Random Crashing while Gaming

[ItsHunterM8]

I havent been playing other games recently but for all I know is this crashing only happens in valorant. After playing in a multiplayer lobby (it doesnt consistently crash in singleplayer lobbies) for about 5-10 minutes (it varies), my monitor will just suddenly go black but the fans of the PC and GPU will still spin (RGB is also still on). Until I manually restart or wait for an indefinite amount of time (sometimes immediately after the screen goes black, sometimes after 30 minutes) my PC will restart automatically.

I dont know whats causing this issue. I thought my windows was corrupted as I was having input lag issues after gaming for around 3 hours and so i tried reinstalling windows to no avail and so i bought a new HDD and had someone install windows 10 on it for me. I then also moved Valorant over to different drives to see if that was the problem, ive tinkered with drivers for my GPU, CPU, and motherboard. I've tried switching out my ram as it seemed like that was the issue (and I still think that its probably the ram). I have 2 sticks of 4GB 2400Mhz CL-16 that Ive been using alongside a single 8GB 3200Mhz CL-22 (seems to run fine at CL-16 at around 2800Mhz), for the past 3ish years with almost no problems. 
I managed to fix the problem by switching out the memory and running only the 8GB stick but then after a day the problem came back. Then I ran the 8GB stick and 1 of the 4GB sticks at 2900Mhz at CL-20 for around half a day before the problem came back again. Because of this I assume the issue is memory related and as of current I dont have the budget to go out and buy spare ram sticks (also because in Indonesia no one sells used RAM for anything less than retail new price) to test out if this is really the issue or not. I also tried changing the page file manually as amidst all this testing I managed to get a BSOD with error code fault in page file or something along those lines. This noticeably decreased input lag in valorant for some reason but if anything caused the crashing to happen what seemed to me to be faster than usual tho I didnt measure it.

 

My System:
PSU: EVGA Bronze 450W

GPU: GTX 970 Strix (I tried underclocking the GPU and CPU to see if it was a power related problem but it persisted so I dont think its that either)

CPU: R5 3600

Mobo: MSI B350-M Pro-VDH

RAM: TeamGroup Elite+ (2x4GB 2400Mhz CL-16) + TeamGroup Elite+ Red (1x8GB 3200Mhz CL-22)

I run 1 NVME SSD for games and a new Seagate Barracuda HDD for Windows

Ive been using this system for around 6 years now with no problem as bad as this one, I upgraded the initial 2x4GB with the 1x8GB around 3+ years ago and ive never really had any issues I havent been able to resolve on my own

Any help would be greatly appreciated, im going insane not able to grind ranked 

 

[podkall]

30 minutes ago, ItsHunterM8 said:

This noticeably decreased input lag in valorant for some reason

The pagefile? That's because pagefile is offloading RAM to disk, which is much slower even if it's an SSD,

 

there's 2 instances where PC offloads to pagefie, when the files aren't that important but used often or something like that, and the more obvious reason when you run out of RAM space.

 

If PC runs out of RAM while your pagefile is off it can cause the PC or apps to crash or have BOSD.

 

To check if it's a power related problem, you could try running your games at capped fps or vsync and on lower settings.

 

Does Valorant blackscreen crash PC too?

[leclod]

You should have your OS (Windows) on the SSD. Otherwise you're hurting performance.

When you decide a Ram setting you should first check if it's stable with a program like Testmem5 or better Memtest86

I'm sure you can get it stable

Edited April 30 by leclod

[ItsHunterM8]

7 minutes ago, podkall said:

The pagefile? That's because pagefile is offloading RAM to disk, which is much slower even if it's an SSD,

 

there's 2 instances where PC offloads to pagefie, when the files aren't that important but used often or something like that, and the more obvious reason when you run out of RAM space.

 

If PC runs out of RAM while your pagefile is off it can cause the PC or apps to crash or have BOSD.

 

To check if it's a power related problem, you could try running your games at capped fps or vsync and on lower settings.

 

Does Valorant blackscreen crash PC too?

Well yea the only issue ive found is in valorant though to be fair I cant say much as I havent played any other games in a while since ive been busy with uni.

I always run on lowered settings though i never thought to test out capping FPS, ill have to give that one a try tho does it really have such a significant impact on the power consumption?

[podkall]

1 minute ago, ItsHunterM8 said:

Well yea the only issue ive found is in valorant though to be fair I cant say much as I havent played any other games in a while since ive been busy with uni.

I always run on lowered settings though i never thought to test out capping FPS, ill have to give that one a try tho does it really have such a significant impact on the power consumption?

what's the difference when GPU is tasked with 140max fps, compared to when it's unbound crunching 500+fps?

[ItsHunterM8]

7 minutes ago, leclod said:

You should have your OS (Windows) on the SSD. Otherwise you're hurting performance.

When you decide a Ram setting you should first check if it's stable with a program like Testmem5 or better Memtest86

I'm sure you can get it stable

Yea I know that much but it got unbelievably finicky when i tried installing Windows on to the SSD (the original reason I got it was to serve as the boot drive due to my slow boot times tho that isnt a problem anymore). Valorant lags when I run it off the HDD but i typically run it on the SSD.
Now another problem I faced was that both the USBs I tried to use to install memtest failed to fully format. This was the same issue I faced with trying to use it as a media drive to install windows but there was no problem formatting or getting the media onto the drives but instead it just never worked when trying to boot off the USB.
Im guessing its both USBs that have issues though I also dont have any USBs lying around and id have to order more

[leclod]

2 minutes ago, ItsHunterM8 said:

Now another problem I faced was that both the USBs I tried to use to install memtest failed to fully format.

I've had a similar problem lately. I did use EaseUS Partition Manager to handle partitions while Windows wouldn't.

[ItsHunterM8]

10 minutes ago, podkall said:

what's the difference when GPU is tasked with 140max fps, compared to when it's unbound crunching 500+fps?

So far it hasnt crashed when capping it at 186 from 328 but if this ends up working i cant believe that theres that massive of a difference when it comes to fps. It does make sense though, ive been running this PSU for 6 years now and it barely covers the 970 at mint condition let alone its probably less efficient self now. Ill keep testing and updating if anything does happen though

[podkall]

1 minute ago, ItsHunterM8 said:

So far it hasnt crashed when capping it at 186 from 328 but if this ends up working i cant believe that theres that massive of a difference when it comes to fps. It does make sense though, ive been running this PSU for 6 years now and it barely covers the 970 at mint condition let alone its probably less efficient self now. Ill keep testing and updating if anything does happen though

I doubt power draw is the main issue:

[podkall]

7 minutes ago, ItsHunterM8 said:

So far it hasnt crashed when capping it at 186 from 328 but if this ends up working i cant believe that theres that massive of a difference when it comes to fps. It does make sense though, ive been running this PSU for 6 years now and it barely covers the 970 at mint condition let alone its probably less efficient self now. Ill keep testing and updating if anything does happen though

though assuming your GPU is in good condition and you have a 6 year old PSU, could be PSU related

[ItsHunterM8]

10 minutes ago, podkall said:

I doubt power draw is the main issue:

Yea it might not be it, I just crashed again capping fps at only 180

 

4 minutes ago, podkall said:

though assuming your GPU is in good condition and you have a 6 year old PSU, could be PSU related

I did buy it used but yea this is just another part that i cant really test due to budget and time. I might just end up having to bring it to a store so they can better troubleshoot what might be causing the problem.

Im just really focused on the fact that when i switched out the ram at first it was working completely fine and i had been playing for half a day straight without crashing and then after restarting my pc it seemed like the problem just re-arose.

Im currently running the 2x4gb kit without the other stick in fear that mixing them might cause the problem but that doesnt seem to be the case

[podkall]

1 minute ago, ItsHunterM8 said:

Yea it might not be it, I just crashed again capping fps at only 180

event viewer logs?

[podkall]

1 minute ago, ItsHunterM8 said:

Im currently running the 2x4gb kit without the other stick in fear that mixing them might cause the problem but that doesnt seem to be the case

and have you tried single 1x8GB? though I don't see how that would fix anything, it's actually slower than 2x4

[ItsHunterM8]

1 minute ago, podkall said:

and have you tried single 1x8GB? though I don't see how that would fix anything, it's actually slower than 2x4

yes I tried it before making this post and it was still happening

 

2 minutes ago, podkall said:

event viewer logs?

Ive never used the event viewer logs but do they not reset after restarting your pc? If they do then I wont be able to view them anyway as the screen goes black and theres no way to turn it back on without restarting

[ItsHunterM8]

I find 3 recurring errors in the viewer logs

The same 3 errors repeat every few page scrolls down though thats a CertificateServicesClient-CertEnroll error so not sure if its related

one of the errors mentions License activation which I believe to be caused by my windows not being original 

[podkall]

2 minutes ago, ItsHunterM8 said:

I find 3 recurring errors in the viewer logs

The same 3 errors repeat every few page scrolls down though thats a CertificateServicesClient-CertEnroll error so not sure if its related

and if you click on it there's details below

[ItsHunterM8]

2 minutes ago, podkall said:

and if you click on it there's details below

I think those ones might be unrelated as i read the details and it was just license activation errors but these ones might be a bit more interesting

 

This was right around where I crashed, it comes right after this error

 

There are multiple of these TPM errors and Critical Kernel-Power related crashes

[podkall]

1 minute ago, ItsHunterM8 said:

I think those ones might be unrelated as i read the details and it was just license activation errors but these ones might be a bit more interesting

 

This was right around where I crashed, it comes right after this error

 

There are multiple of these TPM errors and Critical Kernel-Power related crashes

I think ID41 is when the PC shuts down suddenly or you hold down power button to force turn it off.

 

If that's the case, then you should see if any logs before this event have any clues in them or not.

[ItsHunterM8]

5 minutes ago, podkall said:

I think ID41 is when the PC shuts down suddenly or you hold down power button to force turn it off.

 

If that's the case, then you should see if any logs before this event have any clues in them or not.

Nothing anytime near the crash, this is the closest thing which just states the previous system shutdown

 

The force shut down is me holding down the power button after the monitor turns off as nothing i do is reponsive to the system in that state.

Ill be back here tomorrow as its really late already. I appreciate the help so far

[podkall]

26 minutes ago, ItsHunterM8 said:

Nothing anytime near the crash, this is the closest thing which just states the previous system shutdown

 

The force shut down is me holding down the power button after the monitor turns off as nothing i do is reponsive to the system in that state.

Ill be back here tomorrow as its really late already. I appreciate the help so far

yes, also one of troubleshooting steps is thinking about possible changes to system/hardware right before the problem started happening

[Bjoolz]

Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post. Please follow the instructions to the letter as Windows doesn't like you messing with files in this location.

[ItsHunterM8]

11 hours ago, Bjoolz said:

Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post. Please follow the instructions to the letter as Windows doesn't like you messing with files in this location.

well technically these arent "crashes" so none of the minidump files here i think are related at all but there are 3. All of which have times on them that dont correspond to any of the crash times i had. ill try this tho just in case theres anything there

Minidump.zip

[ItsHunterM8]

So a small update
I was testing the game out while running Rivatuner to see my ram, GPU, and overall power consumption and I realized that the amount of ram being used was constantly increasing before the screen just went black. I then installed all the Ram sticks to try and see what would happen this time. I had set the page file to 16 gigs equal to the amount of ram i had and then i booted up the game.

 

Side Note, I tried uncapping FPS and my GPU went up to 140W without having any problems at all before I then capped it back down to 180.

 

This time the ram consumption kept going up and it did pass 8 gigs but didnt crash this time and I thought that maybe it was fixed. I then kept playing for around an hour in various gamemodes apart from competitive and it seemed completely fine. So then i tried queueing into competitive and low and behold after 2 rounds my PC crashed again and I made sure to keep checking the ram after every round. It was on 9500MB right around when I crashed so that seems to also not be the problem.
My biggest confusion is how i managed to play other gamemodes (all of which had equal or more players in the lobby and were all multiplayer) with 0 problems and if anything really smooth gameplay to what I was getting before all this troubleshooting.

Everytime I rejoined the game the black screen kept happening faster and faster each time and I even tried just booting the pc without running any other programs with again 0 success.

[Bjoolz]

7 hours ago, ItsHunterM8 said:

well technically these arent "crashes" so none of the minidump files here i think are related at all but there are 3. All of which have times on them that dont correspond to any of the crash times i had. ill try this tho just in case theres anything there

Minidump.zip 727.58 kB · 0 downloads

The crashes are from Vanguard so it's likely related. One was the Nvidia driver, but with that specific crash error it can be really hard for the OS to know which driver is actually at fault. I'll leave a copy paste below with some explanation if you are interested. We'll deal with that later if needed. Reinstall Vanguard.

 

DPC_Watchdog_Violation summary:

Quote

You can think of DPC as the CPU's time schedule. Drivers and processes "book time" at certain priorities, they are then put in a queue for a CPU thread and wait for execution. If something doesn't follow the rules, you crash. It's really hard for the OS to know which driver/process actually did the bad thing here. When you open these files in the debugger, it shows you the CPU thread queue of where Windows thinks the issue was. Windows guesses wrong very often.

 

Because these are minidumps, Windows has removed all data it doesn't think we need. So it has removed all the other DPC queues. The way we normally have to debug these is to look at all the DPC queues of 3-4 dump files and compare which drivers are loaded. If every crash has the same driver/process in one of the queues, it's probably that one. We can't do that with minidumps most of the time because Windows has discarded that data.

[ItsHunterM8]

On 5/1/2024 at 7:55 PM, Bjoolz said:

The crashes are from Vanguard so it's likely related. One was the Nvidia driver, but with that specific crash error it can be really hard for the OS to know which driver is actually at fault. I'll leave a copy paste below with some explanation if you are interested. We'll deal with that later if needed. Reinstall Vanguard.

 

DPC_Watchdog_Violation summary:

Wait thats actually really interesting. I've already reinstalled valorant and vanguard twice but that was before reinstalling windows and updating my drivers. I'll try it again but I'll only be able to do that by Monday.

I'll update again