PC/Games crashing when using 1440p, resolution

Denpepe

Denpepe

2[H]4U
Joined
Oct 26, 2015
Messages
2,269
So my computer keeps crashing every now and then when playing games, had it on my old GTX780 and still happens on my GTX1080 but it usually only happens when playing on 1440p, not if I play on 1080p but it can work ok for days and then crash 2 times in 5 minutes.

I have no idea what going on, is it caused by windows 10? Is the fact that I'm using a G-sync monitor at 144hz responsible in some way?

I have looked up the error logs, but I'm not getting far with those, if I disable/remove what causes them, it changes to something else (so far Nvstreameragent, MSIactivex something and now some microsoft installer package)

Does anyone have any idea how to fix this, next try will be a clean windows install on a new SSD (just so I can keep my current install in case it does not fix it)

On a side note, I also have some image artifacting but only when using the Edge browser, pc has been tested by the shop but they could not find anything wrong.
 
PSU voltages? PSU make and model?

CPU temps?

GPU temps?

Try HWInfo64 for info.
 
If your cpu is overclocked try stock clocks. Also, what psu are you using? Could be that a higher gpu load stresses the psu and it can't hold the voltage where it should be.
 
PSU is a Corsair RM850

CPU temps 60-65°C

GPU temps are around the same

tried stock clocks on the CPU does not help
 
Edge artifacting isn't a GPU issue if it happens quite rarely. Its a known bug in Edge. Tend to be only 1 tab it happens in.

What does the crash errors say?

But try with a clean install, specially if you upgraded Windows 10 to a newer Windows 10. And make sure to use stock CPU settings while doing the install.
 
Those corsair RM power supplies don't have the best reputation
 
http://i.imgur.com/MYYt4Zv.jpg

screenshot of voltages and temps while playing

RM series might not be the best around, reviews were still good but's it's on the list of things that might need changing if OS reinstall does not work

Errors are the usual ID 41 from turning of the blocked PC and
"De Appx-bewerking RegisterPackageAsync op Microsoft.AccountsControl_10.0.14393.206_neutral_neutral_cw5n1h2txyewy is mislukt voor gebruiker"
"De Appx-bewerking RegisterPackageAsync op Microsoft.Windows.ShellExperienceHost_10.0.14393.206_neutral_neutral_cw5n1h2txyewy is mislukt voor gebruiker"
which are both id 10

Sorry for the dutch but that's the language windows is in don't have logs of the older ones anymore it seems but there were some NVstreameragent ones
 
Try running with a single stick of ram. Try changing that single stick of ram to another single stick of ram. Maybe that is causing problems. If a fresh install doesn't help then anything can be causing the problem. Motherboard, cpu, psu etc. Even if your voltages seem okay, a short small spike/drop in voltage may cause a crash and you wouldn't see it. You may have to resort to changing parts just to find where the problem lies.
I had a system that would crash or turn off seemingly randomly. After almost a year i figured out that the cpu socket was coming loose and the motherboard was bowed behind the socket. Probably from all the pressure my thermalright cooler was putting on it. Moving the cooler (or even slightly bumping the table) would cause the pc to shut down/crash.
 
try bumping the voltage on the RAM. almost every corsair kit ive ever used needed just a little bit more to be totally stable.
 
So I did a clean win 10 install on Friday evening, checked my logs today and I'm shocked

I have:
1 critical event
202 errors
258 warnings
3512 information messages
4937 succesfull controls
6 failed controls

I did turn my OC of for the moment, could all this come from bad ram or something? So far I haven't had any crashes yet but this amount of errors looks abnormal.
 
Post the log of the more important events. What was the critical and the majority of the errors/warning?
 
let's see, the critical one
"Er is door de Task Scheduler-service een RPC-initialisatiefout in 'RpcServerUseProtseq:ncacn_ip_tcp' gevonden. Aanvullende gegevens: foutwaarde: 14."

errors 88 of these
"Het wijzigen van de status van AppModel Runtime voor het pakket Microsoft.KillerInstinct-Win10_3.4.1.0_x64__8wekyb3d8bbwe voor de gebruiker (huidige status = 0x0, gewenste status = 0x20) is mislukt met 0x490."
"Het wijzigen van de status van AppModel Runtime voor het pakket Microsoft.WindowsAlarms_10.1608.2312.0_x64__8wekyb3d8bbwe voor de gebruiker (huidige status = 0x0, gewenste status = 0x20) is mislukt met 0x490."
all with different packages

and 45 of these
"In de machtigingsinstellingen toepassingsspecifiek wordt de machtiging Activeren niet verleend aan Lokaal voor de COM-servertoepassing met CLSID
{8D8F4F83-3594-4F07-8369-FC3C3CAE4919}
en APPID
{F72671A9-012C-4725-9D2F-2A4D32D65169}
aan de gebruiker NT AUTHORITY\SYSTEM SID (S-1-5-18) met het adres LocalHost (via LRPC) die wordt uitgevoerd in de toepassingscontainer Niet beschikbaar SID (Niet beschikbaar). Deze beveiligingsmachtiging kan worden gewijzigd met het beheerprogramma van Component Services."

96 warnings like this
"The application \Device\HarddiskVolume6\Windows\System32\svchost.exe with process id 8508 stopped the removal or ejection for the device PCI\VEN_8086&DEV_2822&SUBSYS_79771462&REV_31\3&11583659&0&B8"

Also ran an 8 hour memtest tonight, 0 errors
 
I've seen recently a guy online that had simular issues and it was power related.. that was everyone's guess but nothing would fix it.. and then he finally noticed a loose power connection thst wasn't fully seated at the modular psu side. So the satsuma power plug wasn't fully pushed in to the psu.

So check all the power connections!

Sometimes the actual wires can get pushed out at the connector if ea individual wire connection don't mate correctly. So even try pushing in the wires after checking to make sure the wires are fully seated in the connector.
 
Also I feel like a newbie here asking but we're are these error logs located so I can check my pc since it's a new build.. not that I'm having any issues yet but just to be sure.. I did notice the new Samsung 950 pro 512gb m.2 drive the first night showed 8 power down errors (in samsung software program.. can't remember error the name now :/ ) which was alarming as I didn't notice anything..
 
eclypse said:
Also I feel like a newbie here asking but we're are these error logs located so I can check my pc since it's a new build.. not that I'm having any issues yet but just to be sure.. I did notice the new Samsung 950 pro 512gb m.2 drive the first night showed 8 power down errors (in samsung software program.. can't remember error the name now :/ ) which was alarming as I didn't notice anything..
Click to expand...

What OS? Easiest way in 10 is Cortana search for event and click view event log or Event Viewer. Also you can go into the Windows Administrative Tools folder.
 
Most issues like this typically stem from a component overheating, lack of good/clean power, or driver/software configuration issues. That typically accounts for near 98% or more of the problems out there in my experience. A weak or failing power supply is a quite common source of issues - especially if you have an older unit and you've recently added a higher end graphics card that pulls more juice. The other 1-2% is caused by actual component failures. Re-seating all of your PC's connection and going back to stock on all OC's, with a fresh OS install and then running some diagnostics/stress software for your CPU/RAM/GPU is a good first step in attempting to locate the source of your issue.
 
other than a fresh install of windows, have you tried any of the other suggestions? one stick of RAM, bump RAM voltage, unplug and reseat ALL cables, disable g-sync, disable OCs etc? you errors appear to be random so look at your psu and ram first. more logs about random crashes aren't gonna help.
 
Like I said, I did an 8 hour memtest last night which came back clean, windows was reinstalled on a clean SSD with OC turned off and it is still off. I checked my cables this evening again, and all are firmly in place (I also reseated the GPU and soundcard). The PSU is a little over 1 year old, and I'm assuming my new GPU GTX1080 draws less power then the old GTX780.

I appreciate all the suggestions, and while I do have a lot of errors, so far the computer has not crashed yet (knocks on wood). I'm going trough things step by step so that when I find it I will know what exactly did the trick.

If the errors keep comming I will swap in another PSU (seasonic 650w) I bought 6 months ago to build a spare computer.
 
Somethings a miss. Remove everything you don't need encase it's something faulty.. run it as bare as you can cpu/memory/os drive.


Edit:

If it's easy to do remove the motherboard from the case and run your system layer out on cardboard.. that would rule out something grounding out like a standoff out of place or a loose screw that fell behind something that you didn't notice.

Yeah if you have a different psu I'd try that as well.

One last thing would be to get a kill-a-watt power tester and plug it in to the wall outlet your using and select voltage. It will show you the voltage you pulling from the wall with nothing plugged into it. Safe if it shows 118-124v. If you get a real low reading you got a problem but it's possable during peak power hours and if your getting a really high number like 130-140+ volts then your getting way too much power from your box and that can ruine electronics. So if you have a lot of stuff breaking around the house it could be way.
 
Seems I have found the issue (only 1 error in the last 24 hours anymore) Seems I clumsily :oops: bent some of the pins on the USB 3 header on the mobo when installing the GPU, I could not bend all of them back but if there was a short it looks like it's gone now.

Well I hope that was it and it's fixed, still keeping an eye on it to be sure.
 
Denpepe said:
Seems I have found the issue (only 1 error in the last 24 hours anymore) Seems I clumsily :oops: bent some of the pins on the USB 3 header on the mobo when installing the GPU, I could not bend all of them back but if there was a short it looks like it's gone now.

Well I hope that was it and it's fixed, still keeping an eye on it to be sure.
Click to expand...

I had that type of issue too. Tho on the front of my chassis. I couldn't understand why my PC wouldn't start and it was due to a bent pin making a short. It could also randomly crash for the same reason until it got completely angry. Very rare tho.
 
You must log in or register to reply here.
Back
Top