amd 1090t 3.2ghz 6 core black edition
asrock 890fx deluxe 4
coolermaster v6 in push pull configuration
12gb gskill ddr31600(4gbx2 2gbx2)f3-12800cl9d 9-9-9-24
sapphire radeon 6950 2gb dual bios one moded to unlock the aditional shaders on the board
wd 500gb 7200rpm 16mb cache green
wd 1.5tb 5400rpm 32mb cache green
wd 2tb 5400rpm 64mb cache green
mascool 20mm side intake fan
windows 7 home premium 64bit
this thing was the epitome of stability until 2 weeks ago when it turned into a bsod magnet 2 weeks ago playing skyrim.
bsods are random as all hell with one exception, all but one of them involved the ntoskernel.
low level formatted and reloaded windows seven and the ther software, bsods continued still random as all hell.
ran memtest86 4.10 87 errors found on pass 6 but outside
expected memory range starting at 12440.4mb 12441.4mb ect.
got got the updated version 4.20 and ran it sucessfully through 43 passes 124hr 45min 41sec, no errors what so ever.
bsods running prime95 blended tests within 30 mins.
disabled all automatic overclocking and running at stock everything, still bsods. furthermore i accidentally entered dr lan in the bios last night, and saw some more weird errors.
channel a-normal b-normal c-short d-short problem point at 3m
switched out for unopened spare, same errors, switched ports on router and plugged directly into modem, same thing, unplugged cable and got even more anomalies.
cable disonnected, channel a-open,b-normal,c-short,d-short
channel a-open, b-open c-open, d-short
channel a-open,b-normal c-open d-open, ect.
in the case, i came to realize that while my push fan on the cpu was running, the bearings were going and it wasnt up to full speed, so i pulled it off the v6 and took the masscool fro the side of the tower and put it on the cpu as push for now, got a new fan on order. still errorific; running prime95 small fft, runs successfully but bsods within a minute after stopping the tests, same thing with large ffts, and still bsods before 15mins during blended tests. im really really suspecting the motherboard at this point. disabling the onboard nic appears to have allowed me to run the small and large fft tests, but not the blended, all tests would bsod before. also, when i reinstalled the os, and before, speedfan could run without a problem, then the day after reinstalling the os, and speedfan and makin sure it still ran, it started hanging everytime while scannin the isabus address $0ccb, this is constant. last 5bsods
the system lost the other 8 dump files from after the os reload
and it has also lost 2 copies of bluescreenview and 1 prime95 from the desktop, so i have run the tests from hirens bootcd 14.0 since.
prime95 log reports all tests passed, then bsod and no failed test, this is damn peculiar
when it bsods, ithe screen artifacts like a nintendo(old school) game that has just frozen and goes to the dump procedures.
i still need to run a harddrive stress test and gpu stress test but i dont see how those could corrupt the networking card and make it show errors like those it exhibited in the bios, however if the northbridge is the problem, i can see that causing the errors in the the networking card aswell as the video system as they both run on the pci express bus.
any thoughts or suggestions? i can use all the help i can get, im a power user, but im damn vexed by this quagmire of corruption.
all help will be greatly appreciated
-one frustrated fat guy-
since it seems everyone ho has viewed my problem seems to be as vexed as me, and im relatively sure i have determined the problem, im going to post an update. so i rigged a fan up the northbrige heatpipes, an 80mm case fan from a previous build. skyrim is now playable and i can run the prime 95 blended tests for well i stopped it at 16 hours+. it bsod'd after running the blended test, instead of during as it did previously. my conclusion, when put under pressure, the northbridge can hang until demand ceases and then it is simply unable to handle the change in oscillation frequency associated with the drop in processing and mb bandwidth needs and commits an error that windows cant stand and the system crashes. i will report back once i replace the motherboard. considering there is plenty of cooling in this case, and all temps were well within tolerance before adding the 80mm fan, it seems the northbridge is unable to handle higher within tolerance temps and acts up, so as long as i keep the fan operational on the nb heatpipes, it is stable enough to run with until i get a new mb. also, i lowered the memory to 1333 instead of 1600 to lessen the workload on the mb a bit. well ill report back when mb is replaced, although i may opt to replace the proc too with a new 8 core apu or something like that as well, and i know im going to upgrade the memory while in there since to do so, the mb has to be removed to the coolermaster v6 can be removed to get to the dimm slot. well i will report back, and the graphics card and hard disk did pass their tests aswell