Asus M4A87TD EVO incompatibility?

[Sp3ktr3]

Windows 7 Pro 64
Asus M4A87TD EVO bios 1005
Gskill Ripjaws F3-10666CL9D-8GBRL 4GBx2
AMD PhenomII x4 965BE 3.4GHz
XFX Radeon 5850 video card
Sound Blaster X-Fi Xtreme Gamer

I built this system about 2 months ago. I use it as my primary computer as well as running BOINC projects on it. It functioned properly for a few weeks, and then I started getting BSOD's and crashes, both at random times and under specifically reproducible circumstances. Particularly, when starting or stopping a CPU-intensive application it will often BSOD or black screen. It will also occasionally crash when unloading the GPU after it has been running a project for awhile. Additionally, it sometimes freezes at POST, or in BIOS. Every crash is accompanied by the MemOK! light on the motherboard coming on. I have a sneaking suspicion this motherboard does not like the 4GB memory modules, even though it is rated for up to 16GB. I see very few 4GB modules of any brand in the certified memory list. I'm also aware that some PhenomII MC's have trouble with certain types of memory. However I have another computer running 2x2GB Ripjaws 1333MHz on a PhenomII 945 3.0GHz with a Biostar motherboard, and it's been running under load 24/7 for several months without issues. Here are the tests and changes I have tried thus far:

complete Windows reinstall
Windows memory diagnostic: no errors
memtest86 v3.5: fails at address test walking ones 4096-6144 MB
memtest86 v3.4: fails at random points in various tests
1333MHz/1067MHz 8GB dual channel A/B
1333MHz/1067MHz 8GB single channel
1333MHz/1067MHz 4GB single module, all slots
1333MHz 9-9-9-24-1T/2T
1067MHz 8-8-8-20-1T/2T
NB voltage 1.1V -> 1.2V
DDR voltage 1.5V -> 1.6V
unganged -> ganged mode

None of these changes has affected the instability of the system. I don't have any spare DDR3 to swap out, and don't really want to rip apart my science machine to try the smaller modules. If anyone has any input on this situation, it would be greatly appreciated.

edit: CPU temp is 50-52?C max under load, and all other temperatures are well within acceptable ranges.

[GSKILL TECH]

Use a newer memtest86+. This memory should be plug and play no problem. Test each individually to see if one is defective causing the problem.

Thank you
GSKILL TECH

[Sp3ktr3]

I ran memtest86+ 4.1. Initially it crashed, same as always, with 8GB dual channel. Windows Memory Diagnostic also showed a hardware error at this point. I then ran each stick individually and they checked out. I also tested them in the other DIMM slots, they also checked out. I then ran both together again in dual channel, and they checked out. >< If I hadn't experienced the same errors after multiple swapouts earlier, I would say it was a seating issue. I've been building systems for 10+ years and I've never encountered this amount of frustration in identifying a problem.

[Tradesman]

May want to raise the voltage to your 965, the MC might be straining at the current voltage, maybe + 0.05 or + 0.07

[mkw87]

Try re-seating your heat sink on your processor. I recently had that same set of ram sticks on my Asus M4A88TD-V work for 3 days then go nuts throwing blue screens, hard locks, etc. I have an x6 1055T, but the same problem can occur with any similar processor.

Hope that fixes it for you, I was going nuts with mine.

[Sp3ktr3]

Thanks, but I don't believe it's a cooling issue. My CPU hits 52? max under load WITH a high ambient temperature. When it's cooler in the room it hits 46-48? max. Adjusting the fan speed also directly impacts the CPU temp, so there is good thermal contact.

The system has been running stable for about 24 hours now with CPU and GPU under max load. The last set of DIMM swaps seems to have fixed the problem, for now. It's odd because I re-insterted the DIMMS and verified proper seating a dozen times beforehand for various tests. TBH I prefer it when something just blows up, then I know exactly what went wrong.

[mkw87]

Mine wasn't a cooling issue either, my cpu never got hot (stayed under 50?C). Just the random freezes under load and not under load. It would freeze checking email but not running super pi. Then it got to where it would freeze at random not long after loading into windows every time. Mine ran fine for 3 days until I put a third hard drive in, then it went nuts. Removing the hard drive didn't fix the problem. Unhooking then rehooking the heat sink fixed it. Strange I know, but that's what I did.

Was just tossing it out there as I was at a loss as to what to do. Like you said when things smoke it makes it easy............

[Tradesman]

Good to hear it seems okay, if problems crop back up let us know

[Sp3ktr3]

It messed up again after a couple days. I installed the new memory I ordered, it seems to be working so far without any errors. Time will tell.

[Tradesman]

Keep us informed