I have that probelm too! Did you find a solution yet?
Greetz, Obi Wan
Allright, I also had these memtest problems, different mobo and memory though.
Mine looked way worse than yours - I was unable to pass neither memtest86+/memtest86/Windows Memory Diagnostic with any of my corsair tr3x2g1600c8d modules (2x3 channel kits, different versions-mem vendors), in any possible configuration (single/dual/triple).
Only my reboots weren't random. I had a 'rock solid' reboot on test #3 'moving inversions 8bit pattern'.
Windows (7pro 64bit) experience was stable though, even while overclocking (@4ghz). Prime95, OCCT, Intel XTU's test all showed no problem (some ran overnight).
I started suspecting the cpu's mem controller (windows memory diagnostic with cache disabled didn't cause reboots), but than again - stable overclock...
I spent a day flashing to different bios versions and working through bios settings, disabling one at a time 'till I had nothing to disable)) No luck!
Then I tried this software. Some claimed it more aggressive, other (memtest86+ fans) claimed it to be rubbish. Anyway, most of them agree it reveals stuff, other sw might not. So I gave it a run...
8 hours in windows environment, running several instances of sw to utilize all free memory; and then i ran the boot version (linux kernel) for 20 hours. No errors.
that was enough for me.
No, I haven´t found a solution. The system is rock solid with any 8 GB of RAM. I run with several Boinc projects so the CPU keeps at 100% all the time. However if I put the full 16 GB of RAM into the system I get random reboots, even disabling Boinc and keeping the processor at 0% most of the time. As I said before I´ve used several memory testing programs and all 16 GB seem to be OK. I don´t know what to do. It seems to be a BIOS problem and I keep installing the newest versions in the hope that it will solve the problem. In the mean time I have 8 GB of RAM in my desk....
I don´t think I get your point.... My system is not overclocked and is 100% stable with any 8 GB of RAM. All 16 GB of RAM have passed ALL memory tests. My system consistently crashes with 16 GB of RAM regardless of combination and even with the processor close to 0% use, but is 100% stable with 8 GB of RAM even with 100% processor use.
If I understand your post you have the inverse problem. Your system is 100% stable, even with and overclock, with 16 GB of RAM. However some memory testing programs consistently flag your memory to have problems... If this interpretation is correct then I don´t think your problem has anything in common with mine and I think you have a bad DIMM and you have been very lucky not to hit that particular memory area in general use.
Thank you for your interest.
If the normal volts on that ram is 1.5 volts you may have to raise it to 1.6 volts .
Yeah, I finally got it. One 4 GB piece was broken, found it out with memtest86+.
Cheers, Obi Wan
Has anyone found a solution or root cause for this problem?
Did a little research on this , Intel says , Double sided Dimms with 16X , Not supported . What most people are doing is they are running , Something like Corsair CMZ16GX3M4A-1600C9B , which is 1600Mhz ram , But they let run at JEDEC speed default 1333. And they are running 16GB Solid & Stable.
Saved me a lot of time..Much appreciated!!
OK. Maybe that's the problem, but....
Lets see... My memory is Kingston Part number 9905471-009.A00LF and has 8 chips on each side. The chips are Elpida EDJ2108BCSE -DJ-F 256MX8 with 8 internal banks. I'm not sure if this is 16X or not. Any ideas?
Think thats x16