• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

My lengthy, ongoing, and confusing issue with RAM...

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

skoolz

Registered
Joined
Sep 20, 2010
I'll make this as brief as possible. Over a year ago, I updated my mobo to the Gigabyte 890FXA-UD5. I had 4 GB of DDR3 1333 RAM with timings at 9-9-9-24. I also got a 1090t CPU at the same time. Everything worked great, except occasionally I would get a temporary lock up followed by the "Display Driver has stopped responding, but recovered blah blah" error. This would happen in 3D games mainly. After some research, I ended up RMAing my gfx card (HD 5870) thinking that had to be the problem. Nothing was wrong with it. After much more research, I learned that RAM could be causing this problem, so I toyed with my RAM settings, and discovered that the system is stable when I change the timings to 10-10-10-24. I just went straight to 10-10-10 and didn't try anything else, but it worked. I am a fan of overclocking, so you can imagine how this bugs the hell out of me. My 1090t is OC'd @ 4ghz stable, and my HD 5870 @ 940Mhz. I want to OC my RAM, but this ridiculous issue is forcing me to underclock the timings! I recently upgraded to 12 GB of RAM using the same RAM as my previous 4GB. Does anybody have any idea what could be causing this, because I still haven't found a reasonable answer. And perhaps someone has an alternative fix? Maybe I need to play with some voltages? (I played with my RAM voltages, but it didn't work. I went up to 1.6v using the standard 1333 speed and timings.) Perhaps one of you here are the savior I've so desperately needed for over a year!
 
Have you tried it with some of the new RAM that you bought and took out the old? Maybe some of the sticks that you had previously were just bad. Did you ever run them under memtest86 to test to see if they were faulty?
 
I've thought about doing that, but what kept me from trying just the new modules was that others have had the same problem as me. I'll probably do that next, or run memtest tonight as I have not done it yet (embarassingly enough).
 
Well, I ran memtest86+ last night for around 6 hours, completed 4 passes with 0 errors. I know I could run it a whole lot longer, but any errors appearing after that would be really rare. I think it's safe to say that my modules aren't faulty as I think a symptom as bad as making me raise my timings would cause an error in just the first pass!
 
It's possible the error would be rare for the memory diagnostic but not for some other programs you run. Try MemTest86 as well because I've often gotten consistently different results with it than MemTest86+, and I don't know why because they use the same test methods. Also Gold Memory has sometimes found errors that others missed, and vice-versa. One person had memory problems that were missed by one diagnostic, and another diagnostic took over 9 hours to find one error a second time. Replacing the memory cleared the problems.
 
Back