My 4 year old system has run fairly reliably, with the majority of crashes being game related. However after a power cut last week it has been crashing, restarting and locking up regularly (we had a switchboard mounted surge suppresor, now dead, yet the cut also took out the dishwasher and an old amplifier). CPU: C2D E6600 MB: Gigabyte 965G-DS3 RAM: 4x1GB DDR2 800 Apacer GPU: Sapphire Radeon 6850 1GB PSU: Antec Neo HE 430W OS: W7 HP HD: Crucial M4 128GB SSD It will boot and run normally, in fact the crashes are rarely while I'm active - I gave it a thrashing in BF3 this morning for 90mins to test. When it does crash it does so slowly, with some programs (say Opera, or windows explorer) locking up while others (Firefox) continue to run. Then eventually everything locks up, the screen goes blank and it reboots, only to hang on the interrupt screen. Most of the time it crashes when I'm not at the pc and I come back to the interrupt screen. Typically it crashes within an hour of my ceasing any activity. Obviously I strongly suspect the PSU has been damaged by the power cut but I've been trying to confirm that before I start throwing money at spare parts, particularly as I'm planning on a new system anyway in a few months (although I would like to keep the old one for a home server). I've run check disk, which finds no faults. I've run Memtest86+ for 12 passes with no faults. Using Speedfan the temperatures aren't running overly hot, but the reported 12V and -12V readings are 0.77V and -16.97V respectively. These seem suspect but I've read that Speedfan can have trouble reporting accurately and I'm amazed my pc would run at all if the were correct. Unfortunately there don't seem to be any minidump files created by these crashes, although there are some older files there. Nor is event viewer particularly enlightening. Does anyone have any other sensible checks or suspicions before I get a new power supply (and a UPS this time)?