Well, my i7-870 is offline again, same "crash and stuck on BSOD in 30 minutes or less after I leave work, but somehow run all afternoon without an issue" bullshit. I even took a video card out of it today (now I'm glad I did). Taking 100MHz off tomorrow and seeing what it chooses to do after that.
My FX-8150 also had a BSOD (one of those extremely rare 1000007e ones that my Vista based DV5000 clunker also got yesterday) but recovered like any normal computer.
I also FIXED the CMOS issue on my A8N32-SLI Deluxe (which has now crashed as well because I was a dummy and overclocked it before I left, unless it somehow died or something). It's a long story, but here are the key details:
-It's an old 32 pin PLCC package CMOS chip. I popped it out of its socket and found the datasheet online. I looked for the ground and voltage pins (there are 2 3V pins that are linked as well as a ground).
-Took my $3 multimeter, put it on the diode setting and checked for continuity. The ground corresponded with the battery but both +3V pins seemed "intermittent" e.g. I tested them once but for some reason I couldn't get anything out of them after that. I also removed the CLR CMOS jumper as it didn't seem to have any effect on anything (though maybe it solved my issue).
-Decided to take a twist tie with both ends stripped and made a ghetto wire connected from the + contact on battery holder to one of the VDD pins, reseated the CMOS chip and put the battery in.
-Battery voltage rapidly dropped to 1 volt, indicating a large load that I assume exceeded the battery's capacity. System powered on alright, and was still losing its CMOS settings, so I decided to remove the ghetto jumper completely.
-Tested battery again, had risen to near 3V so I just threw it back in. Powered on, set the time/date, and killed/unplugged the system to pull another rig to put my GTX 465 into from the flaky i7-870 system, and tested that system for 20 minutes.
-Plugged the 939 system back in after 30 minutes, entered BIOS and the time/date were correct (I was like WTF). In disbelief, I unplugged it for another 15 minutes and it is still holding settings like a champ, even without the jumper cap on the 3 pin CLR CMOS header.
So, now that my CMOS seems to be working fine (but the dead CPU temperature sensor and DC memory problems still exist, although I set it for 133MHz and increased HTT to 260MHz and it works) I am happy until I screw it up and have to somehow clear CMOS. Just need to put the multiplier back to 9x and it should be good.