📢
9
c/computer-technicians•river190river190•12h ago

Found out today that most RAM failures happen in the first 90 days

Been doing this for roughly 7 years now. Always assumed dead sticks showed up whenever they felt like it. Read a report from a refurb shop in Austin, they tracked 400 returns over a year. 70% of the bad RAM they saw died in the first quarter. The rest trickled in slow over years. Makes me rethink how I test used gear. I normally just run MemTest once and call it good. Now I'm wondering if I should burn in new sticks for a week before trusting them. Anyone else see early failures way more than random old ones?
1 comments

Log in to join the discussion

Log In
1 Comment
lane.drew
lane.drew11h ago
Man, that report lines up with what I've seen too. I once had a brand new stick pass MemTest for two straight days, then die on day 89 right in the middle of a big file transfer. The weird part was the machine kept running, just started corrupting everything. I spent a whole weekend chasing a driver issue before I swapped the RAM on a hunch. Now I'm paranoid about doing a full week of heavy load testing on any new build, but honestly, who has the time for that? Most people just want their rig running same day. Still, I've started keeping a cheap spare stick around for the first few months, feels like cheap insurance.
1