours see what happens. may not be a bad idea to memtest 2 passes before continuing with windows stability testing in fear of os corruption

It was running fine though - for 6 months at 4.2 (nothing has been changed, always been 12gb) - then just started doing this... Actually it had another problem where it disabled the ethernet port then today it locked up, the mouse would move but the cursor just sat spinning.

 

[neybis]

It was running fine though - for 6 months at 4.2 (nothing has been changed, always been 12gb) - then just started doing this... Actually it had another problem where it disabled the ethernet port then today it locked up, the mouse would move but the cursor just sat spinning.

just bc it was stable at those speeds for 6 months doesn't mean it has to be stable at those speeds today. electronic components go through wear and tear when being overclocked and sometimes you can only achieve full functionality by reducing everything to stock settings. in intels spec sheet it says something to this effect.

hate to bear bad news but if ur 100% stable at 4.1, your hardware is probably no longer capable of running at 4.2 with your current settings. I recommend tweaking your oc or finding out if it's your ram that's causing issues as it is much easier than a CPU or mobo to replace

EDIT: might be a PSU issue

 

[cory1234]

I doubt it is a psu issue. I'm guessing your motherboard didn't like the temps that you were running it at for that long. I would back down the oc and lower the volts. Run prime95/linx for a good 8-12hours.