Multiple Different BSOD with Windows 10

akatashi · 09 Mar 2016

I've been working on this problem for 4 days now, and i have made no headway, i get all sorts of different BSOD's, have done 3 fresh installs of windows 10, 2 of windows 7, i will include the latest 2 dumps (only had this install up and running for 2 hours now...

Hopefully someone else can make heads or tails of this, i've more or less assumed its the motherboard, my old ram tested with 1.6m + errors on test 11 with memtest, replaced it with new and it wasn't the issue, but i assume the motherboard affected the ram, or i was getting false positives with memtest because of a faulty motherboard. I can return the motherboard and just get it replaced (2 year replacement plan with microcenter) so thats the next step, but i figured i'd post here and see if maybe someone else thought it may be software, or something else i could test/fix at home, thanks a lot, and i will update with any necessary infoAttachment 68529

Arc · 10 Mar 2016

akatashi said:

Hopefully someone else can make heads or tails of this, i've more or less assumed its the motherboard, my old ram tested with 1.6m + errors on test 11 with memtest, replaced it with new and it wasn't the issue, but i assume the motherboard affected the ram, or i was getting false positives with memtest because of a faulty motherboard

All the possibilities are there. There are multiple different BSODs, but all those are memory failure related.

Code:

BugCheck 19, {20, ffffc001f983fcb0, ffffc001f983fe00, 5150309}

fffff8022a7f9520: Unable to get MiVisibleState
Probably caused by : ntkrnlmp.exe ( nt!AlpcpReceiveMessage+1174 )

Followup:     MachineOwner
---------

Code:

BugCheck 50, {ffffd00023234dd4, 2, fffff961f025fce5, 1}


Could not read faulting driver name
Probably caused by : win32kfull.sys ( win32kfull!xxxSendMessageToClient+55 )

Followup:     MachineOwner
---------

Have you tested the new RAM too? If not, Test it. Test your RAM modules for possible errors.

MemTest86+ - Test RAM

Run memtest86+ for at least 8 consecutive passes.

If it start showing errors/red lines, stop testing. A single error is enough to determine that something is going bad there.
Take a camera snap of the memtest86+ window before closing the program. Let us see it.
Screenshots and Files - Upload and Post in Ten Forums - Windows 10 Forums

If it is free from errors, apply the RAM sticks in the two non-used ports. See if it behaves the same or not.

Let us know.

akatashi · 10 Mar 2016

hey Arc, i did test the memory, my first sticks of ram came back with 1.6 million errors in 1 test during memtest, so i attempted with different ram, same outcome, determined it must be the mobo, returned it and got a new one as of today, now that thats solved, i think i've figured out it may be a CPU issue, the CPU is failing prime95 tests (6/8 cores failing in under 1 minute) i am attempting to raise the voltage to solve the rounding error, if all else fails i will return the CPU and get a new one as it may be faulty.

If there is any other input it would be VERY much appreciated though!

Arc · 10 Mar 2016

OK, as you have identified one problem and have taken the necessary actions, see how it goes.
Report back for any further issue. We will be here.

akatashi · 10 Mar 2016

thank you Arc! I am currently running at 1.5v and 1.285v NB i'm a bit worried to take it higher as i have solid temps (69 with full load on prime95) but i do still have core #1 failing >.< though increasing the voltage has stopped the blue screens thus far, and prime 95 doesn't freeze while running. I plan to run prime95 while sleeping tonight to see how long it takes to get a fatal error with any of them (hopefully none) but what would be safe temps to try and equalize core #1? (ASRock Fatal1ty Performance 970 + AMD FX8370)

Arc · 10 Mar 2016

IMHO 80-85 degrees may be considered as maximum allowable increase of temperature.

akatashi · 10 Mar 2016

Okay i will increase a bit more and see where i am at! I ran Prime95 for 8 hours and 27 minutes last night, results were core 2 giving a Fatal Error at 4 mintues in, Core 1 giving a fatal error at 22 minutes in, and the other 6 passing for all 8 hours. So for some reason raising the voltage on a 4.4mhz clock speed is stabilizing my cores. To be honest i don't quite understand WHY its working, or why the normal voltages out of box for the CPU were not enough. (the MOBO i have supports the CPU i have) but at least it is working as of now. I am going to push it to up a bit higher, then give it a few days, i have a 2 year warranty on the CPU so if anything goes wrong i can return it, i'm just afraid it won't matter and i'll have to raise the voltage on a new one as well.

Thank you for the help Arc! I will again try to post back here in a few days to let you know how all is going and see if i can get stable around 1.6v/1.3v NB