There's something wrong with my system, but it's fairly subtle. For the last few months I've just been living with it occasionally rebooting/freezing, as I haven't had time to fight with the damned thing.
i7 920, 12gb corsair dominator, gigabyte UD5. 860W PC P&C, watercooled.
It passes IBT and prime at stock, and at 3.6ghz, but occasionally crashes under general use regardless. It's borderline stable (IBT and prime95) at 4ghz, but again occasionally crashes under general use.
It's previously been stable at 4.5ghz under reasonable voltages, and cpu temperature is not limiting. Motherboard/ram temperature might be, though there are fans blowing cool air over both. It's probably not the hard drive, in that windows & linux can both crash when running from a range of drives. It's possible the installations are all corrupt.
My gut feeling is that I've damaged the imc on the cpu by pushing too much voltage through it, and unobservantly running the cpu at 90 degrees for a month or so (blockage in tubing). Diagnosing this with any degree of certainty is proving difficult however, as the instability could be in the ram or motherboard (or indeed OS) instead. I'm starting to think the best approach is to replace a component, and retest, but that's an expensive game to play and I'm currently broke. Otherwise I can RMA components, but as they haven't failed catastrophically, I don't expect them to fail basic testing.
So, any ideas? Just living with the occasional crashes, and running at under 4ghz seems the only sane way forward at the moment. I'm hoping a reader will see something I've missed. Cheers
i7 920, 12gb corsair dominator, gigabyte UD5. 860W PC P&C, watercooled.
It passes IBT and prime at stock, and at 3.6ghz, but occasionally crashes under general use regardless. It's borderline stable (IBT and prime95) at 4ghz, but again occasionally crashes under general use.
It's previously been stable at 4.5ghz under reasonable voltages, and cpu temperature is not limiting. Motherboard/ram temperature might be, though there are fans blowing cool air over both. It's probably not the hard drive, in that windows & linux can both crash when running from a range of drives. It's possible the installations are all corrupt.
My gut feeling is that I've damaged the imc on the cpu by pushing too much voltage through it, and unobservantly running the cpu at 90 degrees for a month or so (blockage in tubing). Diagnosing this with any degree of certainty is proving difficult however, as the instability could be in the ram or motherboard (or indeed OS) instead. I'm starting to think the best approach is to replace a component, and retest, but that's an expensive game to play and I'm currently broke. Otherwise I can RMA components, but as they haven't failed catastrophically, I don't expect them to fail basic testing.
So, any ideas? Just living with the occasional crashes, and running at under 4ghz seems the only sane way forward at the moment. I'm hoping a reader will see something I've missed. Cheers