rebooting issue Computer Restarting at Random Intervals ONLY when playing any game.

[Delta38ECHO]

I built this PC back in December of 2020. It ran perfectly with no issues that I can recall. Here lately though, whenever a game is run, it opens nicely, runs well with no lagging or tearing, and the clock speeds and alike for the system have all well been within normal operating parameters; suddenly the screens (3 screens connected using display ports) go black and I'm greeted by the boot screen again. (The one you see when you start/restart your computer).  There were a few other issues that arise as the crashes stared up to. The SATA ports on my ROG STRIX BF450-F Motherboard all stopped working except for two. Finally found the culprit -> The latest BIOS update auto-enables fast boot and CPU overclocking (auto). With those two disabled, the SATA ports all work again, but haven't fixed the rebooting issue... I've done the driver updates for CPU, GPU, BIOS, Chipsets, USB hubs, Bluetooth, and Reinstalled Windows.  I even went as far as to revert back to a 2020 version of bios and all respective drivers for the rig, but nothing worked. (All drivers and BIOS up to date now.)

I've cleared all errors in the windows "Event Viewer" and there are no pending critical errors being posted other than EVENT 86 and EVENT 46.

• Event 86:

  • SCEP Certificate enrollment initialization for WORKGROUP\G-RIG001$ via https://AMD-KeyId-578c545f796951421221a4a578acdb5f682f89c8.microsoftaik.azure.net/templates/Aik/scep failed:

    GetCACaps...  (Windows states that it's an AMD issue)

• Event 46:

  • Computer yelling at me to stop having the power turn off unexpectedly

After pulling my hair out and trying to diagnose this issue myself, I've run a series of tests using Cinebench, Heaven, and MemTest5.
So far my conclusions for:

1.)  Cinebench - (Monitored with AMD Ryzen Master) Ran with multi-core option ~5 times for 10 mins each time. No crash. No wavy clock speeds, and now dips /raises in CPU temp.  - I believe the CPU is fine in this instance.

 

2.)  MemTest5 - Ran tests for all aspects of the 24gb of RAM. No fails detected by program.  - I believe the RAM is fine.

 

3.)  Heaven Benchmark 4.0 (Basic Edition) - (Monitored with AMD Software: Adrenaline Edition) Video starts off choppy and slow as each new transition loads in.  Noticed the GPU Utilization was ranging from 0% to 80% to 100%, then back down to 0% quite wildly as the sample intervals were set to show every 0.5 seconds.

 

3.a.)  Stopped the Heaven from playing and started it up again after it running for about 7 minuets of the first run. Video loaded smoothly and everything seemed fast and quick. GPU Utilization ranged from 100% to 80% (mainly 100%).  This lasted for about 18 minuets and then the system crash -> automatic reboot happened again.

 

3.b.)  Started Heaven again, but had the GPU overclocked through AMD Adrenaline.  Same result as "3.a" only there hasn't been a crash in about 2 hours now.  (Note - No Overclocking is ever used when playing games or anything on this PC.)

 

3.c.)  Used default settings for GPU again through AMD Adrenaline -> Ran Heaven once more and the system restart/crash happened again while GPU usage was at ~100% and all system heat levels were well within normal ranges. 

 

Note - Heaven Settings are:

• API: DirectX 11
• Quality: High
• Tessellation: Extreme
• Stero3D: Disabled
• Multi-Monitor: Disabled
• Anti-Aliasing: x8
• Fullscreen: No
• Resolution: 1920x1080

 

As I said before, everything about this Rig had worked well for about a year and a half. Not sure why it would even be having this sort of issue now.  I'm attaching some  images of the computer build here as well as the PSU since this problem MIGHT be related to another post here: 

https://linustechtips.com/topic/1214710-help-pc-crashes-when-playing-games-and-on-heaven-benchmark/?tab=comments

Specs:
MotherBoard: ASUS ROG STRIX B450-F Gaming

System Type: x64

Processor: AMD RYZEN 5 1600 Multicore

O.S.: Windows 10 Edu Ed.

GPU: XFX Radeon RX 570 RS XXX Edition 1286MHz, 8gb GDDR5, DX12 VR Ready... AMD Graphics Card (RX-570P8DFD6)

BIOS: American Mega Trends

RAM: 24 GB - (XPG z1 DDR4 3200MHz (PC4 25600) 288-Pin CL16-20-20 Memory Modules)

PSU: Cooler Master Elite v3 600 Watt ATX Power Supply (See picture for specs -> detailed).

(If I missed any spec detail, I hope it'd be in the images provided.)

Sorry for the long post, but I'm almost at a loss as to what to do/try. Don't know if the PSU became "not good enough" suddenly or what, but... any and all help is appreciated. Apologies in advance for any rambling.

 

 

 

[kx9987]

It could be psu or gpu, try to monitor power consumption with overlay from amd and if it spikes and thes shuts off its probably psu, also try amd pro driver - fixed my issue with lagging and sudden shut downs

[Delta38ECHO]

2 minutes ago, kx9987 said:

It could be psu or gpu, try to monitor power consumption with overlay from amd and if it spikes and then shuts off its probably psu, also try amd pro driver - fixed my issue with lagging and sudden shut downs

Understood, Will get AMD Pro Ed, select my hardware and have it reinstall the drivers. Then, will run overlay with Heaven and see if the power consumption spikes.  Thanks for the suggestion. Implementing now.

[Crunchy Dragon]

16 minutes ago, Delta38ECHO said:

Understood, Will get AMD Pro Ed, select my hardware and have it reinstall the drivers. Then, will run overlay with Heaven and see if the power consumption spikes.  Thanks for the suggestion. Implementing now.

Just to be on the safe side, I would recommend downloading and running Display Driver Uninstaller to clean out the old drivers before you reinstall new drivers.

[kx9987]

When installing be sure to check factory reset and if not uninstalled with 3rd party uninstaller then make sure you choose same destination and it says downgrade or upgrade

 

[Delta38ECHO]

GPU So Far:

 

.) Used driver Uninstaller to get rid of previous display drivers in system.

 

.) Used AMD Pro to get all new drivers for GPU, chipsets, etc.

 

Started up and ran Heaven again with GPU not Overclocked. Video started off laggy like in instance 3 from my original post. Power consumption was very low as well as GPU utilization.  As Heaven proceeded to run longer, GPU usage shot up to 100%, clock speed maintained around 1284 - 1287MHz, VRAM Clock speed: 1750MHz, and GPU power consumption went from 22W to 103W  - 110W.

 

Currently, the Heaven video has "hung up' on a transition screne/frame and hasn't displayed any motion in a while. (Frozen), while GPU continues to operate at mentioned clock speeds. (No crash yet...but... ya know...)

[Delta38ECHO]

So after some time, the video for Heaven ran smooth, like you'd expect it to. I left it to run while getting some water and upon my return "Tada" login screen!   ...meaning it crashed and rebooted again.

 

I didn't see any fluctuations with power and clock speeds once the graphics card got going, but... Idk..

 

Do you all think it's the GPU that's having issues since the problem is only related to tasks intensive to said GPU, or is it a BIOS issue that maybe needs more power to the GPU, or less? 

[kx9987]

Do you have a spare gpu or kntegrated gpu, could you swap it an then test?

10 minutes ago, Delta38ECHO said:

So after some time, the video for Heaven ran smooth, like you'd expect it to. I left it to run while getting some water and upon my return "Tada" login screen!   ...meaning it crashed and rebooted again.

 

I didn't see any fluctuations with power and clock speeds once the graphics card got going, but... Idk..

 

Do you all think it's the GPU that's having issues since the problem is only related to tasks intensive to said GPU, or is it a BIOS issue that maybe needs more power to the GPU, or less? 

 

[kx9987]

And try reseting bios or updating it

[Delta38ECHO]

56 minutes ago, kx9987 said:

And try reseting bios or updating it

I don't have a spare GPU to use for this thing..yet.  Although the board has an HDMI slot. Do you think trying with that would work? See if it's still crashing there?

 

I did update the bios already. It's the most current bios out there.  Also tried rolling back windows and all bios + drivers to the 2020 configuration (from when I first built the PC.)  Nothing has been giving...

 

 

 

[Delta38ECHO]

Well, today is winding down so here's my current findings:

 

Got rid of anything ASUS related in terms of it trying to overclock anything.

 

Used UDD to rip out any and all display drivers by AMD and then reinstalled them using suggested PRO version of AMD Adrenaline.

 

Ran Heaven and watched the clock speeds, wattage, temp, etc.  Of the GPU, CPU, and Ram.  All operated well within standard levels.   Also note that after using PRO version of Adrenaline the stuttering and "hanging up" I was initially seeing is now gone.

 

After a good 8 minutes, the system crashed and rebooted itself again, so problem still persists.

 

Afterwards, tried the test again and it lasted a good hour before I stopped the test.

 

Final thoughts for the night:

 

It's maybe the ASUS board still trying to do some shady stuff in the background.

 

PSU is maybe not enough, but it's well past the 450W requirement.

 

Research to see if the Processor might have a dead spot for heat that the sensors may not detect...in which case... it'd be a problem no tool can find other than just straight up replacing the processor. 

 

This is karma for something I did when I was too young to remember what I was doing OR something I did in a past life.

 

The voltage going to the GPU is too much / Voltage to the CPU is too much. (Which is unlikely cause it's only a year old...)

 

 

[kx9987]

I just went thru my asus mbo bios and found that when reseted bios sets some overclocking shits like xmp

 

try manually disabling everything that is somehow connected to overclocking

[cns00]

I want to ask you have you been playing multiplayer games that have anti cheat? 

[Delta38ECHO]

32 minutes ago, cns00 said:

I want to ask you have you been playing multiplayer games that have anti cheat? 

Yes, quite a few games like to throw in some Anti-Cheat software. The only two I can think of, that I play, is Bungie's "PunkBuster" and whatever CallOfDuty uses for their free version of the game.

[Delta38ECHO]

57 minutes ago, kx9987 said:

I just went thru my asus mbo bios and found that when reseted bios sets some overclocking shits like xmp

 

try manually disabling everything that is somehow connected to overclocking

You know what?  I woke up in the middle of a "why isn't this working?" sleep.  I originally thought this was SSD issue when it came down to it.  Remember, in my first wall of text, where I said that all but two (left and right ports from bottom) were the only ones working? 

 

I, at the time, had 2 SSDs and a DVD/RW hooked up. After the first crash, I checked bios and it only ever saw the first SSD and DVD/RW (2nd SSD didn't exist)  It would crash constantly while playing any game while 2nd SSD was plugged in and not being seen. HOWEVER, with it removed, it ran and played fine. 

 

Also remember how I said messing with the BIOS and turning off the overclocking settings allowed BIOS to then "see" the other SATA ports come back into existence? 

 

I betcha this Motherboard damaged itself with its default settings and if I were to remove these other SSDs and whatnot EXCEPT for what's in the first two slots then it'll work again.

 

I will update once this hail-mary shows results... and if it does, man...  This better be under warranty.

[Penziplays]

A completely new Windows install helped me with this. It was the only thing I came up with after testing the ssd with the "corrupted" windows install on another pc and still having those crashes.

[cns00]

2 hours ago, Delta38ECHO said:

Yes, quite a few games like to throw in some Anti-Cheat software. The only two I can think of, that I play, is Bungie's "PunkBuster" and whatever CallOfDuty uses for their free version of the game.

That's why your pc is restarting. The anti cheat is messing up and it's causing the pc to restart.

 

Uninstall all the multiplayer games that mess things up, restart your pc, only play offline games that have no anti cheat and see what happens.

[kx9987]

9 hours ago, Delta38ECHO said:

You know what?  I woke up in the middle of a "why isn't this working?" sleep.  I originally thought this was SSD issue when it came down to it.  Remember, in my first wall of text, where I said that all but two (left and right ports from bottom) were the only ones working? 

 

I, at the time, had 2 SSDs and a DVD/RW hooked up. After the first crash, I checked bios and it only ever saw the first SSD and DVD/RW (2nd SSD didn't exist)  It would crash constantly while playing any game while 2nd SSD was plugged in and not being seen. HOWEVER, with it removed, it ran and played fine. 

 

Also remember how I said messing with the BIOS and turning off the overclocking settings allowed BIOS to then "see" the other SATA ports come back into existence? 

 

I betcha this Motherboard damaged itself with its default settings and if I were to remove these other SSDs and whatnot EXCEPT for what's in the first two slots then it'll work again.

 

I will update once this hail-mary shows results... and if it does, man...  This better be under warranty.

Try disconnecting the optical drive and connecting ssd in that same port. What ssd model is that? AAnd how old is it?

[Delta38ECHO]

31 minutes ago, kx9987 said:

Try disconnecting the optical drive and connecting ssd in that same port. What ssd model is that? AAnd how old is it?

That was the idea, Assuming the only first two sata ports were working as intended, the system would crash if any other devices were in any other range of SATA ports.  I then tested the voltage of the other SATA ports with a multimeter; at random times during testing. The board would drop voltage to random SATA ports frequently. Tested the PSU and the connecting wires gave continuous power. PSU has been deemed "working as intended."

 

The drives are:

 

SAMSUNG 870 EVO SATA III (Bought in 2023)

 

Samsung 860 EVO SATA III (Bought in 2020)  (all 1 TB, respectively).

 

---

 

This morning I disconnected all but two SATA ports on the bottom (main SSD with OS on it and extra storage SSD).  I then ran Heaven for a few hours with no overclocking the GPU and everything ran fine. 

 

Note: BIOS had all its self-harming auto-overclocking settings switched to manual,  fast boot disabled, and anything system wise related to overclocking were never on.

 

Conclusion: No crashes when only using the first two SATA ports. ASUS's default settings harmed the MOBO; a replacement is required in order to use the full functionality of this board.

 

Rant: Board was bought in 2020. Checked with asus customer support about warrenty.  My warrenty expired 4 days ago on July 8th 2023. No exceptions. No excuses.  So...  that's fun.

 

--

Marking this post as done and dusted.  Wanted to thank literally all of you so much for your continued help and dedication to my plights and dealing with my whinning.  You're all saints and I'm ever grateful.  I wish y'all the best.

 

[Whatisthis]

2 hours ago, Delta38ECHO said:

That was the idea, Assuming the only first two sata ports were working as intended, the system would crash if any other devices were in any other range of SATA ports.  I then tested the voltage of the other SATA ports with a multimeter; at random times during testing. The board would drop voltage to random SATA ports frequently. Tested the PSU and the connecting wires gave continuous power. PSU has been deemed "working as intended."

 

The drives are:

 

SAMSUNG 870 EVO SATA III (Bought in 2023)

 

Samsung 860 EVO SATA III (Bought in 2020)  (all 1 TB, respectively).

 

---

 

This morning I disconnected all but two SATA ports on the bottom (main SSD with OS on it and extra storage SSD).  I then ran Heaven for a few hours with no overclocking the GPU and everything ran fine. 

 

Note: BIOS had all its self-harming auto-overclocking settings switched to manual,  fast boot disabled, and anything system wise related to overclocking were never on.

 

Conclusion: No crashes when only using the first two SATA ports. ASUS's default settings harmed the MOBO; a replacement is required in order to use the full functionality of this board.

 

Rant: Board was bought in 2020. Checked with asus customer support about warrenty.  My warrenty expired 4 days ago on July 8th 2023. No exceptions. No excuses.  So...  that's fun.

 

--

Marking this post as done and dusted.  Wanted to thank literally all of you so much for your continued help and dedication to my plights and dealing with my whinning.  You're all saints and I'm ever grateful.  I wish y'all the best.

 

Where are you located? In certain states there is a four year implied warranty period for all items above 14.99 in price.

 

Your credit card might offer an extra year of warranty as well. 
 

Do some digging, you might be able to get it replaced still. 

 

https://www.consumerreports.org/cro/magazine/2013/05/the-word-on-warranty-protection/index.htm