• Happy holidays, folks! Thanks to each and every one of you for being part of the Tom's Hardware community!

[SOLVED] getting inconsistent spontaneous black screen crashes

Retrikaethan

Prominent
Jun 7, 2019
20
4
515
hey so i've been trying to diagnose an issue and i'm pretty sure it's software but no matter what i do this crash still happens (save for switching to my backup gpu). i thought it was a MB issue at first and upgraded the MB/cpu/ram cuz they were all well over 5 years old at that point, crash still happens. i think then it's the gpu (gtx 1080) since the crash doesn't happen on my now backup gpu (gtx 960 ftw) and rma'd it but having gotten a different card back, the crash still happens. as a last ditch attempt i replaced the psu and, of course, the crash still happens. since then i've been updating everything i can get my hands on and so far there's been no change (except maybe that DDUing then reinstalling the gpu drivers caused the crash to change its trigger? a game i was using to reliably cause the crash no longer causes it where i was testing it). so far it has been extremely random when these crashes will occur, save that they happen in games after some period of time of running them (somewhere between 30 minutes and a few hours, sometimes shorter as was in the testing game. temps are all fine, highest is 68C but usually doesn't get there)

i am almost completely at a loss for what to do now with the only thing i know i haven't tried yet being a clean install of windows 10 but i really don't want to have to do that but will be my next course of action if i can't figure this out.

ps: this is probably a dumb question, but can a monitor cause this kind of crash? (mine is pretty old, 1920x1080, 60hz at least 3-4 years old)

specs:

OS: windows 10 enterprise (updated from windows 7, crash was occurring there, too)

cpu: intel i5-8600k

mb: msi z370m (gaming pro ac)

ram: 4x corsair vengeance lpx 1x16GB DDR4 3000MHz (YES I KNOW IT'S OVERKILL)

main gpu: geforce gtx 1080 (gigabyte)

backup gpu: geforce gtx 960 ftw (evga)

power supply: corsair cx750m

shit i've updated: cpu chipset driver. bios, gpu driver, a few other minor things i can't remember.
 
Solution
I have the exact same issue with an entirely different setup. i have an amd build thats doing this so the whole pc is entirely different EXCEPT we are using the same power supply. did you recently get a new case? im starting to consider maybe my case is drawing way to much power. please let me know if you find a fix. i feel like ive tried everything i can to including reinstalling and completely reformatting all drives. dont do it, it aint the move bud
So, if you use a Geforce gtx 1080 your system crashes but if you use another GPU, your backup Geforce gtx 960 ftw, it doesn't, correct? The issue occured with windows 7 and and is again occuring with Windows 10. You have RMA's the Geforce gtx 1080 and replaced the motherboard /cpu/psu/ and ram . As far as a memory test, go to https://www.memtest86.com/ and make a bootable RAM tester.

that’s correct. also doing the test now, 2 hours in with the first pass almost done and no errors so far.
 
You say you upgraded to Windows 10 from Windows 7 and you replaced your motherboard. When you replaced your MB, did you buy the same board as before. If so, have you ever done a clean install of Windows 10, meaning deleting the partitions and installing from a Windows 10 USB and then reinstalling all your programs and drivers?

I'm wondering if using the Enterprise Edition of Windows 10 could be the issue (the Windows 7 problem could have been different). https://www.techrepublic.com/articl...ion-what-are-the-key-features-for-businesses/ From what I read the Enterprise Edition designed with gaming in mind, it is for businesses and is very strict on what drivers it allows and delays upgrades by months or more. Since you have replaced everything else, perhaps using Windows 10 Pro might work better.
 
Last edited:
You say you upgraded to Windows 10 from Windows 7 and you replaced your motherboard. When you replaced your MB, did you buy the same board as before. If so, have you ever done a clean install of Windows 10, meaning deleting the partitions and installing from a Windows 10 USB and then reinstalling all your programs and drivers?

I'm wondering if using the Enterprise Edition of Windows 10 could be the issue (the Windows 7 problem could have been different). https://www.techrepublic.com/articl...ion-what-are-the-key-features-for-businesses/ From what I read the Enterprise Edition designed with gaming in mind, it is for businesses and is very strict on what drivers it allows and delays upgrades by months or more. Since you have replaced everything else, perhaps using Windows 10 Pro might work better.
mb is a newer model along with the cpu and ram. i upgraded the w7 to w10 keeping most of the old stuff. clean install is something i want to avoid (cuz setting everything up again is a pain) but is gonna be what i do next barring any suggestions.

if/when i do a clean install i’ll be sure to use pro instead of enterprise:
 
There is little doubt the lack of a clean install is the basis of your problem. When you upgraded to Windows 10, you likely migrated the problems from Windows 7. The are differences in MB's even a newer model, that make the lack of a clean install very problematic.(your lucky it even works). The issues you are having with your graphics card are very common with someone that is using an old OS with a new MB . Yes, it's going to be a pain, but if you want to use the Geforce gtx 1080 with your system you are going to have to do a complete reinstall. It is also probable your PC's performance has been steadily degrading, you simply haven't noticed it, and at some point a reinstall will be necessary for it's operation.
 
Last edited:
again, is it possible the monitor is the cause? or the hard drives [2 ssds (500gb new, 120gb old) and an hdd (2tb old)] as well, is it possible 750watts isn't enough at peak? and what's the likelihood of getting a refurbished gpu back with the same problem?
 
have tested it with one stick of ram multiple times before, still happens. as for the crash, it's the same one. total system shutdown or whatever it's properly called. whole thing turned off then restarted, as per normal. no blue screen.
 
have been messing around with settings seeing if anything changes the crash and so far it's still totally random, far as i can tell. none of the temps go beyond acceptable levels, if they change from baseline at all. gonna disconnect most of the stuff and clean the contacts, see if that does anything. failing that, i'm gonna rma the card again.
 
Last edited by a moderator:
previously: https://forums.tomshardware.com/thr...ent-spontaneous-black-screen-crashes.3487489/

tl;dr (sorry it's still kinda long) of the previous is that this is a long term issue with my system, i replaced the motherboard/cpu/ram because initially i thought it was my MB finally ----ing out on me again (it was a refurbished board so i expected it) buuut the crash persisted through systems and switching from windows 7 to w10. rma'd the gpu next since the crash did not happen on my backup gpu (no longer true, just a lot more rare) and got a different card back, error persisting. completely reinstalled w10 fresh and did some diagnostic checks(ram, malware/adware sweeps) and core software updates (bios, chipset, gpu drivers forwards and backwards) before and after said reinstall.

the crash itself seems to only happen when the system is running a game (benchmarks do not seem to cause it no matter how long i run them) and the only game that seems to cause it more frequently than others is path of exile. lowering /increasing graphical settings reduce/increase its likelihood across the board.

since the previous post, i finally had the crash happen on my backup gpu before i sent in my current gpu to be rma'd again. since then, i replaced my ram with ram appropriately compatible with my MB which did improve performance a bit but the crash still happened. i am at a complete and total loss for what to do next.

the only thing i can even think to try is to replace the monitor and the psu(again) but with a higher wattage psu. other than that i am out of ideas and at the end of my rope.

i really don't want to have to lug this thing into a bestbuy or whatever to get a professional look at this (especially since the crash is seemingly random and i currently do not have a way to reliably cause it) or to have someone come to my home to look at it (for similar reasons). please send help.

edit: i don't know why this only just occurred to me, but is it possible this is an electrical wiring problem rather than a computer one? ie, takes too much amperage from a single line but with a mildly faulty line, causes said crash? crap now i need an extension cable.

edit2: been running the different psu for a few days now and i have not been able to get the crash to happen. it seems that it was specifically the corsair cx750m causing the issue even though i had a completely new one.
 
Last edited by a moderator:
  • Like
Reactions: andsettle
I have the exact same issue with an entirely different setup. i have an amd build thats doing this so the whole pc is entirely different EXCEPT we are using the same power supply. did you recently get a new case? im starting to consider maybe my case is drawing way to much power. please let me know if you find a fix. i feel like ive tried everything i can to including reinstalling and completely reformatting all drives. dont do it, it aint the move bud
 
  • Like
Reactions: Retrikaethan
Solution
I have the exact same issue with an entirely different setup. i have an amd build thats doing this. did you recently get a new case? im starting to consider maybe my case is drawing way to much power. please let me know if you find a fix. i feel like ive tried everything i can to including reinstalling and completely reformatting all drives. dont do it, it aint the move bud
my case is old, all the internals are new, tho, except for some of the cables and hard drives (also cases don't draw power unless they have like internal peripherals or something, mine doesn't have any non-fan components for it, though). i just now tried using different power outlets and it still consistently crashes so that's probably not it in my case. gonna try a socket in a different part of the house (electrically speaking) next to confirm for sure but i'm not super hopeful about that.

only thing i've really learned now is that in my case playing games in borderless fullscreen apparently increases the likelihood of the crash (making it a bit more reliable to test, if nothing else). not sure why that's a thing.
 
im almost certain after reading your issues and the issues of the other person in the corsair forum that it is almost definitely the psu. i will be exchanging mine tomorrow so ill keep you updated if that works. i think there is something with the psu that is incompatible with maybe certain hardware? im not sure but im almost certain its the psu now
 
  • Like
Reactions: Retrikaethan
i just replaced mine from a cx750m to a brand new from best buy cx750m. i think if we just venture away from corsair itll be fine.
if we're both experiencing the same issue with different builds with the only common factor(for all three of us including the guy from the link you mentioned) being that particular power supply then yeah, i'd say running away from that psu like our lives depended on it is the obvious answer here. it's kinda tenuous as a guess but i genuinely cannot see any other source. all of my other components have been replaced, that's the only one which was the same model when replaced.
 
  • Like
Reactions: andsettle