I've heard rumor and hearsay that this is the place to go for help with such things. Please help me get to the bottom of this! My computer, which I ran over-clocked without issue since June, within the last two or three weeks has started to crash on me. My first course of action was to run memtest86+. Even though I didn't get any errors, I returned the computer to stock speeds. When the problem persisted, my second course of action was to remove the resistor I had slowing my main exhaust fan. Temperatures at load improved significantly (especially chipset and PWMIC temps), but my computer still crashes. It seems that something is more permanently wrong. Here is the error code I wrote down: STOP: 0x0000009C (0x0000000000000004, 0xFFFFF80000123BC0, 0x00000000B2000000, 0x0000000000070F0F) Unfortunately, I have no dump file to share, at least not yet. Microsoft support article KB 130536 has led me to believe that I need to create a "minidump" subdirectory in my windows (%systemroot%) directory, since a dump file will not be written to a non-existent path. I have done so - when I crash again, hopefully I'll have a .dmp file to share. Originally, I only had a page-file on my secondary hard drive, the smaller Hitachi drive. I've since added a small (256MB) paging file to my boot drive in order to get a memory dump. I have my computer set to create small memory dump (128k). Should I increase the page file size on my C: drive and set it to do a Kernal dump instead? These are my system's specifications: Operating System: Windows XP Profession x64 Edition service pack 2 Mother Board: DFI LanParty UT nF4 SLI-DR (04/06/2006 BIOS) Chipset Cooling: Jing Ting JTS 0006, no fan. CPU: AMD Athlon 64 X2 3800+ CPU Cooling: Thermalright XP-90, 92mm Nexus Real Silent Case Fan Memory: 2x1024MB Patriot PC3500LLK (PDC2G3500LLK) Graphics: XFX GeForce 7800GT (factory over-clocked) Graphics Cooling: Arctic Cooling NV Silencer 5 (Rev. 3) Hard Drive 1: 200GB Western Digital CaviarSE WD2000JD 7,200RPM SATA Hard Drive 2: 80GB Hitachi Deskstar 7K80 7,200RPM SATA Optical Drive: Lite-On LH-20A1s SATA Power Supply: Seasonic S12 500 Watt Case: Antec P180 with three 120mm Yate Loon D12SL-12 fans (one at stock speed, two currently under-volted to 5V) The crashes occur in the middle of the night, when Diskeeper has various operations scheduled to run. As near as I can figure, there are a lot of things that could be wrong: CPU or Cache failure, which pretty much mean its time for a new computer Memory Failure. memtest86+ didn't return any errors, but I'm going to try again. In this case, I have the warranty to fall back on Chipset or SATA controller failure - once again, new computer time Hard Drive failure is easier to deal with, I suppose Bad power from the wall or surge protector - what kind of UPS should I get to ensure I have clean power? Power Supply failure: Seasonic makes good stuff, but nobody's perfect. What kind of digital multimeter should I buy, and how should I go about testing for this or for the above scenario? Advice and recommendations would be most appreciated. Thanks everyone, George Austin EDIT: Hitachi Drive Fitness Test and Western Digital Data Lifeguard Diagnostic have found absolutely nothing wrong with either drive after full media tests, so I'm definitely focusing on CPU, memory, MB, and power issues.