Okay, so I've been having an annoying problem with a completely new build, basically any game (or at least games I've tried so far, Eve Online, Civ V, Crysis) will sooner or later lead to a system crash (typically around 1-2 hours, though can happen within 30 minutes, can also keep going for hours on end) where the screen blacks out, makes a buzzing noise, and forces a hard restart in order to move on. Because of this, I can't get a BSoD/any memory dump (that being said, I can't seem to get it to BSoD in any circumstance, the Ctrl+(ScrLk)x2 doesn't work).
-So far I can all but rule out the RAM issues (memtest86+, 39 hours with both sticks in, 34 with one stick and like 29 with the other, no errors).
-Overheating doesn't seem to be a problem either as the temperatures never get any higher than 60C (watched them in-game with Afterburner, as well as tabbing out to an already open HWMonitor).
-Furmark on the GPU leads to only 70C at 99% and levels off there long before finishing the 15 minutes.
-Passed all basic tests using SeaTools on the HDD.
-No errors or noteworthy events show up in the Event Viewer until rebooting after crash, at which point the only errors are just pointing out there was an unexpected crash.
-Everything completely up-to-date as far as Windows Update is concerned.
-Prime95 seems concerning, shoots the cores up to 89-95C range and the overall CPU reading being 87C, but levels off at those temps. This is using the stock cooler that comes with the i5, is this within expectations for stock?
-MemtestCL also gave off some worrying stats, averages around 25 million errors in only the random blocks test, but there is apparently problems with AMD cards and this test. Still, of all the cases I've read of other people doing it, 25 million seems ridiculously high (didn't see anyone report higher than 10,000), anyone have experience with this test? Is this grounds to get an RMA?
System specs:
i5 2500k
MSI HD 7870
ASUS P8Z68-V/GEN3
2x Corsair Vengeance 4GB 1600MHz (CMZ8GX3M2A1600C9)
850W Seasonic M12II-850
WD Caviar Black (WD1002FAEX)
TP-Link 300Mbps wireless card (TL-WN951N)
HAF X
Windows 7 Ultimate 64 Bit
-So far I can all but rule out the RAM issues (memtest86+, 39 hours with both sticks in, 34 with one stick and like 29 with the other, no errors).
-Overheating doesn't seem to be a problem either as the temperatures never get any higher than 60C (watched them in-game with Afterburner, as well as tabbing out to an already open HWMonitor).
-Furmark on the GPU leads to only 70C at 99% and levels off there long before finishing the 15 minutes.
-Passed all basic tests using SeaTools on the HDD.
-No errors or noteworthy events show up in the Event Viewer until rebooting after crash, at which point the only errors are just pointing out there was an unexpected crash.
-Everything completely up-to-date as far as Windows Update is concerned.
-Prime95 seems concerning, shoots the cores up to 89-95C range and the overall CPU reading being 87C, but levels off at those temps. This is using the stock cooler that comes with the i5, is this within expectations for stock?
-MemtestCL also gave off some worrying stats, averages around 25 million errors in only the random blocks test, but there is apparently problems with AMD cards and this test. Still, of all the cases I've read of other people doing it, 25 million seems ridiculously high (didn't see anyone report higher than 10,000), anyone have experience with this test? Is this grounds to get an RMA?
System specs:
i5 2500k
MSI HD 7870
ASUS P8Z68-V/GEN3
2x Corsair Vengeance 4GB 1600MHz (CMZ8GX3M2A1600C9)
850W Seasonic M12II-850
WD Caviar Black (WD1002FAEX)
TP-Link 300Mbps wireless card (TL-WN951N)
HAF X
Windows 7 Ultimate 64 Bit