Question Old PC and now new PC shuts down and restarts under load ?

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
Need some help from someone with a big brain please and thank you.
So after 7 years of constant use with my old gaming laptop it finally let go a month and a half or so ago so i pick up a rig off marketplace for $900

The specs were -
CPU - i7 4790k
COOLER - NZXT Kraken X61
MOBO - ASUS Z97 Pro
RAM - Corsair 16GB DDR3 1600 Vengeance LPX
GPU - EVGA 980Ti
SSD - HyperX Savage 240GB
SSD - Samsung 850 EVO 500GB
PSU - Corsair HX850i
CASE - NZXT H440
FANS - 4x NZXT 120s
Also with my 2 usb HDDs 1tb and a 2tb

It ran perfectly for like a month then it randomly shut down then restarted when i was loading reserve on EFT, it was plugged into a surge protector so i thought it just reset so tried to load a map on EFT again and the same thing happened so i tried some other games and i was able to play GR breakpoint for like 30 mins without issue so i tried SWTOR for like 20 mins and it was fine until i tied to load my legacy page then it shut down since it has to load alot also tried War thunder and it was able to play full match on max settings then at the end i tabbed out then back in and it shut down.
After all that i was digging around and from what was said everywhere my conclusion was that it could be either the PSU or ram could be on its way out so it just decided to build the rig i was going to build before i found that pc. So i used its AIO kraken, the 980ti and the SSDs in my new rig.

The current rig specs are -
CPU - i7 10700K
COOLER - NZXT Kraken X61
MOBO - Asrock B560 PRO4
RAM - Corsair Vengeance LPX 32GB (4x8 3000mhz)
GPU - EVGA 980Ti
SSD - HyperX Savage 240GB
SSD - Samsung 850 EVO 500GB
PSU - Gigabyte 850W
CASE - Cougar Panzer
FANS - 4x Corsair AF120s Red + 2x Corsair ML140s Red
With the 2 usb HDDs aswell.

I put it together over the last couple of days but only booted it up last night after i got a brand new surge protector just to be safe, updated all the drivers, downloaded all my programs, games and what not including the bios. Doing just desktop things were a breeze and had no issues so i decided to boot up eft and see if its all good. Loaded into a private factory map and was fine so tried reserve scav run and it got did the full like 10-20min scav run without shutting down so though yay its fixed then tried to load my eft stash and it shutdown and restarted like before. I also tried a match of War thunder and it did the full match with like 30 seconds to go then my discord on my 2nd monitor went completely white so i tabbed out to check it and i minimized it then full screened it then it was good again but there was a NVidia high definition audio or something pop up was only there when it went white but not 100% if it cause it eitherway i tabbed back into war thunder and it shut down and restarted.

Considering everything is brand new except the GPU, AIO cooler and SSDs i assume its the GPU but just want opinions before i do anything further.
Got alot of screenshots of Speccy, Hardware monitor, eventviewer, cpuz, msi afterburner so you can see what i cannot probably.

- Overview Of Usual Needed Info-
cBmi0BA.png


- Hardware Monitor Idle Shots -
View: https://imgur.com/a/RQBJyrt


- Event Viewer Logs Of Crashes -
View: https://imgur.com/a/GyDiBiU


- Another Possible Crash Cause? - Alot of poeple had the issue where there hyperx cloud 2 headsets were causing random crashes and restarts but after alot of back and forth that error seems to be gone or atleast for now it is
0tn2E4O.jpeg

YUmWFzU.jpeg


Im working on trying to find someone local to borrow there GPU and wack it in to see if its just my GPU but that could take some time so thought id pop this up and see if anyone could help
Thank you :)
 

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
A bloke on Linus tech tip said to try clearing the motherboard CMOS, updating the bios, and changing XMP profiles before sending the GPU in. So being still fairly new to full on pc stuff but google was my best friend. I cleared the cmos, updated the bios from 1.80 to 2.10 and i wasnt on a xmp profile so the ram was running @2133 but turned the xmp on so now the ram runs @3000.

After i did all that i ran EFT and tried to load a shoreline run and it shutdown and restarted instantly so no good but that rules out those areas.
 

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
But now im even more confused. I had steam open and with dumb luck the 3dmark, pcmark 10 and vr mark bundle popped up and was on sale for 12 AUD instead of the usual 115 AUD so i thought well this will be a good thing to test with.



I brought it and ran the standard Timespy benchmark to test all components and it did it fine with no issues.

- Results - Max GPU Temp was 67.78c

View: https://imgur.com/a/zNwqN0s




Then i did the time spy Pure GPU 20 minute stress test and it did the whole thing with no issues.

- Results - Max GPU Temp was 69.19c

View: https://imgur.com/a/MtfsoVv




So i have no clue wtf is wrong now then lol
 
First thing you want to do it get rid of that gigabyte PSU before it explodes and takes half or more of your PC components with it. This most likely will not solve your issue. Did you reinstall windows when you made the new PC? I would try to uninstall and reinstall the graphics drivers using DDU just to rule out bad drivers completely. If I had to guess, there is something wrong with the 980 ti. Where did you get the old PC?
 

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
First thing you want to do it get rid of that gigabyte PSU before it explodes and takes half or more of your PC components with it. This most likely will not solve your issue. Did you reinstall windows when you made the new PC? I would try to uninstall and reinstall the graphics drivers using DDU just to rule out bad drivers completely. If I had to guess, there is something wrong with the 980 ti. Where did you get the old PC?

I got the first PC off facebook market place then the 2nd i built was with all brand new parts from Umart except the kraken CPU cooler, the 980Ti and the hyperX SSD also cheers for the warning about the PSU, didnt realise that the gigabyte PSUs have a 50/50 chance of being a time bomb so i decided to go big or go home and just ordered a Seasonic focus 850w 80+ gold for $225 AUD. As for the windows question, yes i completely wiped both my SSDs and reinstalled a fresh windows with a USB. I also just used DDU then to uninstall the 980Ti's drivers then used the latest drivers in Nvidia experience. Might be just the weather but the card was running cooler at idle then before but tried to run EFT and load into shoreline to test it and it still shutdown and restarted.

-Event Viewer Log -
View: https://i.imgur.com/y8JQYrA.jpeg
 

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
Things ive done so far -
  1. When i built the pc i completely wiped both my SSDs and reinstalled a fresh windows with a USB.
  2. Cleared the cmos, updated the bios from 1.80 to 2.10
  3. I was on the default ram settings so i put the xmp profile on and the went from running @2133mhz to now running @3000mhz.
  4. Brought 3D Mark so i ran the standard Timespy benchmark to test all components and it did it fine with no issues or crashes. - Results - Max GPU Temp was 67.78c
  5. Since the first one worked i did the time spy Pure GPU 20 minute stress test and it did the whole thing with no issues or crashes. - Results - Max GPU Temp was 69.19c
  6. ran CHKDSK on both my SSDs and both came back with no problems found.
  7. I didnt realise that the gigabyte PSUs have a 50/50 chance of being a time bomb so i decided to go big or go home and just ordered a Seasonic focus 850w 80+ gold for $225 AUD.
  8. I used DDU then to uninstall the 980Ti's drivers then used the latest drivers in Nvidia experience. Might be just the weather but the card was running cooler at idle then before.
  9. Switched from a single cable to 2 separate cables for the GPU.
Just tried to run EFT and load into shoreline to test it and it still shutdown and restarted.
I dont understand how it can do both benchmarks and not crash but when it loads into a map on eft it does.
 
Last edited:
Things ive done so far -
  1. When i built the pc i completely wiped both my SSDs and reinstalled a fresh windows with a USB.
  2. Cleared the cmos, updated the bios from 1.80 to 2.10
  3. I was on the default ram settings so i put the xmp profile on and the went from running @2133mhz to now running @3000mhz.
  4. Brought 3D Mark so i ran the standard Timespy benchmark to test all components and it did it fine with no issues or crashes. - Results - Max GPU Temp was 67.78c
  5. Since the first one worked i did the time spy Pure GPU 20 minute stress test and it did the whole thing with no issues or crashes. - Results - Max GPU Temp was 69.19c
  6. ran CHKDSK on both my SSDs and both came back with no problems found.
  7. I didnt realise that the gigabyte PSUs have a 50/50 chance of being a time bomb so i decided to go big or go home and just ordered a Seasonic focus 850w 80+ gold for $225 AUD.
  8. I used DDU then to uninstall the 980Ti's drivers then used the latest drivers in Nvidia experience. Might be just the weather but the card was running cooler at idle then before.
Just tried to run EFT and load into shoreline to test it and it still shutdown and restarted.
I dont understand how it can do both benchmarks and not crash but when it loads into a map on eft it does.
Do you have another graphics card you can try with the PC? You can try setting the power limit of the card to like 85% and then change the voltage to -50mv and see if that helps, if it does there is probably something slowly degrading the graphics card. This is expected for a 6 year old card factory overclocked card. I wonder what EVGA would say about getting it serviced or something like that. Usually EVGA goes above and beyond for their cards. If it turns out to be the card then I would not hurt to jump on their forums and ask.
 
Last edited:

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
Do you have another graphics card you can try with the PC? You can try setting the power limit of the card to like 85% and then change the voltage to -50mv and see if that helps, if it does there is probably something slowly degrading the graphics card. This is expected for a 6 year old card factory overclocked card. I wonder what EVGA would say about getting it serviced or something like that. Usually EVGA goes above and beyond for their cards. If it turns out to be the card then I would not hurt to jump on their forums and ask.
Sadly no only got the Ti.
I tried to lower the voltage but i can only set it to 0 in afterburner so i looked up a way around it and most said to set the power limit to 75 and the core clock to +100 to start with so i did that and tried to run eft and as soon as it tried to load the map my other monitor went black and both froze so i tried to alt tab and then the main monitor went all like kinda white with chequeredy and jumbled <Mod Edit> all over it so i had to hold the power button to stop it.
Since ive never messed with stuff besides fans speeds and whatnot, i dont know if what i changed caused it to do that or it proves that it is the GPU causing issues?
 
Last edited by a moderator:

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
I still have no clue what is causing it cause theres plenty of games that run perfectly. Like just then i was playing middle earth for 2-2 and a half hours straight on ultra with the hd mod at a constant capped 100fps and the hottest the card was during that time was only 73c, another is Mass effect andromeda on ultra ive played for like 30 mins each day with zero issues and at like 70-100fps also ghost recon breakpoint i can play on ultra at like a constant 70-90 fps without issues.

If i can actually get past a games loading the card preforms fantastically. Everyday browsing and all apps run like a breeze and it hasnt once crashed in windows, the pc only seems to shut down and restart when a game has a heavy loading section or some thing intense but at the same time it did both benchmarks without issues.
 

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
Good news but bad news at the same time, someone on reddit said to unplug the 980 and try stuff with the intergrated gpu so i unplugged the gpu and just use the intergrated one, did the update for it then ran EFT on the lowest res and settings. It took awhile to load into the maps as you expect but it loaded into the maps and out of them perfectly fine without crashes even had 35-40fps with the intergrated GPU lol. Then tried to load my stash and hideout and all worked perfect so closed the game and no issues.

So new gpu it is
 
Good news but bad news at the same time, someone on reddit said to unplug the 980 and try stuff with the intergrated gpu so i unplugged the gpu and just use the intergrated one, did the update for it then ran EFT on the lowest res and settings. It took awhile to load into the maps as you expect but it loaded into the maps and out of them perfectly fine without crashes even had 35-40fps with the intergrated GPU lol. Then tried to load my stash and hideout and all worked perfect so closed the game and no issues.

So new gpu it is
Yup, as I expected. So tarkov must be hitting your GPU in such a way that it uncovers some physical issue or degradation in that 980 ti causing it to crash and artifact the screen as it did in your last attempt. Here's the slightly good news, you could probably sell that card "for parts" for a decent chunk of money (100+ US). What part of the world are you in?
 

Ob1CanKnowMe

Prominent
Sep 28, 2021
18
0
510
Yup, as I expected. So tarkov must be hitting your GPU in such a way that it uncovers some physical issue or degradation in that 980 ti causing it to crash and artifact the screen as it did in your last attempt. Here's the slightly good news, you could probably sell that card "for parts" for a decent chunk of money (100+ US). What part of the world are you in?
I live in Australia so not sure if I can't palm it off over here for anything.