OK, here’s the situation: I just ran memtest86+ and started getting errors from the first test on. How do I best go about determining whether the modules are bad (2 2GB Corsair PC26400 @ 800 MHz --I think it’s this stuff: TWIN2X4096-6400C5) or whether it’s the slots on the motherboard, a Gigabyte EP45-UD3R (v. 1.1)? Do I first just pop out one of the DIMMS from its existing slot and test the other one where it is, or just move the two DIMMs to the two empty slots and re-test?
I’m just trying to figure out the most efficient way to do this. Maybe one DIMM at a time in each of the 4 slots?
BTW, WTF is up with RAM prices now, geez! I’m sure that stuff cost me a lot less last year.
If it is the RAM and not the motherboard, I’d appreciate recommendations on what to get as a replacement.
Finally (and I know this is long, sorry) is there a good guide to interpreting the results of the tests? In the early passes I kept seeing stuff like this under addresses for the errors:
0008f2a9b60 – 2290.6 MB
0001476f558 – 327.4 MB
Those two “MB” addresses kept coming up, although the memory addresses before them would vary.
The fact you’re seeing the errors at two locations tells me that there might actually be 2 bad DIMMs. Slots RARELY go bad, so I’d be more inclined to think it was a RAM problem. You can try both of them in slots 1 & 2, then try both in 3 & 4, and see if the error happens around the same memory location.
I’ve never dealt with Corsair’s RMA dept. Mwave will have a copy of your order records though, so that should be all the proof that Corsair needs. In the meantime, got any spare DDR2 laying around to hold ya off until the RMA is over?
Bummer. Is this motherboard/CPU/VGA combination something you see yourself keeping around for another two years or more? If so, might as well grab matching 1GB x 2 DIMMS and then you can run 6GB of total memory once you get your stuff replaced.