Papageno Posted December 27, 2015 Share Posted December 27, 2015 I think I've isolated the root cause of random spates of BSODs (every few months I'd have like a group of them) I've had with this machine I'm using since I built it four years ago... The likely culprit: bad RAM (the original two-DIMM 8GB kit), which, annoyingly, never tested as defective with memtest86+. I'd recently started having the latest bunch (5 in the last month) and was tearing my hair out, so I went to the bleepingcomputer.com site's forums for help. The guy helping me said my crash minidumps were all over the place, which implied a hardware defect of some sort. He recommended I use Prime95 x64 as a preliminary step, so I did finally last night, using the Blend test. Got errors pretty much IMMEDIATELY. Today I figured, well, if it's RAM, it would be the original RAM, not the extra 8 GB I just got a month ago (the same Corsair p/n as the original, but obviously a different lot), since the BSODs (the bulk of them over the years) preceded their installation. Opened the case, took out the two suspect DIMMS, moved the other two to their slots, rebooted, ran P95, and nary an error nor warning. Let it run for two hours straight, and the CPU temp never went above 60C either, while the mobo stayed around 30C. I think the above evidence argues strongly for the original kit of Corsair RAM being the (subtly) bad component, despite operating fine most of the time. Not overclocking, case clean, power supply and case changed out less than 12 months ago along with video card. Moral of the story: always test a new build with P95, I sure wish I had back then. Now to see whether Corsair will honor its warranty on this bad kit despite the memtest86+ results. Link to comment Share on other sites More sharing options...
This topic is now archived and is closed to further replies.