This is a story of a ocz ram nightmare.....
Ok, so I purchased the following ram about a week ago (running an e6600 c2d, bfg 800w psu, and old 7800gtx(yep needs changing) )....
2 kits of the OCZ 2GB (2 x 1GB) PC2-8500 1066MHz SLI-Ready Edition Dual Channel DDR2 (OCZ2N1066SR2GK)
Put a kit into an asus p5n32-e sli nforce 680i mobo. Reset the bios, insert kit, went back in, set up the vdimm to 2.1, mem timings to 5-5-5-15-2t, stock mhz. Ran memtest for about 20 mins, no probs. Booted into windows, it froze after about 20 seconds. So ok, figure perhaps a bad set of ram that just died after a while.
Did some diags, and the machine wouldnt boot with one of the sticks.
So, reset bios, remove kit, insert second kit. Set up memory volts etc. Reboot, memtest, ok fine. Went into windows for an hour or so. Froze up. Here we go again. One of the sticks was dead again like the other kit.
So, we figure it is a bad mobo. Or is it? Perhaps it was just two dodgy ram kits. Low probability, but hey. Mobo was past its 28 days return/7day return, so it would have to go back to ocuk then off to asus. No pc for weeks? I dont think so. So forked out for a abit in9 32-max wifi.
The two kits of ocz get rma'd back, they were only like 3 days old. They send me out two new kits. Bloody great service as they turned up 3 days later.
So yesterday (sat), started rebuilding machine. Out with the asus mobo, in with the new abit in9 32-max. It is a rev 1.0 mobo even though it was delivered on saturday (12th may) from ocuk. (I think there is 1.1 out now, not sure what the changes are, as the bios update readme talks about rev1.0 and rev1.1). Bios on the mobo was id 10 , the release one, dated 12/25/2006 (some people work on xmas day). I left that bios on there for now.
Inserted new kit. Reset cmos, powered up for the 1st time. All seemed ok. Took my chances and didnt run memtest this time. Found the nvidia array, etc, booted into xp. Working for for a while. Then freeze up.
omg wtf etc etc. So thought perhaps need new bios. Slapped on the bios version 11, dated 2/13/2007. What a nightmare. Everytime I came out the bios, it would say 'checksum error' and reload the defaults. So I couldnt change anything. I did some reading up on this and it seems there is a problem with epp and this sli ram. So i rolled back to version 10. Thankfully it would still boot from floppy. Flashed fine etc. Back to square one.
At this point I cranked up memtest. I always run test 5 first, as this seems to find stuff quickly. 5,6,7 itterations no probs. On the 8th, screen fulls of red errors. Check mem volts, yep 2.1, 5-5-5-15 timing, 800mhz data rate, that is like pc2 6400, this stuff is pc2 8500 (1066mhz). So, after checking ocz tech site, I crank the volts to 2.3 and perform memtest again. Same errors. I feel the temps of the sticks, they are very cool (I am running a coolermaster geminII heatsink with 2x120mm noctua's).
So I think great. Another duff pack of ram. Shut down, reset bios. Dig out the 2nd pack that got rma'd back to me. Install that. Power on. Beeeeeeeeeeeep. No post, no nothing. Just beeeeep beep etc. Great. So I remove those, put back in the set that was just erroring. Posts fine, set up the ram volts 2.1 etc etc.... back into memtest 1.70+, errors again.
Ok, so I do a reset, and download M622A_12.B03 beta bios. Seems like the oc chaps over at xtremesys are using it. Flash that (risky I know with duff ram but hey), reset cmos etc, set up ram again, run memtest. It errors again.
Now what the heck ???? My head is about to hit the desk a few times.
Anyone advise me on what I should do?
Richie.
PS: sorry for the ramble, but damn I needed to get that off my chest.
Edit: I cross posted this into ocz tech forums to see if they can help out.
Ok, so I purchased the following ram about a week ago (running an e6600 c2d, bfg 800w psu, and old 7800gtx(yep needs changing) )....
2 kits of the OCZ 2GB (2 x 1GB) PC2-8500 1066MHz SLI-Ready Edition Dual Channel DDR2 (OCZ2N1066SR2GK)
Put a kit into an asus p5n32-e sli nforce 680i mobo. Reset the bios, insert kit, went back in, set up the vdimm to 2.1, mem timings to 5-5-5-15-2t, stock mhz. Ran memtest for about 20 mins, no probs. Booted into windows, it froze after about 20 seconds. So ok, figure perhaps a bad set of ram that just died after a while.
Did some diags, and the machine wouldnt boot with one of the sticks.
So, reset bios, remove kit, insert second kit. Set up memory volts etc. Reboot, memtest, ok fine. Went into windows for an hour or so. Froze up. Here we go again. One of the sticks was dead again like the other kit.
So, we figure it is a bad mobo. Or is it? Perhaps it was just two dodgy ram kits. Low probability, but hey. Mobo was past its 28 days return/7day return, so it would have to go back to ocuk then off to asus. No pc for weeks? I dont think so. So forked out for a abit in9 32-max wifi.
The two kits of ocz get rma'd back, they were only like 3 days old. They send me out two new kits. Bloody great service as they turned up 3 days later.
So yesterday (sat), started rebuilding machine. Out with the asus mobo, in with the new abit in9 32-max. It is a rev 1.0 mobo even though it was delivered on saturday (12th may) from ocuk. (I think there is 1.1 out now, not sure what the changes are, as the bios update readme talks about rev1.0 and rev1.1). Bios on the mobo was id 10 , the release one, dated 12/25/2006 (some people work on xmas day). I left that bios on there for now.
Inserted new kit. Reset cmos, powered up for the 1st time. All seemed ok. Took my chances and didnt run memtest this time. Found the nvidia array, etc, booted into xp. Working for for a while. Then freeze up.
omg wtf etc etc. So thought perhaps need new bios. Slapped on the bios version 11, dated 2/13/2007. What a nightmare. Everytime I came out the bios, it would say 'checksum error' and reload the defaults. So I couldnt change anything. I did some reading up on this and it seems there is a problem with epp and this sli ram. So i rolled back to version 10. Thankfully it would still boot from floppy. Flashed fine etc. Back to square one.
At this point I cranked up memtest. I always run test 5 first, as this seems to find stuff quickly. 5,6,7 itterations no probs. On the 8th, screen fulls of red errors. Check mem volts, yep 2.1, 5-5-5-15 timing, 800mhz data rate, that is like pc2 6400, this stuff is pc2 8500 (1066mhz). So, after checking ocz tech site, I crank the volts to 2.3 and perform memtest again. Same errors. I feel the temps of the sticks, they are very cool (I am running a coolermaster geminII heatsink with 2x120mm noctua's).
So I think great. Another duff pack of ram. Shut down, reset bios. Dig out the 2nd pack that got rma'd back to me. Install that. Power on. Beeeeeeeeeeeep. No post, no nothing. Just beeeeep beep etc. Great. So I remove those, put back in the set that was just erroring. Posts fine, set up the ram volts 2.1 etc etc.... back into memtest 1.70+, errors again.
Ok, so I do a reset, and download M622A_12.B03 beta bios. Seems like the oc chaps over at xtremesys are using it. Flash that (risky I know with duff ram but hey), reset cmos etc, set up ram again, run memtest. It errors again.
Now what the heck ???? My head is about to hit the desk a few times.
Anyone advise me on what I should do?
Richie.
PS: sorry for the ramble, but damn I needed to get that off my chest.
Edit: I cross posted this into ocz tech forums to see if they can help out.
Last edited:
