Question PC randomly restarting when gaming - please help me

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
Suddenly my computer start to restarting while gaming (PUBG, OVERWATCH, CS GO) .

What is happening?
Everytime I start playing my computer suddenly restarts without any warning or error. Sometimes takes 20 minutes after starting playing to restart.

Things I tried so far;
  • DDU reinstalling nvidia driver (I already tried downgrade, didn't work either)
  • Formating the computer
  • Upgrading all drivers
  • Changing energy plan to perfomance ( and other options that I found on the internet)
  • Scandisk
  • Memory check
  • Clean the computer
  • Update BIOS
  • Checking temperature on gpu and cpu
  • Changing admin account (Works for 20 minutes than the computer restart again)
  • Turning off fast start up

Information I gathered and noticed so far;
  • In the event viewer there is a critical problema called Kernel 41
  • When I stress the gpu or the cpu alone in the software FurMark nothing happens, however if I stress both simultaneously the PC restart automally

This just happening when I am playing or stressing gpu and cpu simultaneously. I don't know what to do anymore.

I found some topics that says the problem isn't the PSU on the internet, even switching the PSU the problem persist. My bet is on the PSU, but I don't know if this is correct. If someone could help me, I will be very gratefull.

My computer spec;
CPU: I7 - 4790k
GPU: Geforce GTX 1060 6gb
Memory: 8 gb ram 1600 Mhz
MOBO: Asus ranger VII
PSU: EVGA 600W 80 PLUS BRONZE
2 SDD: 120 gb and 240 gb
Water cooler corsair h80i
 

PC Tailor

Illustrious
Ambassador
Formating the computer
By this do you mean you clean installed windows?

Checking temperature on gpu and cpu
What were these temperatures? and did you monitor each individual CPU core not just the overall package?

In the event viewer there is a critical problema called Kernel 41
Don't worry about this error. It basically means power was lost unexpectedly, you'd get the same error appear simply by pressing the reset button. It doesn't indicate what the problem is, just that the problem is causing a shutdown (which you already know from your screen going black!).

even switching the PSU the problem persist
What did you do exactly to test this? Also what exact model EVGA are you using?

Does the issue only happen under load?
Can you verify if the issue occurs in safe mode?
Are you running any overclocks at all?
 

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
By this do you mean you clean installed windows?
Yes.

What were these temperatures? and did you monitor each individual CPU core not just the overall package?
I didn't, just overall package. How can I do it?

What did you do exactly to test this? Also what exact model EVGA are you using?
I took off the GPU and tried to stress the onboard, however I did not try to do it with the cpu simultaneously . I don't think this provide any kind of information, I just follow internet instruction.
The model is ATX EVGA 600W 80 Plus Bronze - 100-B1-0600-KR.

Does the issue only happen under load?
Yes

Are you running any overclocks at all?
So far as I know, I am not. Not sure.

Can you verify if the issue occurs in safe mode?
Ok.
How can I run the stress tests on safe mode? I tried with FurMark but required the geforce drivers.

Thank you very much for your time!
 
Last edited:

PC Tailor

Illustrious
Ambassador
The fact that it has occured even after a clean install can often indicate a hardware problem.
I didn't, just overall package. How can I do it?
Don't get me wrong, usually nowadays most software say the package temperature is actually the highest core temp, but not always. You can use software like HWInfo to check the temps of each individual core. Because basically it only takes one core to overheat to cause a problem.

The model is ATX EVGA 600W 80 Plus Bronze - 100-B1-0600-KR.
So the EVGA B1 series is low end PSU range. Random restarts can be anything practically, but are usually PSU or GPU related once software has been eliminated. A poorer quality PSU can certainly cause this as it may struggle to output a stable power when put under load, or when other components are put under load and require more power.

Equally a GPU faulting can also cause this kind of issue. I would probably remove GPU and stress test CPU to see if the issue repeats. But you would usually look at PSU first, especially if they are a poorer quality PSU.
So far as I know, I am not. Not sure.
Resetting CMOS/BIOS will remove any OC if you had any.

How can I run the stress tests on safe mode? I tried with FurMark but required the geforce drivers.
If the issue only happens under load, this becomes more difficult as obviously all other drivers are disabled, the only real way you would be able to stress them in a "safe mode" like environment would be doing it from a bootable repair disk/USB that has a stress testing utility on it.

To confirm again, did you say you actually swapped the PSU entirely and the issue still occurred?
If this is the case, I would be looking at GPU.
Obviously it never hurts to cover all of the basics first also, and the most common causes of errors (for example bent pins on the CPU socket) so following this guide will cover all of the essentials also: https://forums.tomshardware.com/thr...ng-about-post-boot-no-video-problems.1285536/
 

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
Don't get me wrong, usually nowadays most software say the package temperature is actually the highest core temp, but not always. You can use software like HWInfo to check the temps of each individual core. Because basically it only takes one core to overheat to cause a problem.
I checked again, the CPU (all cores) temperature on stress didn't pass the 80ºC.

So the EVGA B1 series is low end PSU range. Random restarts can be anything practically, but are usually PSU or GPU related once software has been eliminated. A poorer quality PSU can certainly cause this as it may struggle to output a stable power when put under load, or when other components are put under load and require more power.

Equally a GPU faulting can also cause this kind of issue. I would probably remove GPU and stress test CPU to see if the issue repeats. But you would usually look at PSU first, especially if they are a poorer quality PSU.
I hope is only the PSU, would be easiest to fix this. Seems to be the PSU, once the problem only occur when CPU and GPU are both over load, if is just only one of them, nothings happens. Would you recommend any kind of PSU for this setup?


Resetting CMOS/BIOS will remove any OC if you had any.
I already did, nothing changed. I will tried again to make sure.

To confirm again, did you say you actually swapped the PSU entirely and the issue still occurred?

I didn't swapped the PSU. I read on the internet that some people on similar situations that swapped the PSU didn't have success fixing the problem.

Obviously it never hurts to cover all of the basics first also, and the most common causes of errors (for example bent pins on the CPU socket) so following this guide will cover all of the essentials also: https://forums.tomshardware.com/thr...ng-about-post-boot-no-video-problems.1285536/
I will check.


Again, thank you very much!
 

PC Tailor

Illustrious
Ambassador
I checked again, the CPU (all cores) temperature on stress didn't pass the 80ºC.
80 degrees certainly is on the higher side, it is advisable to keep temps below 80 degrees at all times, but thermal throttling would likely occur around 90 degrees.
Either way, i wouldn't expect normal load to push your cores further than a CPU stress test. Prime95 was it?

I didn't swapped the PSU. I read on the internet that some people on similar situations that swapped the PSU didn't have success fixing the problem.
Random restarts again, can be anything, but is often indicative of a PSU issue. I would swap the PSU out with a good quality, known working unit and retest. Unfortunately random restarts can have a different cause for anyone, so you will undoubtedly come across forums where the PSU swap didn't help, but it certainly is the first port of call for restarts if you have eliminated firmware/software.

Follow the guide posted above, then I would look at swapping PSU. The B1 series is poor quality so it could well be.

Would you recommend any kind of PSU for this setup?
If I do give a recommendation, just keep this in mind:

The PSU is the component that supplies power to all other components, so it should be the last component to skrimp on, as a cheap PSU = poor quality PSU = risk to your other components. So I would never personally recommend a PSU below a tier 3 (and even then I will always aim Tier 2+). So it does depend on your budget, but paying more for your PSU on a good quality unit, does ensure the safety of your components and usually gives you a much better warranty. It's not always the case, but more cost usually means better quality from reputable brands.

But as a basic level - I would refer to this https://www.gamingscan.com/psu-hierarchy/ and aim for anything Tier 2 and above.

For example some good pointers:
  • Basically ANY seasonic PSU.
  • Basically ANY superflower PSU.
  • Newer Corsair CX series (not old ones), HX, TX-M
  • Antec Edge or True Power Classics
  • EVGA GS, G1, G2, G3 etc.
 
Last edited:

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
80 degrees certainly is on the higher side, it is advisable to keep temps below 80 degrees at all times, but thermal throttling would likely occur around 90 degrees.
Either way, i wouldn't expect normal load to push your cores further than a CPU stress test. Prime95 was it?
I was using the FurMark, the print below show how is the temperature on overload test in this program.
D9xyB72XsAAptR4.jpg:large


Random restarts again, can be anything, but is often indicative of a PSU issue. I would swap the PSU out with a good quality, known working unit and retest. Unfortunately random restarts can have a different cause for anyone, so you will undoubtedly come across forums where the PSU swap didn't help, but it certainly is the first port of call for restarts if you have eliminated firmware/software.

Follow the guide posted above, then I would look at swapping PSU. The B1 series is poor quality so it could well be.

Ok.

If I do give a recommendation, just keep this in mind:

The PSU is the component that supplies power to all other components, so it should be the last component to skrimp on, as a cheap PSU = poor quality PSU = risk to your other components. So I would never personally recommend a PSU below a tier 3 (and even then I will always aim Tier 2+). So it does depend on your budget, but paying more for your PSU on a good quality unit, does ensure the safety of your components and usually gives you a much better warranty. It's not always the case, buy more cost usually means better quality from reputable brands.

But as a basic level - I would refer to this https://www.gamingscan.com/psu-hierarchy/ and aim for anything Tier 2 and above.

For example some good pointers:
  • Basically ANY seasonic PSU.
  • Basically ANY superflower PSU.
  • Newer Corsair CX series (not old ones), HX, TX-M
  • Antec Edge or True Power Classics
  • EVGA GS, G1, G2, G3 etc.


Thank you very much for clarifying it.
 

PC Tailor

Illustrious
Ambassador
Temps are definitely higher than they should be and some thermal throttling might be occurring during the stress test, however most stress tests far exceed the load that would be put on your CPU in nearly all applications, so I'd expect stress test results to be much worse.

However it does indicate you may want to improve your cooling as you'll want to keep all cores under at least 85 degrees C, under 80 degrees ideally. (You can see core temps below the temperature section of your screenshot).
 
  • Like
Reactions: Abhygail

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
Temps are definitely higher than they should be and some thermal throttling might be occurring during the stress test, however most stress tests far exceed the load that would be put on your CPU in nearly all applications, so I'd expect stress test results to be much worse.

However it does indicate you may want to improve your cooling as you'll want to keep all cores under at least 85 degrees C, under 80 degrees ideally. (You can see core temps below the temperature section of your screenshot).

Thank you for replying. I sent the computer for maintenance, the technician said that he tested the psu on overload, nothings happened. He also said that swapped the psu and tested on overload aswell, nothing happened. However when I got home and tested the computer its still restarting on overloard.


And it seems the temperature got worse, even though the technician said that he changed the thermal paste. I'll return the pc for them check again, but I have no idea why this is only happening at home, makes no sense.

Could this be a problem with water cooler? There is a way to test it?
 

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
Update
I went at maintenance store and the technician show me that was working normal there. One thing I noticed that was very different from my house it was the monitor, he was using one with 14 inches.

I think this is might be software problem.
 
Last edited:

PC Tailor

Illustrious
Ambassador
Update
I went at maintenance store and the technician show me that was working normal there. One thing I noticed that was very different from my house it was the monitor, he was using one with 14 inches.

Could the monitor cause this problem?
what the technician actually test?

The monitor itself won't cause a shutdown, but load on the GPU could. But directly, no, the monitor can't be causing the shutdown. The PC definitely shuts down? not just loses signal?
 
  • Like
Reactions: Abhygail

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
what the technician actually test?

The monitor itself won't cause a shutdown, but load on the GPU could. But directly, no, the monitor can't be causing the shutdown. The PC definitely shuts down? not just loses signal?

The technician said he test CPU, PSU, USB, didn't get more details. I tested another monitor, nothing changed.
The PC definitely shuts down? not just loses signal?
PC just restart.


I was searching on internet and found some articles that indicates this could be a problem related to driver of windows update. I remember that, when this problem started, I downgrade the nvidia driver and the computer stayed stable for while. Later I updated W10 and everything changed after a while.
 

PC Tailor

Illustrious
Ambassador
Doubted this one would NOT be hardware because you said you clean installed previously.
Problem with technicians is you don't know what tests they performed exactly. It could still well be PSU depending on how they did the test.

Confirm if the issue still occurs if you remove the GPU entirely, or better put a different GPU in and see if the issue repeats.
Goes back to my first post, probably either GPU or PSU initially.

RAM can cause all manner of issues, but is harder test without more than 1 module.
 
Last edited:
  • Like
Reactions: Abhygail

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
Doubted this one would be hardware because you said you clean installed previously.
Problem with technicians is you don't know what tests they performed exactly. It could still well be PSU depending on how they did the test.

Confirm if the issue still occurs if you remove the GPU entirely, or better put a different GPU in and see if the issue repeats.
Goes back to my first post, probably either GPU or PSU initially.

RAM can cause all manner of issues, but is harder test without more than 1 module.
I see.

I was able to play a game (overwatch) for more than an hour without any problem, usually don`t pass the lobby, however in low settings, this indicate that a problem with psu?
Again, thank you very much.
 

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
Was this after you removed the GPU?
With the GPU.
I noticed another event, if I set water cooler to run more faster during this game that mentioned before the pc restart. This corroborate with the hypothesis that problem might be the PSU? Once the computer is requiring more energy...
 
Last edited:

rawkenrawbie

Reputable
Apr 27, 2016
1
0
4,510
I am having the EXACT SAME ISSUE:

Previously - Windows 8.1
ASRock Z97 Pro4 - P2.20
Pentium G3258 - STABLE OC @ 4.5Ghz @ 1.280v - Tested stable 8 hrs Prime95
Dual Channel DDR3 2 x 4GB G.Skill (NO OC)
GTX 960 4GB (NO OC)
EVGA 650W PSU [80 Plus Gold CERTIFIED]
AiO Corsair Double Radiator Sealed Cooling Loop

- Checked temps during all testing and with both GPU and CPU, there is no apparent overheating; GTX 960 throttles as intended according to TechpowerGPU logs.

-Event viewer states unexpected shutdown:
1) The computer has rebooted from a bugcheck. The bugcheck was: 0x0000009f (0x0000000000000004, 0x000000000000012c, 0xffffa78e24cd6080, 0xfffff8053c9e79e0). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 4485f52a-4078-4f23-9dcf-e640ea03b1fa. (This is not likely for this issue, but it shows up as critical.

2) 11 events with the following message: {This is EXACTLY the event that occurs each time I have been shut down.}
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Log Name: System | Event ID 41 | Level: Critical | Source: Kernel-Power | Task Category (63)
Keywords: (70368744177664),(2)

3) 11 events with the following message: {Also in conjunction with the above event}
The previous system shutdown at 3:54:47 PM on ‎7/‎6/‎2019 was unexpected.
Log Name: System | Event ID: 6008 | Level: Error | Source: EventLog
Task Category: None | Keywords: Classic

The following trouble shooting has been performed:

- Uninstalled/reinstalled the GTX 960 drivers and Game Ready Software; no change

- Double Checked with MSI Afterburner to ensure GPU wasn't OC'd; attempted to move the power profile between 90-98%; STILL SAME RESULTS

- Logged temps with CoreTemp and TechpowerGPU: CPU/GPU NEVER remotely reached critical temps; GPU throttles and hovers around 80C as expected, CPU is watercooled and never reaches about 52degrees C

- Reseated GTX 960

- Plugged GPU and CPU into different available slots; still the same results

- Getting desperate.... I purchased Windows 10 Pro, performed a hard format and clean installed, reinstalled drivers and Game Ready drivers; STILL THE SAME FRIKKIN result!

- Contacted nVidia support (They suggested updating WIndows 10, updated and verified; STILL THE SAME RESULT

The following games have been tested to see if this was specific to any game:

FFXV (BEST RESULT: Able to access Tutorial Mode and complete, but once main game is accessed, it immediately shuts down for 10 seconds, then restarts)

Fortnite (BEST RESULT: 75% of the time, it will allow me to play a full game session, but immediately shuts down/reboots once the game is over and attempts to go back to the main menu; attempted to scale down to 720p: I was able to get through 1 1/2 games before the computer abruptly shuts down, then restarts, mid-game) [Again, Checked GPU/CoreTemp logs and no abnormal or critical temperatures reached] {One other quirky thing I tried... Downscaled to 720p, then reverted back to 1080p, immediate shutdown, then restart}

Roblox (Works fine in Windowed Mode, cuts out between 30-60 minutes in Full Screen Mode)

Emulators: NES - PSX/N64 works fine, no problems
 
Last edited:

PC Tailor

Illustrious
Ambassador
I am having the EXACT SAME ISSUE:

Previously - Windows 8.1
ASRock Z97 Pro4 - P2.20
Pentium G3258 - STABLE OC @ 4.5Ghz @ 1.280v - Tested stable 8 hrs Prime95
Dual Channel DDR3 2 x 4GB G.Skill (NO OC)
GTX 960 4GB (NO OC)
EVGA 650W PSU [80 Plus Gold CERTIFIED]
AiO Corsair Double Radiator Sealed Cooling Loop

- Checked temps during all testing and with both GPU and CPU, there is no apparent overheating; GTX 960 throttles as intended according to TechpowerGPU logs.

-Event viewer states unexpected shutdown:
1) The computer has rebooted from a bugcheck. The bugcheck was: 0x0000009f (0x0000000000000004, 0x000000000000012c, 0xffffa78e24cd6080, 0xfffff8053c9e79e0). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 4485f52a-4078-4f23-9dcf-e640ea03b1fa. (This is not likely for this issue, but it shows up as critical.

2) 11 events with the following message: {This is EXACTLY the event that occurs each time I have been shut down.}
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Log Name: System | Event ID 41 | Level: Critical | Source: Kernel-Power | Task Category (63)
Keywords: (70368744177664),(2)

3) 11 events with the following message: {Also in conjunction with the above event}
The previous system shutdown at 3:54:47 PM on ‎7/‎6/‎2019 was unexpected.
Log Name: System | Event ID: 6008 | Level: Error | Source: EventLog
Task Category: None | Keywords: Classic

The following trouble shooting has been performed:

- Uninstalled/reinstalled the GTX 960 drivers and Game Ready Software; no change

- Double Checked with MSI Afterburner to ensure GPU wasn't OC'd; attempted to move the power profile between 90-98%; STILL SAME RESULTS

- Logged temps with CoreTemp and TechpowerGPU: CPU/GPU NEVER remotely reached critical temps; GPU throttles and hovers around 80C as expected, CPU is watercooled and never reaches about 52degrees C

- Reseated GTX 960

- Plugged GPU and CPU into different available slots; still the same results

- Getting desperate.... I purchased Windows 10 Pro, performed a hard format and clean installed, reinstalled drivers and Game Ready drivers; STILL THE SAME FRIKKIN result!

- Contacted nVidia support (They suggested updating WIndows 10, updated and verified; STILL THE SAME RESULT

The following games have been tested to see if this was specific to any game:

FFXV (BEST RESULT: Able to access Tutorial Mode and complete, but once main game is accessed, it immediately shuts down for 10 seconds, then restarts)

Fortnite (BEST RESULT: 75% of the time, it will allow me to play a full game session, but immediately shuts down/reboots once the game is over and attempts to go back to the main menu; attempted to scale down to 720p: I was able to get through 1 1/2 games before the computer abruptly shuts down, then restarts, mid-game) [Again, Checked GPU/CoreTemp logs and no abnormal or critical temperatures reached] {One other quirky thing I tried... Downscaled to 720p, then reverted back to 1080p, immediate shutdown, then restart}

Roblox (Works fine in Windowed Mode, cuts out between 30-60 minutes in Full Screen Mode)

Emulators: NES - PSX/N64 works fine, no problems
You will probably want to start your own thread my friend.
 
  • Like
Reactions: Abhygail

Abhygail

Reputable
Jan 10, 2015
15
0
4,510
I am having the EXACT SAME ISSUE:

Previously - Windows 8.1
ASRock Z97 Pro4 - P2.20
Pentium G3258 - STABLE OC @ 4.5Ghz @ 1.280v - Tested stable 8 hrs Prime95
Dual Channel DDR3 2 x 4GB G.Skill (NO OC)
GTX 960 4GB (NO OC)
EVGA 650W PSU [80 Plus Gold CERTIFIED]
AiO Corsair Double Radiator Sealed Cooling Loop

- Checked temps during all testing and with both GPU and CPU, there is no apparent overheating; GTX 960 throttles as intended according to TechpowerGPU logs.

-Event viewer states unexpected shutdown:
1) The computer has rebooted from a bugcheck. The bugcheck was: 0x0000009f (0x0000000000000004, 0x000000000000012c, 0xffffa78e24cd6080, 0xfffff8053c9e79e0). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 4485f52a-4078-4f23-9dcf-e640ea03b1fa. (This is not likely for this issue, but it shows up as critical.

2) 11 events with the following message: {This is EXACTLY the event that occurs each time I have been shut down.}
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Log Name: System | Event ID 41 | Level: Critical | Source: Kernel-Power | Task Category (63)
Keywords: (70368744177664),(2)

3) 11 events with the following message: {Also in conjunction with the above event}
The previous system shutdown at 3:54:47 PM on ‎7/‎6/‎2019 was unexpected.
Log Name: System | Event ID: 6008 | Level: Error | Source: EventLog
Task Category: None | Keywords: Classic

The following trouble shooting has been performed:

- Uninstalled/reinstalled the GTX 960 drivers and Game Ready Software; no change

- Double Checked with MSI Afterburner to ensure GPU wasn't OC'd; attempted to move the power profile between 90-98%; STILL SAME RESULTS

- Logged temps with CoreTemp and TechpowerGPU: CPU/GPU NEVER remotely reached critical temps; GPU throttles and hovers around 80C as expected, CPU is watercooled and never reaches about 52degrees C

- Reseated GTX 960

- Plugged GPU and CPU into different available slots; still the same results

- Getting desperate.... I purchased Windows 10 Pro, performed a hard format and clean installed, reinstalled drivers and Game Ready drivers; STILL THE SAME FRIKKIN result!

- Contacted nVidia support (They suggested updating WIndows 10, updated and verified; STILL THE SAME RESULT

The following games have been tested to see if this was specific to any game:

FFXV (BEST RESULT: Able to access Tutorial Mode and complete, but once main game is accessed, it immediately shuts down for 10 seconds, then restarts)

Fortnite (BEST RESULT: 75% of the time, it will allow me to play a full game session, but immediately shuts down/reboots once the game is over and attempts to go back to the main menu; attempted to scale down to 720p: I was able to get through 1 1/2 games before the computer abruptly shuts down, then restarts, mid-game) [Again, Checked GPU/CoreTemp logs and no abnormal or critical temperatures reached] {One other quirky thing I tried... Downscaled to 720p, then reverted back to 1080p, immediate shutdown, then restart}

Roblox (Works fine in Windowed Mode, cuts out between 30-60 minutes in Full Screen Mode)

Emulators: NES - PSX/N64 works fine, no problems

Did you try to switch the psu? My problem was the PSU, I bought a new one and worked. Games that do not demand too much from the GPU or the CPU works, but just for a while, that is because the PSU isn't supplying enough energy (In my case).