[SOLVED] Newly built 3700x system shuts off while gaming

Aug 3, 2020
7
0
20
1
Hello everyone. I have just built my dream PC in July and am having some issues with it shutting itself off while gaming. There doesn't appear to be any overheating issues, I check Temps with Ryzen Master for the cpu. It seems like if I leave the computer on during the day, when I go to play a game in the evening I can almost expect a shut down. Usually after a couple of hours gaming but sometimes as little as 15 or 20 minutes. My system is as follows:
Asus Rog Strix B550e motherboard
AMD 3700x with stock cooler
2 16gb ddr4 3600 gskill trident z neo
Gigabyte rx580
Evga 600w 80 plus bronze
All inside a well ventilated cooler master td500 mesh case with 3 fans in the front.
I plan on updating to a nvidia 3070 when they come out and will upgrade the power supply at that time as well.
This is my 3rd build and the only one I've ever had a problem like this with.
Just to clarify, it doesn't reboot, or give me a blue screen it straight up shuts off instantly like its a psu failure or cpu overheating issue. I had an identical psu swapped it out, problem persists. When overclock settings are on auto in ai tweaker it seems to shut off much sooner.
I hope that's enough info to start with
 
Aug 3, 2020
7
0
20
1
Happy to update that once I was able to switch the psu to a Corsair RMx 850 everything is running smoothly after considerable testing. Now to start overclocking!
 
Aug 3, 2020
7
0
20
1
Update: ran occt, stress tested system for 15 minutes no problems, temps stabilised around 80c for both cpu and gpu at 100% usage. Tried to run 3D test, instant shutdown. Im thinking I have a faulty gpu.
 
This can be one of many things AmonAten and it may be unstable Bios settings.
As you can get into Windows then go into safe mode and check to see if any drivers are causing the shutdown or if any files have been corrupted.

You can get into safe mode by spamming the F8 key after post has finished. Choose safe mode with networking.
When in windows safe mode, go into your Control Panel>Device Manager and check for yellow triangles indicating a bad driver. If there are any then uninstall them and try a reboot.
You can also bring up an elevated command prompt and key in "sfc /scannow" without the quotation marks to check the integrity of your system32 files for corruption. sfc will attempt to correct the issue.
Try the above first.
 
Last edited:
Reactions: AmonAten
Aug 3, 2020
7
0
20
1
This can be one of many things AmonAten and it may be unstable Bios settings.
As you can get into Windows then go into safe mode and check to see if any drivers are causing the shutdown or if any files have been corrupted.

You can get into safe mode by spamming the F8 key after post has finished. Choose safe mode with networking.
When in windows safe mode, go into your Control Panel>Device Manager and check for yellow triangles indicating a bad driver. If there are any then uninstall them and try a reboot.
You can also bring up an elevated command prompt and key in "sfc /scannow" without the quotation marks to check the integrity of your system32 files for corruption. sfc will attempt to correct the issue.
Try the above first.
ok i did that, scannow found no problems, when in safe mode on device manager my audio drivers say they have issues, but only in safe mode, when i check device manager normally everything is fine? when i tried to do the OCCT #D test again this morning it got about 20 seconds in before black screen shutdown instead of instant.
 
Aug 3, 2020
7
0
20
1
UPDATE: I went into device manager, right clicked the gpu and searched for updated driver. Surprisingly it found that i needed an update (i had manually installed what i thought was the latest driver) i installed it and restarted, ran the OCCT 3D test and it ran for 20 mins no issues. I think it was a driver problem. Will update if there are any further issues but I think I fixed it. Thank you for the assistance.
 
Aug 3, 2020
7
0
20
1
Update: it shut itself off again. uff. I think this is faulty hardware and i suspect it is either the PSU or GPU but I don't know how to test this.
 
It may well be the PSU which is not well rated in the first place. You can check your rail voltages (must be under load) in HWInfo64 or AIDA64 . Acceptable voltages are within + or - 5%. on all rails.

The best way is to swap the unit out with a known gold certified working unit of the same or higher wattage.
To test for temperature issues use AIDA64 for 10-15 mins. It's free and will tell you what you need to know for the CPU,GPU and RAM. I can guide you if your not sure.

Also: You are using 32GB "OC"RAM at 3600MHz which needs Bios intervention to get them to work at their rated SPD.
Just have one module inserted in the MB (as per your manual) initially to test, then swap them around. This will also determine if you have a MB ram slot issue.
You should also test your DIMM modules using Memtest86 to check for errors.
 
Last edited:
Reactions: AmonAten
Aug 3, 2020
7
0
20
1
It may well be the PSU which is not well rated in the first place. You can check your rail voltages (must be under load) in HWInfo64 or AIDA64 . Acceptable voltages are within + or - 5%. on all rails.

The best way is to swap the unit out with a known gold certified working unit of the same or higher wattage.
To test for temperature issues use AIDA64 for 10-15 mins. It's free and will tell you what you need to know for the CPU,GPU and RAM. I can guide you if your not sure.

Also: You are using 32GB "OC"RAM at 3600MHz which needs Bios intervention to get them to work at their rated SPD.
Just have one module inserted in the MB (as per your manual) initially to test, then swap them around. This will also determine if you have a MB ram slot issue.
You should also test your DIMM modules using Memtest86 to check for errors.
Thank you, A friend with more knowledge than my own came over and helped test the RAM and rest of the components. I'm confident now that this power supply is causing the issue. I will be ordering a Corsair 800w 80+ gold. I had only gone with this PSU because I had a spare and halfway expected it to not work properly, at least I can still run Premiere pro and other cpu intensive programs, it is only when the gpu is very active that the PSU shorts out. when it would turn off I would need to flip the PSU's switch to off a few seconds and then back on again to get the PC to turn back on, was told this is the giveaway that the PSU is being overloaded.
 
Aug 3, 2020
7
0
20
1
Happy to update that once I was able to switch the psu to a Corsair RMx 850 everything is running smoothly after considerable testing. Now to start overclocking!
 

ASK THE COMMUNITY