Need Advice on hardware malfunction

[Suck.]

OK, so I'm doing my normal stuff on my computer yesterday, browsing the web looking for Hamachi networks, when Firefox crashes. Upon trying to reopen it, it crashes again. I rebooted, and upon reboot, I quickly opened Firefox. All good. Then, out of the blue, like 7 or 8 programs or system services have errors and are terminated, some important stuff too. I reboot again. Same thing happens, but it's about 10 this time. I'm like, "wtfh4x," so reboot again. At least 12 programs or services catastrophically fail.

So... the only thing I can think of is to run Memtest86. I run it, and it gets to the 2nd test at which point my first stick of memory causes the process to enter a never-ending loop between 2,500 and 3,500 errors, although the report says only 33 in another place. ?? At this point, I decide to test each stick independently to see how bad the damage is, and make sure I isolate the correct bad stick of memory. Upon unseating my RAM and running one Memtest86 pass on each stick, with several runs for the first stick that had reported so many errors initially, I find no errors.

In the interum, while swapping memory sticks, maybe three times the system board power LED flickered on while removing a stick, perhaps with some latent power discharge from the stick or something. The box was unplugged and hard switched off at all times while manipulating hardware. After this happened each time, I got an overclocking error upon booting up, which was easily solved by just entering my bios and exiting without making changes. I haven't touched anything in it for at least a couple of months since fixing some crashes by upping my memory voltage (recommended by people with the same board and similar memory setup). When I finally got all four sticks back in, the system flat out wouldn't boot. It didn't even reach the post. I reseated all of my memory, and it fixed the problem.

So anyhow, at this point I figure it's something that only happens when I've got all four sticks in (also having run Memtest86 on almost all possible pairs of memory sticks. Still keeping in mind which stick the initial test indicated errors on, I reinstall all my memory, only to find that Memtest86 does 2 full passes that come back clean. H4x?

I'm now running in the clear as far as I can tell, but uncertain what I should do, if anything. I'm going to leave Memtest86 on overnight to get at least a dozen passes in, but if that's clear, I'm up in the air as to whether the problem was the memory, or possibly my main board, as the powering-on while swapping memory/post failure might suggest.

If I keep running in the clear for the next few days, should I still try to take action as long as my hardware is under warranty and try to swap out my motherboard and/or RAM? Does this sort of thing happen ever, where memory just needs to be reseated to fix catastrophic memory errors? Help?

EDIT:
12/23 @ 22:38 - Just had a complete lock-up. Will post further errors if/when they occur.


Thanks,
Grant

[Best_predator]

I would switch it out while the warranty is good and leave the voltage at default.

[Suck.]

Ran about 19 cycles of Memtest86, and found 655 total errors, although most of them looked like paging errors. Still, it's enough for me to think I need to switch out my memory while I can. As far as the voltage, I really don't have a choice. I either set it high or endure daily crashes.


-Grant.

[K2]

So what's the verdict on this Suck? Made any progress?

[Suck.]

Since last post, I've run about 10 Memtest86 passes every couple of days, and haven't found a single error. The last post I made was about a bunch of errors, but they were all reported as paging errors and not attributed to any particular memory stick, so at this point, I'm sorta' chalking this one up to a bad connection that was fixed upon reseating the memory. It's an unsettling situation, but having been unable to replicate a single memory error after dozens of passes, I don't know that I can justify trying to get the memory replaced.


-Grant.