I upgraded my Athlon II X3 435 and MSI mobo to the Bulldozer FX-6200 and Gigabyte 990XA-UD3. I only had this working for hours as it crashed every time it was running. There was no consistency to the crash, it would not happen after a certain time or a certain load. The temperatures were being lively monitored and it hovered between 25-35 for the most part, never got above 45.
I updated the BIOS, had many issues where I was forced to reinstall the OS. I had difficulty with this for a long time but a chkdsk (which returned no errors) managed to resolve the issue.
Since then, I did have one boot where it crashed at the Windows logo, but it has ran smoothly, besides the crashing. Like I say, there is no consistency to when it cuts out. No timing or temperature or load consistency whatsoever, it could just as easily crash when running at 1% load, 23 degrees and having just booted up. Or an hour later under full load with high temps.
There are a few peculiarities in my setup and the issues i've had that I think should be of note:
Firstly, the motherboard takes a 20+4pin main motherboard power lead, and an additional 8 pin socket for the CPU. The 8 pin socket currently only has a 4pin connector as I need to upgrade my power supply. I have checked this online and it seems this should make a major if any difference.
The temp fluctuation is surely odd? To fluctuate between24/35 degrees celsius in less than a second is nothing ive seen before? I know some people have mentioned that AMD's chips arent entirely flat and dont make 100% contact with the heatsink/fan?
Also, the one thing I do notice, every time I get the chance, is that the CPU clock is at 3800 when it crashes, which gave me the idea that maybe I have a setting disabled/enabled in BIOS that prevents my new CPU hitting its 'turbo speed' of 4100 and leads to the crash? Is that a realistic possibility?
The PSU is a 500W ATX and beyond the above components I only have 2 HDD's and a GTS 250 so nothing demanding that it cannot handle. The disk is 2 years or so old but I do not think it is that as it was running fine until all these changes.
Please advise what I can do and what any solutions could be?
Also, all temps (GPU, assembly, chipset) are reasonable, so the temps are real.