Random Memory Errors

[masteroc]

The computer in question was/has been crashing during gaming and internet browsing if I remember correctly (its been a while since I actually used it). I decided to try and find out what is wrong with it now and put it to good use. At first I ran memtest86 on all 4 sticks of ram. I came back after a few hours and found a lot of red errors. So I started trying to narrow it down to which stick of ram was bad. I ran each ram separately for about an hour and no errors came back. I then took a stick of ram and ran each socket for an hour...still no errors.... I decided to run the 4 sticks again...this time, no errors. If I recall correctly I tried this the first time that the computer started acting up and had the same results...first memtest comes back with errors but the following memtests show nothing. What could be happening here?

Specs: Core 2 Quad 6600, 4gb ddr2, 500gb hd, 8800gt gfx, evga 780i mobo

Any ideas?

P.S. Silly me, more detailed specs in sig

Thanks!

 

[MindoverMaster]

How long did you run each stick? it takes a couple loops until it finds an error.

 

[masteroc]

Ran them for about an hour, I know a good test runs them for 24...but the first time errors came up (with all 4 sticks) it was after only approximately an hour.

 

[MindoverMaster]

I'd wait at least 2-3 hours on each stick. Could be anything.. Or just run em overnight.

 

[masteroc]

Put all 4 sticks back in and have left it running for 7 hours so far (letting it go overnight as well). No errors yet....This is strange b/c it had only been running an hour or maybe a bit more when it got tons of errors (all 4 sticks). Could it be some kind of flexing on the motherboard or something....?

 

[Desolation]

might of just been one of the sticks wasn't seated all the way

 

[masteroc]

Normally I would agree, but i'm pretty sure that this has happened before (the main reason that I stopped using the computer in the first place). It seems to happen after I have let the computer sit unused for a while and then tested. Once I start all the testing, the problem disappears....

 

[Slaymate]

It sounds like it could be a problem with the memory controller, which is a part of your motherboard. And the 780i was known to run hot and when it does the memory controller is affected.

 

[masteroc]

Any specific way to test for this, I was thinking it was something motherboard related....although I was telling myself that evga was a good company making good boards....?

 

[Slaymate]

EVGA does make good mobo's, it's really an issue with the Nvidia 780i chipset itself. I had a EVGA 780i mobo a few years ago but I can't remember what I was using to monitor the temps. I would think HWMonitor would work, HWMonitor v1.17