I need immediate help with these problems: The processor was undervolted to 1.25v for 3 years at stock, everything else is at stock. It started having problems corrupting an entire windows install. Then I ran OCCT and the system bluescreened. I raised the voltage to 1.3 and finally reverted to the factory default (auto setting on the motherboard). However, OCCT errors still persist (OCCT reports error in the cpu cores) But then I ran memtest86+ and the memory failed after a few seconds. At stock voltages, these problems don't show up during normal operation. Only in these benchmarks do they get flagged. Here are my questions: 1. Right now I don't know if it is the MB or CPU or the ram that's bad. I don't have the means to test these parts on other systems. 2. I'm not sure if it is still a good idea to "replace" any of the three components at this time. They are over 3 years old, typically operating for long hours on any day. (but at stock and processor undervolted) Now, the motherboard is an AM3 board, so that goes with the CPU. But I'm not sure if getting new RAM is the right thing to do as I can't tell if the MB or processor is damaging the ram. 3. Just as a precaution, I want to know if this causes silent data corruption on all file transfers and files stored on active disks during this time. (like music/picture/video files) My uneducated belief on this tells me that there should be some data verification process going on under the hood the entire time? What concerns me is could there be corrupt files that can't get detected in any way because the files will simply play through the errors. (like any random 1s and 0s can be read as potential wav content) Thank you.