System restart after freeze in games

[Kerbalino]

Hello everyone!

 

My system specs:

• CPU: Ryzen 5 3600
• GPU: Asus ROG STRIX RX5600xt oc
• Motherboard: MSI B450 Tomahawk Max
• RAM: HyperX 2x8GB DDR4-2666 CL16
• PSU: beQuiet! Pure Power 11 500W
• M.2 SSD: Samsung 970 Evo Plus
• HDD: Seagate BarraCuda 2TB
• Cooler: beQuiet! Pure Rock 2
• OS: Windows 10 Pro 64-bit

 

Edit: I have no overclocks or XMP profiles, everything is stock

 

I am experiencing a rather annoying problem. My system, which is not even a week old, restarts suddenly without any error message.

 

To be more exact, I have been playing Titanfall 2 just fine for a couple of rounds, before the game crashes with no error window or anything else. After I've started it again and went back in-game, my system froze up for a couple of seconds before restarting completely. Again, no errors or messages. This has also happened multiple times. The second time, the same thing happened, only with a different game in the loading screen (Subnautica).

 

This is what I have tried so far:

• Reseat RAM, GPU, power plugs
• Run an OCCT CPU and GPU test for 10 minutes, results being
  - CPU max temps: 79°C
  - GPU max temps: 74°C
  Both tests didn't lead to the restart; also the OCCT power test, but I stopped that one after 30 seconds because my CPU went above 90°C ...
• Windows memory test, no errors there
• Updating GPU drivers

 

It's hard for me to recreate the problem, since it seems that it only occurs after playing for quite a while. Temps should be fine after playing games, but I will monitor them the next time when I play. Any help is apprechiated!

 

Marko

 

[Nerdom]

I'm guessing you're overdrawing your PSU and triggering over-current protection.

 

500W is a bit low for a 3600 and 5700XT in my opinion

[SavageNeo]

3 minutes ago, Nerdom said:

I'm guessing you're overdrawing your PSU and triggering over-current protection.

 

500W is a bit low for a 3600 and 5700XT in my opinion

my 3700x and 5700xt bot overcloked use little over 400w. honestly 500w is enough and that psu is decent.

 

are you running with XMP on?

[Kerbalino]

4 minutes ago, Nerdom said:

500W is a bit low for a 3600 and 5700XT in my opinion

@Nerdom But my GPU is a 5600xt, not a 5700xt, or do they draw the same power?

[Kerbalino]

1 minute ago, SavageNeo said:

are you running with XMP on?

@SavageNeo Nope, I will mention that in my post. I have no overclocks or XMP profiles on.

[SavageNeo]

2 minutes ago, Kerbalino said:

@SavageNeo Nope, I will mention that in my post. I have no overclocks or XMP profiles on.

try with XMP on.

4 minutes ago, Kerbalino said:

But my GPU is a 5600xt,

it draws around 50w less than 5700xt so you consume under 350w, since you have no oc. your psu is fine.

[Pitbull Tyson]

12 minutes ago, Kerbalino said:

Hello everyone!

 

My system specs:

• CPU: Ryzen 5 3600
• GPU: Asus ROG STRIX RX5600xt oc
• Motherboard: MSI B450 Tomahawk Max
• RAM: HyperX 2x8GB DDR4-2666 CL16
• PSU: beQuiet! Pure Power 11 500W
• M.2 SSD: Samsung 970 Evo Plus
• HDD: Seagate BarraCuda 2TB
• Cooler: beQuiet! Pure Rock 2
• OS: Windows 10 Pro 64-bit

 

Edit: I have no overclocks or XMP profiles, everything is stock

 

I am experiencing a rather annoying problem. My system, which is not even a week old, restarts suddenly without any error message.

 

To be more exact, I have been playing Titanfall 2 just fine for a couple of rounds, before the game crashes with no error window or anything else. After I've started it again and went back in-game, my system froze up for a couple of seconds before restarting completely. Again, no errors or messages. This has also happened multiple times. The second time, the same thing happened, only with a different game in the loading screen (Subnautica).

 

This is what I have tried so far:

• Reseat RAM, GPU, power plugs
• Run an OCCT CPU and GPU test for 10 minutes, results being
  - CPU max temps: 79°C
  - GPU max temps: 74°C
  Both tests didn't lead to the restart; also the OCCT power test, but I stopped that one after 30 seconds because my CPU went above 90°C ...
• Windows memory test, no errors there
• Updating GPU drivers

 

It's hard for me to recreate the problem, since it seems that it only occurs after playing for quite a while. Temps should be fine after playing games, but I will monitor them the next time when I play. Any help is apprechiated!

 

Marko

 

I would blame the PSU off the bat. A restart without a BSOD or a no freeze or black screen but just a restart means PSU not driving juice and not enough on the ampage on the 12v railing and what not.✝

[Helpful Tech Witch]

Did you clear the cmos? 

[Kerbalino]

3 minutes ago, HelpfulTechWizard said:

Did you clear the cmos? 

@HelpfulTechWizard No, but I will do it now.

 

9 minutes ago, Pitbull Tyson said:

I would blame the PSU off the bat. A restart without a BSOD or a no freeze or black screen but just a restart means PSU not driving juice and not enough on the ampage on the 12v railing and what not.✝

@Pitbull Tyson Shouldnt it be enough wattage to run the whole system even under load?

[Pitbull Tyson]

1 minute ago, Kerbalino said:

@HelpfulTechWizard No, but I will do it now.

 

@Pitbull Tyson Shouldnt it be enough wattage to run the whole system even under load?

Yes sir you need about 400w max for powering up and gaming and full CPU load.  The 500w should be enough but here is why it isn't.  Since it is only 500w PSU and a low tier model that means the capacitors and the entire design is compromised.  You can get a good 500w or a cheap 750w.  Hope that makes sense.  We want a 550w PSU with a driven track record. I always recommend EVGA 550w Bronze 80+ Rating or a SeaSonic for those on budget or want to save money and still get great product.TLDR

Actually now that I look at it your Be Quiet is a very good brand PSU and quality.  Still only 500w and your getting restarts with nothing else no black screen no rolling audio while in black screen no freezes no BSODs.  Only thing I would try at this point to alleviate the PSU is to try one stick of RAM and test out system.  If it works then try the other stick of RAM alone and see what happens. After this things will all make sense and we will know for sure if it is the PSU or not. Also use HWInfo64 or Aida64 or through BIOS and see what your health for PSU is.  See what voltages it shows if your doing it from BIOS.  The 12v rail should get 12v power and 5 get 5v of power and 3.3 get 3.3v of power, just something to keep a eye out for as we try to fix your issue.✝

[cr8tor]

2 hours ago, Pitbull Tyson said:

Yes sir you need about 400w max for powering up and gaming and full CPU load.  The 500w should be enough but here is why it isn't.  Since it is only 500w PSU and a low tier model that means the capacitors and the entire design is compromised.  You can get a good 500w or a cheap 750w.  Hope that makes sense.  We want a 550w PSU with a driven track record. I always recommend EVGA 550w Bronze 80+ Rating or a SeaSonic for those on budget or want to save money and still get great product.TLDR

Actually now that I look at it your Be Quiet is a very good brand PSU and quality.  Still only 500w and your getting restarts with nothing else no black screen no rolling audio while in black screen no freezes no BSODs.  Only thing I would try at this point to alleviate the PSU is to try one stick of RAM and test out system.  If it works then try the other stick of RAM alone and see what happens. After this things will all make sense and we will know for sure if it is the PSU or not. Also use HWInfo64 or Aida64 or through BIOS and see what your health for PSU is.  See what voltages it shows if your doing it from BIOS.  The 12v rail should get 12v power and 5 get 5v of power and 3.3 get 3.3v of power, just something to keep a eye out for as we try to fix your issue.✝

Agree with this 100%

Thanks for saving my some typing, and well said to be honest.

 

3 hours ago, Kerbalino said:

monitor them the next time when I play.

Monitor them with hdwinfo so when it reboots you still have the log file to share with us.

 

Download HWiNFO here - https://www.hwinfo.com/

 

When you open HWiNFO, choose the option for Sensors only, and if prompted, hit continue. 

Also if prompted, do not download the beta version.

 

How to create a log:

Open HWiNFO and choose sensors only.

Then in the bottom right there is an icon that looks like a spreadsheet with a green plus on it.

Click that button to start logging, the green plus will switch to a red X to let you know its saving data. 

It will also ask you for a file name and where to save it. I save them to my desktop, name it whatever you want.

Let it run for a few moments or until the machine locks up, resets, or does whatever its not supposed to do,

Then click the spreadsheet again so the red X turns back to a green +.

You now have a log file.

 

How to view the log file:

You can view the log using the GenericLogViewer addon here https://www.hwinfo.com/add-ons/

Download, extract, run GenericLogViewer.exe, click the open file button in the top right and point it to the log file you made above.

Use the dropdown in the upper right to choose which category to view

[Kerbalino]

1 minute ago, cr8tor said:

Monitor them with hdwinfo so when it reboots you still have the log file to share with us.

 

Download HWiNFO here - https://www.hwinfo.com/

 

When you open HWiNFO, choose the option for Sensors only, and if prompted, hit continue. 

Also if prompted, do not download the beta version.

 

How to create a log:

Open HWiNFO and choose sensors only.

Then in the bottom right there is an icon that looks like a spreadsheet with a green plus on it.

Click that button to start logging, the green plus will switch to a red X to let you know its saving data. 

It will also ask you for a file name and where to save it. I save them to my desktop, name it whatever you want.

Let it run for a few moments or until the machine locks up, resets, or does whatever its not supposed to do,

Then click the spreadsheet again so the red X turns back to a green +.

You now have a log file.

 

How to view the log file:

You can view the log using the GenericLogViewer addon here https://www.hwinfo.com/add-ons/

Download, extract, run GenericLogViewer.exe, click the open file button in the top right and point it to the log file you made above.

Use the dropdown in the upper right to choose which category to view

@cr8tor Thanks a lot! I will make sure to have it running in the background while playing next time

[cr8tor]

Just now, Kerbalino said:

Thanks a lot! I will make sure to have it running in the background while playing next time

No problem, i'm interested to see what we find.

[Kerbalino]

So it has been a while now since the last restart, but it happened again, despite the CMOS reset and newest driver version. Although this time, I happened to get some more infos about what might be wrong:

• While the system froze up before restarting, the red debug LED for VGA lighted up on my motherboard
• Roughly 13 minutes before the system froze, all windows seemed to freeze for some milliseconds, before running fine again (thats when I got concerned and fired up HWiNFO logging right away, just in case the system restarts altogether again, which it did)
• The system was not loaded at all, I was just watching a Discord stream when it happened

 

@cr8torThank you again for the HWiNFO idea. This time I had the logging running, but at first glampse (GPU power, currents, etc.), I couldn't find anything out of the ordinary ...

 

I am really starting to become a little frustrated, because it just happens randomly and without any prior warning (unless that window reseting thing means its going to happen)

 

Edit: Found something weird while looking at the used GPU memory in Generic Log Viewer. Right before it restarted, the used memory was around 6670MB, which is weird, because I wasn't running any games or other programs. After the restart, with the same situation and programs running, it only used 800MB, which is much more reasonable. Could the issue be with that?

[Mark Kaine]

23 minutes ago, Kerbalino said:

So it has been a while now since the last restart, but it happened again, despite the CMOS reset and newest driver version. Although this time, I happened to get some more infos about what might be wrong:

• While the system froze up before restarting, the red debug LED for VGA lighted up on my motherboard
• Roughly 13 minutes before the system froze, all windows seemed to freeze for some milliseconds, before running fine again (thats when I got concerned and fired up HWiNFO logging right away, just in case the system restarts altogether again, which it did)
• The system was not loaded at all, I was just watching a Discord stream when it happened

 

@cr8torThank you again for the HWiNFO idea. This time I had the logging running, but at first glampse (GPU power, currents, etc.), I couldn't find anything out of the ordinary ...

 

I am really starting to become a little frustrated, because it just happens randomly and without any prior warning (unless that window reseting thing means its going to happen)

It might be psu. 

Way to find out: get better psu and test. 

 

It's probably the motherboard however. 

Way to find out : same as above, get better motherboard, test it. 

 

 

If you're really unlucky it's both motherboard and psu... 

 

Note: I had this motherboard, it's trash, kept restarting for no reason, XMP on or off. 

 

Put in my old motherboard (b350) and no random issues like that anymore. 

 

ALLEDGELY a motherboard bios update would fix this tho, did you do a motherboard bios update yet? 

[Kerbalino]

2 minutes ago, Mark Kaine said:

ALLEDGELY a motherboard bios update would fix this tho, did you do a motherboard bios update yet? 

@Mark KaineNo, but I will do it right now.

[Kerbalino]

Update: It seems that the GPU's drivers are just broken. I've tried DDU and the newest two drivers, but it doesn't seem to fix the issues I am experiencing. This is very likely on AMD's side, isn't it?

Spoiler

 

[cr8tor]

On 11/3/2020 at 7:55 AM, Kerbalino said:

Update: It seems that the GPU's drivers are just broken. I've tried DDU and the newest two drivers, but it doesn't seem to fix the issues I am experiencing. This is very likely on AMD's side, isn't it?

 

Knowing that you got that VGA diag light previously during the reboot i wonder if its your GPU going bad. That could explain the GPU memory usage issue also along with everything else.

 

You might share your hardware log so we can look at it also.

[Kerbalino]

8 hours ago, cr8tor said:

Knowing that you got that VGA diag light previously during the reboot i wonder if its your GPU going bad. That could explain the GPU memory usage issue also along with everything else.

@cr8torThere has been much more going wrong in the last week. I had multiple games crash and Battlefield was the only one capable of giving me a crash message (seen in the hidden contents in my last reply). I also had a BSOD and with the help of BlueScreenView, it pointed to amdkmdag.sys. My theory is that the drivers just seem to be broken. I am considering to RMA it.

 

I sadly don't have a recent HWiNFO log to share, only the original one with the high memory usage.