As some of you may have seen in the Sow forums, I have had an issue with my computer simply freezing to black screen. I thought that this was due to a heat issue as it seems the CPU just ‘turns off’.
I’ve just installed a new cooler and after testing in the ATAG server last night, all seemed fined. This evening, after flying for just over an hour, with a Bf109 way out in front of me, the computer froze to black screen again. F**K :evil:
I seriously don’t know what to do. All the temps seem fine. Fans and cooler running at in high mode (I manually set them high). Nothing in the Windows Event Log. Stumped.
I’ve played FarCry4 on the computer and the same thing happened, so I know it’s not related to CLoD.
I have run that test before, just a couple of weeks ago and all was OK. But no harm in doing it again.
On further investigation this evening, seems my GPU is hitting 75 degrees when the computer shuts down. I thought GPU’S could go higher than this though?
I ran a quick search and your assumption seems to be valid.
Nvidia femini should be below 80 C, but max temp ranges seems to be in 100-105 C.
If your RAM have passed tests, you bought a new cooler which keep temps at 75 C, my next guess would be the powersupply or a faulty heat reading/faulty component on the motherboard.
Are your BIOS up to date? or does it have been updated lately?
Do a real memtest using MemTest86 if you haven’t already: http://www.memtest86.com/
The one from MS is not thorough enough to find everything. Still doesn’t necessarily mean that it’s the memory stick themselves that is faulty, could still be the MB.
I went through something similar when I built my first i7 PC. I had random crashes when playing, I don’t remember the exact order of things, but I think that at first I found some RAM problems, and changed RAM sticks, that helped some, but I still had random crashes. Then I thought it was a heat issue (and the CPU was getting a bit hotter then what is preferred), so I changed the cooler to a water cooler (simple closed system), that also seemed to help a bit, but still not completely. Did some more testing of the RAM, and might even have changed memory sticks once more, and by now the crashes was so far between that I more or less gave up until a year later or so I decided to upgrade to a i7-4770K on a Asus Sabertooth Z87 and haven’t had any more hardware crashes since then.
Funny thing (sort of), just after I had pressed send on the above post, my work PC crashed, but that was a software issue (and not due to the forum either). :roflmao:
Thank you for all the great advice. I will be following them all up. However…
This evening I took my GPU out and took it to bits to apply new thermal paste. That all went well.
Put it back in the computer and ran 3D Mark. 3 weeks ago the computer ran 3D Mark all the way through the stress tests. This evening the computer froze within 15 seconds of it starting the stress test.
So 616 Gosling paid me a visit this evening as he lives less than a mile away from me and he brought over his spare GTX 275 card. Got it working quite quickly, but the settings l had it on meant that over Dover it was less then 20fps and low level is was less than 5fps!
So instead of waiting until the new Nvidia cards in Jan, l bought a factory refurbished ATi 480x 8gb for £230 that l’ll sell on in Jan.
Mmmmm…if Ram is ok, new GPU…the problem could be the Motherboard? Or the Win system? Have you got a backup of the system? The problem is only with Clod?
No overclocking anywhere? And you didn’t get any error messages from the RAM when you did that test?
I’m afraid we’re getting close to pointing out the motherboard.
BUT! One last thing before that! What kind of PSU do you have? Is it powerful enough to supply energy to everything, and are you sure it’s working as intended. Digging back in my memory, PSU’s can give strange effects if they are about to give up. If you have another to test with, try that first.
Problem not with Clod, same in FarCry4 and 3D Futuremark. Interesting thing is that when I swapped the GPU’s over, with the new GPU I only have to hit Ctrl Alt Delete to get back to the desktop, where as with the older card I had the BSOD (Black screen of death). I do have a backup of Windows on a spare SSD, so I may run that.
Mikke…
Yes, CPU is slightly overclocked, but nothing major. No messages on the RAM at all, didn’t come up with one error. I’ve a 750W PSU, more than capable enough. In fact, the new GPU uses one less power rail.
But, I have been getting a couple of VCORE under volt warnings. And yesterday I saw a VCORE over volt warning. So now I’m being led down the path of the PSU starting to fail.
Computer shop across the road from where I work, so I may have to go buy one of those now…
:)