Western Digital Caviar SE 500GB 7200 RPM Hard Drive
Western Digital Caviar Green 1TB Hard Drive
All running on Windows Vista 64 bit ultimate
It ran beautifully for the first 2 and a half months or so, when it suddenly froze during gameplay to a gray screen- computer was unresponsive. I rebooted only to find that the computer froze again soon after (this time no gray screen, just complete freeze). Again and again, the computer gradually froze sooner and sooner until it was freezing during bootup. When I initially built the computer, I had some trouble getting the memory voltage correct until it would stop freezing, but I had fixed it. Seeing as I didn't change any of the settings in the bios, I thought it could be a software problem of some sort- the computer was running cool, with no other hardware fault I could see. So I reinstalled windows only to see the problem persisted. I eventually ran memtest to see errors flying everywhere, so I started tweaking the memory again. I eventually found advice on another forum to set the MCH core to 1.28v (from 1.1v) and the Dram to 2.08v. It was suggested that the northbridge on my motherboard was bad- seeing as I had to overvolt it so much- but once my computer resumed normal operation, I sorta forgot about it. Well anyhow, about the same length of time has passed since I reinstalled windows and the problem has revealed itself yet again. I ran memtest and sure enough errors showed up oddly enough... so I set the ram voltage back to "Auto" settings in the bios- still froze. I then changed it back to my settings (MCH core at 1.28v and Dram to 2.08v) and now it is running semi-alright again, albeit a few hiccups on startup (once inproperly loaded my profile, just acted a bit jerky here and there).
Anyhow I don't want to just ignore it this time around. Could it really be my motherboard that is causing these issues? Is it my memory? I've run memtest all the way through and all sticks check out. I want to be sure it is the motherboard before I issue an RMA with Gigabyte- as that would be a pain in the ass.
All suggestions and alternate theories are welcome, please- any advice would be appreciated. I will try to provide as much information as I possibly can so please feel free to ask for any that may aid in solving this problem
This is nominally 2.0-2.1 V RAM, four sticks often require a little 'bump' past that, also reason MCH wants to be high... With four sticks of fast RAM, cooling both the RAM and the MCH is never a bad idea:
This appears to me as a memory problem. Try removing 2 sticks, set the voltage to 2.1. Run it for a while and see what happens. 4 sticks can be troublesome to get running correctly on any board, and could require even higher voltage.
You also have not stated exactly what timings and buss speed you are running your memory at. That can also be a big problem.
Set you memory back to standard 800mhz settings and see what happens. Unless you are pushing the FSB of your processor over 400mhz, running your memory at a higher buss doesn't gain you much of anything...except problems.
Also, you did not mention if you were overclocking the CPU. That again can make a difference as well, if you aren't doing it correctly. So....let us know.
I am running my ram at it's specified timings, 5-5-5-15
I am running the ram at 800mhz, have been this whole time.
I guess I am confused as to why the problem only chooses to present itself randomly every few months? As in why does the computer work normally for a while and then suddenly decide the ram isn't doing alright?
I would like to overclock the memory and processor if I could, but quite frankly- don't know how- and still need to get the system stable, i.e. no freezes at the very least for a year. Haven't made it to three months yet ; )
Thank you for the advice thus far. As of now the computer has not frozen yet with the voltage set to 2.08v- would you suggest increasing it past 2.1v for further-increased stability (with all four sticks)?