• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Question Pc Shutting Down Randomly While Playing Games

andres191

Junior Member
Nov 10, 2020
7
0
6
I apologies for my English. I'm not sure if this is the correct section, sorry if not.

The PC was working fine until a few months ago. I go on to explain the problem.

PC Specifications:

  • CPU: Intel Core I3-9100F
  • Motherboard: Gigabyte H310M-H 2.0
  • Graphics processor: ASUS Radeon RX 560 2Gb
  • Power Source: XFX XT 450w 80+ Bronze
  • RAM: Crucial Ballistix 1x8Gb 2133MHz
  • Storage drive 1: HDD WD 1TB 5400 RPM
  • Storage drive 2: KINGSTON 120Gb SSD
  • Operating System: Windows 10 Pro x64

All the components except the HDD (5 years old) are only 2 or a year and a half years old.

Issue:

The PC suddenly shuts down at a random moment when I play video games with some noticeable load. Sometimes it lasts for a long time, sometimes it shuts off in the first few minutes. There is no freeze, there is no slowness, there is nothing, it turns off as if the power went out. When the PC shuts down, sometimes it takes about 5 minutes or 3 and it turns back on by itself, it is not a reboot. When the PC shuts down, the bios always resets.
The Date and Time always change when this happen.

Tested video games where it has been shutdown: Diablo 3, World of Warcraft, Dota 2, Little Nightmares, League of Legends (The first time in this game never shutdown, but after a few months start to happen).


Some on the HDD and some on the SSD, the same thing happened.

What I have tried so far:

  • Temperatures at maximum load, the CPU reaches 60 ° and the GPU 70 °. I don't think they are high temperatures, I have not controlled the voltages because I do not understand that part very well, nor the speeds of the clocks. The HDD and SSD do not exceed 30 °, the mother neither.
  • Complete partition deletion and complete Windows reinstallation.
  • Bios update.
  • RAM testing with Windows software.
  • Force the CPU and GPU separately with software, they passed them all without shutting down.
  • I changed the power source for a borrowed one to test (Generic 500Watts), it was still the same.
  • Change the GPU for an older one and the same thing happened.
  • I changed the thermal paste of the CPU.
  • I removed all the cables and reattached them, as well as the components, except the cpu.
  • Changed the CMOS battery of the bios (It worked for 2 weeks and returned to the same).
  • I don't think it's a virus being that I deleted the partitions, but I still did an analysis with MalwareBytes AntiMalware, Adwcleaner, CCleaner, etc.
  • Windows 10 is up to date and the GPU drivers too, I have tried using the DDU and reinstalling 2 versions, the most recent and an old one, with the old one shutting down even faster.
  • I used MSI Afterburner and HwInfo to monitor temperatures. For the CPU and GPU forces I used Aida and Furmark. I used another one for the GPU just in case, I don't remember the name, if you need it I'll look it up.


Things I have not tested yet:

  • Remove the HDD and SSD and try another one.
  • Replace the Motherboard (Not enough budget for replace now).
  • Replace the CPU (Not enough budget for replace now).
  • Change the power outlet (Although it sounds illogical, I cannot move the PC because I am having certain problems at home and it is somewhat complicated right now, but I will do it soon).

Thank u so much.
 

Campy

Senior member
Jun 25, 2010
785
171
116
My first thought was PSU, but then I read you've already tried a second unit so it seems likely to be a motherboard issue.
 

shredz

Member
Aug 5, 2010
141
1
81
Same thing happened on one of my pcs. I was convinced it was a graphics card error, but ended up being the PSU. It was a good one, seasonic so I never questioned it. I sent it in and seasonic replaced it. In your case it seems it may be a motherboard issue, but if you have another PSU to try I would try that.
 

andres191

Junior Member
Nov 10, 2020
7
0
6
Same thing happened on one of my pcs. I was convinced it was a graphics card error, but ended up being the PSU. It was a good one, seasonic so I never questioned it. I sent it in and seasonic replaced it. In your case it seems it may be a motherboard issue, but if you have another PSU to try I would try that.
Thanks for Reply, both.
I tried another PSU, i know it was a generic one, but it doesn't make any sense because in the case that the generic one wasn't enough power, i change the video card for and old one that consume nothing. Probably it's the motherboard, seems looks like it's draining all my CMOS battery, or maybe the outlet, not sure tbh.
 

Leeea

Golden Member
Apr 3, 2020
1,060
1,248
96
Brutal.

Low probability but free to try suggestion: memtest86


When the PC shuts down, the bios always resets.
That is a gut punch. Odds are Mainboard or PSU.

I suggest taking a flashlight and a magnifying glass to the mainboard and see if you cannot see something obvious. Specifically with the VRMs, but do the whole thing. You might not find anything, and a null result has limited value. However if you do find something, you will know where your problem is. Sometimes board problems can easily be fixed even by non-techs.
 
  • Like
Reactions: andres191

andres191

Junior Member
Nov 10, 2020
7
0
6
Brutal.

Low probability but free to try suggestion: memtest86




That is a gut punch. Odds are Mainboard or PSU.

I suggest taking a flashlight and a magnifying glass to the mainboard and see if you cannot see something obvious. Specifically with the VRMs, but do the whole thing. You might not find anything, and a null result has limited value. However if you do find something, you will know where your problem is. Sometimes board problems can easily be fixed even by non-techs.
I will try memtest86, thanks for reply.
 

andres191

Junior Member
Nov 10, 2020
7
0
6
Brutal.

Low probability but free to try suggestion: memtest86




That is a gut punch. Odds are Mainboard or PSU.

I suggest taking a flashlight and a magnifying glass to the mainboard and see if you cannot see something obvious. Specifically with the VRMs, but do the whole thing. You might not find anything, and a null result has limited value. However if you do find something, you will know where your problem is. Sometimes board problems can easily be fixed even by non-techs.
I tried memtest and no errors.
About motherboard, i can't see nothing weird, i will attach some pics
 

Attachments

Leeea

Golden Member
Apr 3, 2020
1,060
1,248
96
I zoomed in and those pictures and could not see anything, was worth the try.

A positive mem-test rules out your memory stick, so that is good news.

The PC suddenly shuts down at a random moment when I play video games with some noticeable load.
Two more ideas, both free:

idea 1:
When computer fans age sometimes they stop spinning when they heat up. Take a flashlight and watch the fans and make sure they all keep spinning correctly as the system heats up under load. I am worried about the power supply fan specifically.

----------------

idea 2:
You are close to the max power spec on your Power supply, lets try downclocking and vsync. Vsync will force the cpu to wait until after each frame is displayed, effectively forcing the CPU to idle and do nothing. With the video card nerfed, this should reduce your systems power consumption.

ASUSs website recommends a 400 watt PSU for the ASUS Radeon RX 560 2Gb. An i3-9100F has a "TDP" of 65 watts, but manufacturer TDP figures have been marketing lies for a long time now. Reality is likely between 80 to 140 watts for your CPU alone.

Go into overclocking section on the video card software and hit it with the nerf bat. Sometimes video cards do not like underclocking and will crash. However, most of the time they nerf just fine. With vsync on this will force the entire system to slow down and use less power.
 
  • Like
Reactions: andres191

andres191

Junior Member
Nov 10, 2020
7
0
6
I zoomed in and those pictures and could not see anything, was worth the try.

A positive mem-test rules out your memory stick, so that is good news.



Two more ideas, both free:

idea 1:
When computer fans age sometimes they stop spinning when they heat up. Take a flashlight and watch the fans and make sure they all keep spinning correctly as the system heats up under load. I am worried about the power supply fan specifically.

----------------

idea 2:
You are close to the max power spec on your Power supply, lets try downclocking and vsync. Vsync will force the cpu to wait until after each frame is displayed, effectively forcing the CPU to idle and do nothing. With the video card nerfed, this should reduce your systems power consumption.

ASUSs website recommends a 400 watt PSU for the ASUS Radeon RX 560 2Gb. An i3-9100F has a "TDP" of 65 watts, but manufacturer TDP figures have been marketing lies for a long time now. Reality is likely between 80 to 140 watts for your CPU alone.

Go into overclocking section on the video card software and hit it with the nerf bat. Sometimes video cards do not like underclocking and will crash. However, most of the time they nerf just fine. With vsync on this will force the entire system to slow down and use less power.
I will try VSYNC, and all the Fans are fine when the shutdown happen, anyways if u want to see more pics or other i publish some on here, just i don't want to forget information that other ppl ask me before: https://www.cnet.com/forums/discussions/pc-shutting-down-randomly-while-playing-games/
 

Leeea

Golden Member
Apr 3, 2020
1,060
1,248
96
When the PC shuts down, sometimes it takes about 5 minutes or 3 and it turns back on by itself, it is not a reboot.
That sounds like a thermal circuit breaker:
The wait time you are describing is the thermal breakers bimetal cooldown time.

Most likely the power supply's thermal circuit breaker. If you stick your ear next to it, you might even hear it click as it resets itself.

-------

The Generic 500Watt unit you tried was likely inferior to the XFX unit you already have.
 
Last edited:

Leeea

Golden Member
Apr 3, 2020
1,060
1,248
96
odd thought:

What is your AC line power like? The power coming to where you computer plugs into the wall?

If the AC power is poor, as the voltage drops the amps will climb, generating far more heat per watt the power supply puts out. This will eventually throw the thermal circuit breaker in the power supply.

Frequently this can be detected with a Kill-a-watt meter (or cheap knockoff*), by watching the voltage display under load. If the voltage drops under the power supplies labeled minimum, you have a problem.
*https://smile.amazon.com/s?k=kill-a-watt&ref=nb_sb_noss_2
*https://www.google.com/search?q=kill-a-watt+meter&tbm=shop
*be sure to buy a unit rated for the voltage your testing. In the US we are most commonly 120v, but in Europe it is frequently 240v.
 
  • Like
Reactions: andres191

andres191

Junior Member
Nov 10, 2020
7
0
6
odd thought:

What is your AC line power like? The power coming to where you computer plugs into the wall?

If the AC power is poor, as the voltage drops the amps will climb, generating far more heat per watt the power supply puts out. This will eventually throw the thermal circuit breaker in the power supply.

Frequently this can be detected with a Kill-a-watt meter (or cheap knockoff*), by watching the voltage display under load. If the voltage drops under the power supplies labeled minimum, you have a problem.
*https://smile.amazon.com/s?k=kill-a-watt&ref=nb_sb_noss_2
*https://www.google.com/search?q=kill-a-watt+meter&tbm=shop
*be sure to buy a unit rated for the voltage your testing. In the US we are most commonly 120v, but in Europe it is frequently 240v.
thabk so much, i will try everything what u said and update in one week probably.
 

aigomorla

Cases and Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
19,208
1,603
126
Im also leaning heavy on the board going bad.

When the PC shuts down, the bios always resets.
The Date and Time always change when this happen.
This could be a corrupted bios chip, or the board is heading downhill.
The CMOS should never reset if the battery is there, and the jumper is not set on short.

It means either the bios chip is dying, or the board is somehow shorting itself during the power down, as again, it would need to be a physical short or physical damage on bios chip to have this occur.
 

Leeea

Golden Member
Apr 3, 2020
1,060
1,248
96
This could be a corrupted bios chip, or the board is heading downhill.
Seems unlikely, as BIOS is stored in flash memory, which does not require any power.

The CMOS should never reset if the battery is there, and the jumper is not set on short.
notice what he said: "Changed the CMOS battery of the bios (It worked for 2 weeks and returned to the same)."
I thought that was very curious. But it is not unheard of for PSUs to cut out and reset the CMOS memory*.

It means either the bios chip is dying, or the board is somehow shorting itself during the power down, as again, it would need to be a physical short or physical damage on bios chip** to have this occur.
I disagree. While it is possible his board is failing, it is not the bios chip.

*CMOS memory must be powered at all times, even a momentary interruption for a millisecond will reset it.

**the bios chip is a flash memory chip, it needs no power, and is separate from the CMOS memory chip. When BIOS memory chip fails normally*** it corrupts, and it is all over until it is replaced.

***if you flash it with a bad bios, yes, that corrupts it to, but that is not a normal failure and can be reflashed to a good bios

The CMOS memory chip is connected to the CMOS battery. Since the CMOS battery is not rechargeable, the mainboard must not send power to the CMOS battery (doing so could cause a fire). However, nearly all of the time the CMOS memory is powered from the mainboards power, and not the CMOS battery. Even when disconnected from the wall, there will be a reserve power on to mainboard that will power the CMOS memory for some time***. It is not until that reserve power is mostly drained that the mainboard switches to pulling power from the CMOS battery. This swap is padded with a capacitor, which maintains the CMOS memory for the split moment in time the CMOS memory is powered by nether the mainboard or the CMOS battery.

The way his system is crashing, with the thermal breaker being tripped under load gives no warning to the mainboard that it needs to swap the CMOS memory from its own power to the CMOS battery. So it just crashes hard, resetting the CMOS when the power is cut to the CMOS memory. But then the switch (or solid state relay) is no longer powered, and swaps the CMOS battery into the dead circuit. The CMOS battery is then forced to powerup the pad capacitor, the CMOS memory, and the real time clock, rather then just maintaining a pre-existing charge. This greatly reduces the life of said battery.


*** experiment: pull your CMOS battery leave your PSU on, and unplug it from the wall. It will take several seconds to minutes depending on board/PSU before the CMOS will reset. Most boards will also boot without a CMOS battery. You will notice this has no effect on your BIOS version number, it just resets your CMOS settings and date/time.
 

andres191

Junior Member
Nov 10, 2020
7
0
6
Guys i will send the Motherboard to RMA today, they say 30 days idk, not sure, i live far from RMA place so i need to send it. I will update when back and be sure if it's the Motherboard and if it's can be repaired. Thank u all.
 

ASK THE COMMUNITY