You rarely encounter such an issue; it's quite uncommon.
You rarely encounter such an issue; it's quite uncommon.
This is a troubleshooting-related question, but I'm putting it here as my box is now operational. For those of you with lots of experience troubleshooting systems: how often do you get an issue that eventually turns out to be the RAM - and not even a bad module, but just funky contacts? I just got through a few hours of troubleshooting an older Ivy Bridge 3770k with 32GB DDR3 RAM that suddenly started doing the PSU cycling thing - turns on, fans spin, no display/BIOS, then shuts off after 20 seconds, then 10 more seconds later does it all over again. I pulled the GPU (most recent new component), then swapped the PSU, nothing. Pulled all drives, nothing. Not until I took out half the RAM (2 of 4 sticks DDR3) did it come up. I figure, aha! One of those two sticks is bad. So I swap those two in, pull the "known good" sticks, and... it comes up fine. Reinstall all four sticks and... it comes up fine. Reinstall the old PSU, GPU, drives, and it starts up like nothing ever happened. I pull the RAM and clean the contacts to make sure, and now I'm typing this post on that system. Over 25+ years of nothing but my own builds, upgrades and repairs, I have had my share of mysterious hardware failures, but a ridiculous proportion of those issues have ended up being just like that one. I'd say more than half. It's happened with an Athlon 1.4GHz SDRAM box, a dual-socket DDR1 AMD 1900+ system, a Q6600 quad core DDR2, a Phenom x6 DDR3, this Ivy Bridge system *twice* now (but involving different RAM each time). Knock on wood, none of my DDR4 systems have done it. Only three times in that span have I had a failure that actually was a failure. Two PSU's died (one in 2004, one just last year), and once I actually did have a bad stick of DDR3 RAM I had to RMA in 2012 (coincidentally, the replacement RAM I got for that one is the RAM involved in today's annoyance). It's because of this that I'm such a big fan of contact cleaners. More than a few of those times, just reseating wasn't enough - I had to clean off the contacts (my product of choice is DeOxit). But it seems strange that this has been the gremlin so many times for me. I don't get the impression from reading troubleshooting fora that it happens to others that often, and "Clean your RAM contacts" doesn't seem to get suggested that often either.
In roughly thirty years of experimenting with computers, I've never touched one.
I faced an issue where one of my sticks didn’t arrive, and it seemed a tiny piece of debris was lodged in the slot of the used motherboard I bought. After thoroughly cleaning all slots with canned air, the problem resolved and everything functioned properly.
Frequently, across laptops and desktops, my main system wouldn't boot properly. Swapping the RAM twice didn’t fix it—I might have dropped it while handling something else. My laptop sometimes had tricky RAM issues, and reinserting it resolved the problem. In contrast, my friend’s desktop only displayed 8GB, but after reseating its RAM, it showed the full 16GB.
My experience with the AMD 1900+ dual socket was similar. I was adjusting the IDE configuration for the HD to resolve a small problem, and during the process I accidentally pushed one of the RAM sticks into its slot to reach a cable. That triggered the issue. God saw what I’d done—since I was handling the drives—I spent about an hour focusing on them before finally deciding to remove everything (RAM and cards), clean the contacts, reinstall, and everything worked afterward.