Question PC Restarting When Gaming

Oct 20, 2019
9
0
10
Hey guys,

I am having problems with my PC. It restarts everytime and the screen goes completely black and it turns on again when I play graphical games like Rage 2 and Forza Horizon 4.

Specs
Processor: i7-6700k
GPU: Asus RTX 2070 Super
PSU: EVGA 700B


It happened a week after I installed the new PSU and GPU. It crashed and I let the PC be for a day and it didn't crash in a week now it's crashing more often. In event viewer it says Critical Power Kernel 41 (61) It only crashes when I'm playing games.

I'm thinking of getting a new PSU, but I'm wondering if it's my PSU or GPU, or anything else thanks!
 
Oct 20, 2019
9
0
10
use occt and stress test psu for about 10 minutes and take note of voltages values before and on test also what are the temp in system use hwinfo to find them .
In occt it says only voltage of GPU and CPU am I supposed to write those down when benchmarking? And with hwinfo which temps should I check because my CPU and GPU temps are fine. Sorry if I sound stupid
 
Mainly you just want to make sure it doesn't crash while running the test. Let it run at least 30 minutes.

If I may ask, where did you buy that power supply? Was it listed as new or refurbished? That unit hasn't sold as new here in the U.S. for a couple years (it was first released back in 2013), so just curious.
 
Oct 20, 2019
9
0
10
Mainly you just want to make sure it doesn't crash while running the test. Let it run at least 30 minutes.

If I may ask, where did you buy that power supply? Was it listed as new or refurbished? That unit hasn't sold as new here in the U.S. for a couple years (it was first released back in 2013), so just curious.
I bought it new I think on a Swedish site it says it's new and the sites trusted, but it only crashes when gpu is under load because when I stress tested the CPU nothing crashed but when I did both it crashed in like 42 seconds so I can't stress test
 
Oct 20, 2019
9
0
10
voltage on the 12 volts line and same for 5 volts and 3 volts from occt and hwinfo sensor page these are the one who will tell us if psu is going out also remove gpu after reading and use onboard to see what happens in games .
Does the kernel power 41 (62) critical shutdown not properly in event viewer mean anything should I get a new PSU and what do you mean with 12 volts line and 5 and 3 from occt??
 
Hey guys,

I am having problems with my PC. It restarts everytime and the screen goes completely black and it turns on again when I play graphical games like Rage 2 and Forza Horizon 4.

Specs
Processor: i7-6700k
GPU: Asus RTX 2070 Super
PSU: EVGA 700B


It happened a week after I installed the new PSU and GPU. It crashed and I let the PC be for a day and it didn't crash in a week now it's crashing more often. In event viewer it says Critical Power Kernel 41 (61) It only crashes when I'm playing games.

I'm thinking of getting a new PSU, but I'm wondering if it's my PSU or GPU, or anything else thanks!


Any time you have an obvious reference point the chances the problem lies there is ~100%.
 
Oct 20, 2019
9
0
10
Any time you have an obvious reference point the chances the problem lies there is ~100%.
So I should get a new PSU? If so which one do you recommend at 650 watts and a Gold, and are you sure it's my PSU because I got it new like a month ago is it possible that new PSUs come faulty? Thanks for your help :D
 
I had a similar problem with my rig recently. It would instantly restart in-game, although at least a few things might be different. It was really only happening in 1 game for me as far as I can tell, and it was only in 4k resolution. My voltages, temps, everything was normal at the time of the restarts and it never restarted in 3dMark, only in Far Cry 5.

Since it costs nothing but time, I can suggest to you to try a software solution first before you spend any money to replace the PSU. First, just make sure your voltages (3.3v,5v,12v) are ok as Scout suggested. If they're good, you should uninstall the GPU drivers and Geforce Experience using DDU in Safe Mode while NOT connected to the internet and then install your drivers again from a pre-downloaded installer file for version 431.60 which is an older and presumably more stable driver than the recent ones. To paraphrase the advice given to me over at the nVidia forums (Looks like a lot, but I've just been very specific about each step):

1) Download the driver:
http://us.download.nvidia.com/Windows/431.60/431.60-desktop-win10-64bit-international-whql.exe

2) Download DDU:
https://www.wagnardsoft.com/content/display-driver-uninstaller-ddu-v18019-released

3) Disconnect Network Cable / Disable WIFI adapter

4) Run msconfig by typing it into the Windows start menu. click "Boot" tab, check "Safe Boot" option then "Apply" and reboot your PC.

5) Install and run DDU and select "Options" from the program's menu in the upper left.

6) Check ALL the boxes for nVidia as shown below:
images


NOTE: the warning about C:\NVIDIA sounds worse than it is, but it means that DDU will delete the folder where downloads of previous nVidia driver install files (like the one you manually downloaded in step 1) have accumulated. If you have a slow internet connection and you would like to keep these files just in case, then don't check this box. I just got rid of them all myself.

7) Click "Close" in the options menu, select "nVidia" where is says "Selected Graphics Driver" and then click "Clean and do NOT restart"

8) Open msconfig again and make sure Safe Boot is unchecked (uncheck and Apply if necessary) and then reboot your PC.

9) Now out of Safe Mode and back in normal Windows mode, find the nVidia driver 431.60 file that you downloaded and start the install process. Select CUSTOM and make sure ONLY "nVidia Driver" and "nVidia NGX" are checked (do NOT install PhysX, audio drivers, Geforce Experience, or anything else at this time). Now, check the "Clean Install" box, finish the installation, and then reboot your PC.

10) Reconnect Network Cable / Re-enable WIFI

11) If using any of these, remove them:
Uninstall iCue | Link
Uninstall CAM
If using a PCIe Riser, remove it.

12) Retest with games/software that are giving you restarts.

13) If everything is ok by now and you're not getting restarts anymore, then open the nVidia driver installer again and install PhysX and retest. If still ok, add the audio driver and test again. If one of these items brings the problem back, start over from step 1 and leave out that item. The only thing I added to my installation was PhysX. I think I'm going to do without Geforce Experience for now. I might try installing the Audio driver, but haven't yet.

For reference, here's the link to my question over on the Geforce forums, with greybear's responses:
 
Last edited:
Oct 20, 2019
9
0
10
I had a similar problem with my rig recently. It would instantly restart in-game, although at least a few things might be different. It was really only happening in 1 game for me as far as I can tell, and it was only in 4k resolution. My voltages, temps, everything was normal at the time of the restarts and it never restarted in 3dMark, only in Far Cry 5.

Since it costs nothing but time, I can suggest to you to try a software solution first before you spend any money to replace the PSU. First, just make sure your voltages (3.3v,5v,12v) are ok as Scout suggested. If they're good, you should uninstall the GPU drivers and Geforce Experience using DDU in Safe Mode while NOT connected to the internet and then install your drivers again from a pre-downloaded installer file for version 431.60 which is an older and presumably more stable driver than the recent ones. To paraphrase the advice given to me over at the nVidia forums (Looks like a lot, but I've just been very specific about each step):

1) Download the driver:
http://us.download.nvidia.com/Windows/431.60/431.60-desktop-win10-64bit-international-whql.exe

2) Download DDU:
https://www.wagnardsoft.com/content/display-driver-uninstaller-ddu-v18019-released

3) Disconnect Network Cable / Disable WIFI adapter

4) Run msconfig by typing it into the Windows start menu. click "Boot" tab, check "Safe Boot" option then "Apply" and reboot your PC.

5) Install and run DDU and select "Options" from the program's menu in the upper left.

6) Check ALL the boxes for nVidia as shown below:
images


NOTE: the warning about C:\NVIDIA sounds worse than it is, but it means that DDU will delete the folder where downloads of previous nVidia driver install files (like the one you manually downloaded in step 1) have accumulated. If you have a slow internet connection and you would like to keep these files just in case, then don't check this box. I just got rid of them all myself.

7) Click "Close" in the options menu, select "nVidia" where is says "Selected Graphics Driver" and then click "Clean and do NOT restart"

8) Open msconfig again and make sure Safe Boot is unchecked (uncheck and Apply if necessary) and then reboot your PC.

9) Now out of Safe Mode and back in normal Windows mode, find the nVidia driver 431.60 file that you downloaded and start the install process. Select CUSTOM and make sure ONLY "nVidia Driver" and "nVidia NGX" are checked (do NOT install PhysX, audio drivers, Geforce Experience, or anything else at this time). Now, check the "Clean Install" box, finish the installation, and then reboot your PC.

10) Reconnect Network Cable / Re-enable WIFI

11) If using any of these, remove them:
Uninstall iCue | Link
Uninstall CAM
If using a PCIe Riser, remove it.

12) Retest with games/software that are giving you restarts.

13) If everything is ok by now and you're not getting restarts anymore, then open the nVidia driver installer again and install PhysX and retest. If still ok, add the audio driver and test again. If one of these items brings the problem back, start over from step 1 and leave out that item. The only thing I added to my installation was PhysX. I think I'm going to do without Geforce Experience for now. I might try installing the Audio driver, but haven't yet.

For reference, here's the link to my question over on the Geforce forums, with greybear's responses:
Hey I tried this and it didn't work thanks anyways, but when I said my pc restarted I forgot to say that when it restarts I don't get the usual restart screen. My screen goes completely black my pc justs shuts down it's like I lost power or something and after 5 seconds it starts back up and it's fine but if I start playing a game it just repeats. I let my computer rest and I play the game I can play it for about 7 hours and it crashes.

I turned off power once and turned it back on and it replicated the exact same thing that happens when I play for a long time, I'm assuming that it's the PSU because my GPU seems fine, and it has to be the GPU or the PSU because it started when I installed these parts.

So here are some questions.
Is there anyway to fix a PSU that loses power when under load?
Why does my PSU crash under load is it because it's too little power or is it because it's faulty?
Can losing power when PC is under load damage the PC in anyway?
Last question which is the best 650 Gold watt PSU that I can get that isn't too expensive that I should get?
Thanks for helping me I really appreciate it have a nice day :D
 
Yeah, sad to say my problem still exists too. I jumped the gun thinking it was fixed. Same problem for me - black screen restarts. I got one just last night. I am also looking at my PSU as the culprit. I will be trying a different one, which I have on hand. If it’s available in your area, Seasonic is a common suggestion here on Tom’s forums - let me check for a model number for you. Corsair is also recommended, but they have some budget models to avoid.

No easy way to fix a psu that’s causing a restart if the fault is within. You’d need knowledge of all the internals and how it all works.

It could be causing restarts for either of those reasons - inherently not enough power, or faulty.

Not really sure how long the hardware can tolerate these restarts, but for me it’s happened at least 10 times in the past year, most of them recently, and I am really forced to escalate my actions to determine what is causing this. Personally, I am concerned about an eventual spectacular PSU failure which definitely could harm other components. Not always, but it definitely could.
 
Last edited:
Yeah, I should point out that I have first hand experience with an EVGA 500B that failed on me. I replaced it in 2015 with an HX750i which saw hundreds of hours of gaming over a couple of years. It's not my daily driver anymore, but I still use the PC occasionally and the HX750i PSU even survived a short circuit on the power cables to a USB 3 add-in card. As best I can tell, the wires were resting above my GPU backplate and the heat melted the insulation on the ketchup and mustard cables and positive and negative touched. The sata power connector on the USB card was melted, but the GPU and everything else survived! I found a small pea-sized ball of melted copper wire resting on the back of the GPU where the wires had been. The whole room filled with smoke. Lesson learned. The GPU card has a cutout on the backplate right where the processor is, so there's a lot of heat there. I'll always give proper clearance and tie wires back from hot areas.