[SOLVED] Games crashing out of nowhere ?

[Cactaclysm]

Hi all, I recently upgraded my GPU from a PNY GTX1060 to an EVGA 1080ti Kingpin. Everything went well for the first month or two, until I had to underclock it in MSI afterburner to -90 , because games like overwatch and R6 would crash if it was at a factory clock. It ran well for a few months after that , but just recently the same issues are coming back, and I’m getting game crashes , specifically overwatch and R6 saying “rendering device has been lost”. I have tried everything. I have factory reset my computer, installed newest drivers , updated windows , rolled back drivers , uninstalling both games and reinstalling them, tinkering with the clock speed and voltage of my GPU and CPU, and rewiring my GPU to my PSU. My temps don’t go above 60 for either my GPU or CPU, so I have ruled out overheating as the issue .. is there anything that I may be overlooking? I’m getting quite desperate at this point.

Any help is appreciated.

 

[alceryes]

Did you buy the 1080ti Kingpin used or new?
Try stress testing it with Furmark. GPUs can fail without hitting thermal thresholds.

 

[Ralston18]

Update your post to include full system hardware specs and OS information.

Include PSU: make, model, wattage, age, condition?

What do you mean by "rewiring my GPU to my PSU."? What exactly did you do?

Look in Reliability History and Event Viewer for error codes, warnings, and even informational events that correspond with the game crashes.

Start with Reliability History. The timeline format may make it easier to discover when crashes started, stopped, and started again.

 

[Cactaclysm]

Update your post to include full system hardware specs and OS information.

Include PSU: make, model, wattage, age, condition?

What do you mean by "rewiring my GPU to my PSU."? What exactly did you do?

Look in Reliability History and Event Viewer for error codes, warnings, and even informational events that correspond with the game crashes.

Start with Reliability History. The timeline format may make it easier to discover when crashes started, stopped, and started again.

My current specs are:

Operating System
Windows 10 Pro 64-bit
CPU
Intel Core i7 7700K @ 4.20GHz 37 °C
Kaby Lake 14nm Technology
RAM
16.0GB Dual-Channel Unknown @ 1069MHz (15-15-15-35)
PSU
Rosewill 1000w Bronze
Motherboard
ASUSTeK COMPUTER INC. PRIME Z270-P (LGA1151) 30 °C
Graphics
27GL650F (1920x1080@144Hz)
Acer KG241Q (1920x1080@75Hz)
3071MB NVIDIA GeForce GTX 1080 Ti (EVGA) 25 °C
Storage
931GB Western Digital WDC WD10EZEX-08WN4A0 (SATA ) 19 °C
111GB KINGSTON SV300S37A120G (SATA-2 (SSD)) 17 °C

I have been thinking that perhaps it could be the power supply, seeing that reducing the power intake and underclocking it seemed to fix the problem temporarily before it arose again. When I said I rewired it , I changed a couple of the cords from my PSU and tried connecting them into different PCI-E slots on the back of the PSU. The PSU is 4 years old. I will use reliability now to see if I can pick something up..

 

[Cactaclysm]

Did you buy the 1080ti Kingpin used or new?
Try stress testing it with Furmark. GPUs can fail without hitting thermal thresholds.

I bought it used but I saw it running in the system perfectly fine and it was from a trusted friend.

 

[Cactaclysm]

Update your post to include full system hardware specs and OS information.

Include PSU: make, model, wattage, age, condition?

What do you mean by "rewiring my GPU to my PSU."? What exactly did you do?

Look in Reliability History and Event Viewer for error codes, warnings, and even informational events that correspond with the game crashes.

Start with Reliability History. The timeline format may make it easier to discover when crashes started, stopped, and started again.

Update on the Reliability History, when the game crashed it came up saying that "a problem with my hardware caused windows to stop working correctly"
Problem Event Name: LiveKernelEvent
Code: 141
Parameter 1: ffff830f9c67e1d0
Parameter 2: fffff80789b5bb7c
Parameter 3: 0
Parameter 4: 60c
OS version: 10_0_18363
Service Pack: 0_0
Product: 256_1
OS Version: 10.0.18363.2.0.0.256.48
Locale ID: 1033

 

[Ralston18]

Is that 111 GB Kingston Drive the boot drive? How full is the drive?

4 year old PSU - problematic just on its' own merits.

Swapping power cables (modular) can also cause problems.

Do you have a multi-meter and know how to use it. Or a knowledgeable family member or friend who does?

https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

Even thought the testing is not a full test (PSU not being loaded) any voltage(s) out of spec would be a red flag on the PSU.

 

[Cactaclysm]

Is that 111 GB Kingston Drive the boot drive? How full is the drive?

4 year old PSU - problematic just on its' own merits.

Swapping power cables (modular) can also cause problems.

Do you have a multi-meter and know how to use it. Or a knowledgeable family member or friend who does?

https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

Even thought the testing is not a full test (PSU not being loaded) any voltage(s) out of spec would be a red flag on the PSU.

I swapped the cables because the other ones had the same problem as well, so it didn’t matter.. unfortunately I don’t have a multi meter. The Kingston drive has 39GB left. I could put the 1080ti in one of my friends gaming computers with a newer more reliable PSU and see if it runs properly ?

 

[Ralston18]

Would expect that with 39 GB available space that that would not be a problem per se. However, 240 GB is now the general recommendation....

====

Overall that old PSU is a primary suspect. But, as a matter of elimination there are things you can try:

Power down, unplug, open the case.

Clean out dust and debris.

Verify by sight and feel that all connections, cards, RAM, and jumpers are fully and firmly in place.

Manually download drivers - start with the GPU. Reinstall and reconfigure.

Run the built in Windows troubleshooters. The troubleshooters may find and fix something.

Likewise run "sfc /scannow" and "dism" via the Command prompt.

References:

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

https://www.howtogeek.com/222532/ho...-system-files-with-the-sfc-and-dism-commands/

 

[Cactaclysm]

Would expect that with 39 GB available space that that would not be a problem per se. However, 240 GB is now the general recommendation....

====

Overall that old PSU is a primary suspect. But, as a matter of elimination there are things you can try:

Power down, unplug, open the case.

Clean out dust and debris.

Verify by sight and feel that all connections, cards, RAM, and jumpers are fully and firmly in place.

Manually download drivers - start with the GPU. Reinstall and reconfigure.

Run the built in Windows troubleshooters. The troubleshooters may find and fix something.

Likewise run "sfc /scannow" and "dism" via the Command prompt.

References:

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

https://www.howtogeek.com/222532/ho...-system-files-with-the-sfc-and-dism-commands/

I have done the SFC scan. So I will clean debris and dust , and I will check if I can replace the power supply. Thanks for the help.

 

[Zerk2012]

I have done the sfs scan. So I will clean debris and dust , and I will check if I can replace the power supply. Thanks for the help.

Put your 1060 back in and see if it works, you can put the 1080 back in your friends PC and see if it works.

I would test the video cards before just buying a new power supply.

A video card and start failing you underclocked it and might of bought a bit more time before the total failure.

 

[Cactaclysm]

Put your 1060 back in and see if it works, you can put the 1080 back in your friends PC and see if it works.

I would test the video cards before just buying a new power supply.

A video card and start failing you underclocked it and might of bought a bit more time before the total failure.

Ok, will do

 

[Cactaclysm]

Hey all, bit of an update.
Tried my GPU with my friends PSU, but it was 600w off brand and my GPU didn't fit in the case regardless.
I decided to use my logic of "less power, more success" and it worked. I turned it from -90 to -100 and now somehow the games are running absolutely flawlessly with maximum framerate.
However, I would really like to be able to run somewhere near factory clock, and I certainly have the budget to purchase a new PSU and RAM (preferably upgrading from G-Skill 2133 mHz) so do you folks think that perhaps new, more recognizable ram and a gold/plat certified , good name PSU would make any difference in the performance of my GPU? I feel as if it is being bottlenecked by my RAM and PSU, because my CPU is not too bad either.
Thanks again for all your help.

 

[Ralston18]

This:

"somehow the games are running absolutely flawlessly with maximum framerate. "

Leave well enough alone aka "if its not broke don't fix it".

Determine if, over time, the current performance remains the same and stable.

Use that time to look into viable options (PSU, RAM etc. as you mentioned) that will indeed enhance performance. And remain stable.

Read product documentation, FAQs, manufacturer forums, etc.. Verify all specs and compatibilities. Ensure that there will indeed be measurable improvement in the final build's performance.

Then plan out the necessary purchases and installation requirements.

Just my thoughts.....

 

[dorsai]

A quality PSU is always a good move...something Gold or Plat rated in the 800w range is an all around good idea.

As far as the 1080ti do you know if the previous owner ran it overclocked ? It sounds like age/degradation may be a factor...but I would change out the PSU and retest at full clock speed/volts.

 

[Cactaclysm]

A quality PSU is always a good move...something Gold or Plat rated in the 800w range is an all around good idea.

As far as the 1080ti do you know if the previous owner ran it overclocked ? It sounds like age/degradation may be a factor...but I would change out the PSU and retest at full clock speed/volts.

no, the card was not overclocked previously. I am ordering an 850w gold cert corsair psu as well as 32 gb of 3200mhz corsair ram.. hopefully it all goes well.

 

[Cactaclysm]

This:

"somehow the games are running absolutely flawlessly with maximum framerate. "

Leave well enough alone aka "if its not broke don't fix it".

Determine if, over time, the current performance remains the same and stable.

Use that time to look into viable options (PSU, RAM etc. as you mentioned) that will indeed enhance performance. And remain stable.

Read product documentation, FAQs, manufacturer forums, etc.. Verify all specs and compatibilities. Ensure that there will indeed be measurable improvement in the final build's performance.

Then plan out the necessary purchases and installation requirements.

Just my thoughts.....

Yes, I just had a 5 hour gaming session with mixed gameplay from overwatch & r6 and my temps were around the 50-60 degree mark.. however, at this point getting a better psu and better ram is just an upgrade, but as for now I wont tinker with any of my clocks, -100 seems to be the sweet spot, because the card is running at 1750 mhz even underclocked.. Doesn't that seem a bit high??
All in all though, for the past day or so it has been 100% stable, as I said no frame drops, good temperatures, and no tearing, buffering, or crashes to be reported in the reliability history.
Will update if something happens , or when the new parts arrive.
Thanks for all your comments, and truthfully I wasn't even aware that reliability history existed. You learn something everyday haha.
Cheers!

 

[Cactaclysm]

Hey all! Little update;

So both the RAM and PSU came in the mail and I installed them both. Upon installing them I immediately went to afterburner and reset all the clocks to 0 (as opposed to -113) andddddd
everything works perfectly! It seems that my old 1000w BRONZE psu was struggling to support the newer card, but the TX 850M Gold cert supply is working perfectly, so I am happy to finally have a fully functional computer with no drawbacks.
Thank you all for the help and support, had I not gotten feedback about the possibility of it being my PSU I definitely would not have had the courage to buy new parts as a fear of wasting my time and my money.

Cheers to all!