Trying to isolate cause of BSoD - RAM, mobo or CPU? Please help!

[soulhunter123]

Hello all!

I'm having some incredibly annoying and persistant issues with Blue Screens of Death on my latest build. - see my sig for specifications.

It's about 4 months old, and for the most part runs very well. However over the last month or so I've encountered blue screens, maybe 1 per day or less, all of which seem to point to my memory as being faulty.

So I returned the RAM, got 4 brand new sticks, and it ran fine for a while. Now it's crashing again. I've run the RAM through Memtest in several configurations (individually, paired, all 4) several times - and here's the interesting part... it doesn't find errors every time. Only sometimes. And they're usually "simple" errors like the misplacing of 1 bit of data in an "ffffffff" sequence. Other symptoms include applications (especially Firefox, dunno why) shutting down instantly of their own accord.

The motherboard has been acting up recently - the onboard RAID controller simply would not function correctly for a few hours, no matter what I did, and suddenly came back to life. There's no visible signs of damage to the board and I really don't know how you'd go about testing for mobo faults.

So, my ultimate question is "how do I isolate the cause of this"? Could it be overheating of the RAM, since I run 4 modules right beside each other? Could it be the CPU spitting out random errors? Or the motherboard doing something stupid?

Any advice is appreciated. Thanks!

 

[soulhunter123]

Double posting to include my signature (with specs).

 

[Teaf]

Did you wittle it down to which stick was always in the machine when you were testing Memtest? It could just be one stick out of the 4 you have that could be bad - it's all it'd need to be to cause havoc. I very much doubt it's the CPU or anything heat related with regards to the RAM. It's most likely going to be 1 stick of the 4 you have being bad, or a bad memory controller on the motherboard (or the motherboard being generally broken).

You can wittle this down but it'll take time.

1. Label the sticks #1 to #4
2. Label the memory slots #1 to #4
3. Run memtests in various configurations

What you should be trying to do is identifying which sticks are in when the memtest spits out errors, even 1 error. See if you can find a pattern, ie. it's only happening when a certain stick in the machine, or when any stick is in a certain memory slot. Although because you've RMA'd all your RAM and get all brand new sticks, it would be kinda unlucky and very unlikely that you'll get ANOTHER bad stick of RAM - which could mean it's motherboard related.