BSOD and crashes to dekstop on many video games

[idanlevayev]

Hello everyone,
I have a question, exactly a year ago I bought a new PC, I had no problems with anything, Until now.

Starting playing Rainbow 6 Siege 2 months ago, and I had a lot of BSOD crashes, usually with this error : "PAGE_FAULT_IN_NONPAGED_AREA" and sometimes (like 2 or 3 times) I had this 1 : "KERNEL_MODE_HEAP_CORRUPTION"

I tried to check if the ram is ok using the windows diagnostic tool thing that makes you restart the PC.
It says no errors, so I tried to update the NVIDIA graphics card driver.
Still didn't help, what finally did help is, that I clean installed the drivers, now a new season started which means an update, and after the update the crash came back, but now its not only in this game, I play another RTS game which can run on every system, no intense requirements and it crashes to desktop and I am the only 1 who crashes from my friends, and lethal company (a video game) also BSOD crash with "PAGE_FAULT_IN_NONPAGED_AREA".

It usually crashes randomly, but sometimes on game launch.

I tried formatting the pc, crashed BSOD 3 times in the same game (and only that game) with a different error (using bluescreen viewer) : Caused By Driver: ntoskrnl.exe, and same "PAGE_FAULT_IN_NONPAGED_AREA"
Right after that tried the heaven benchmark thing, did it on extreme and ultra and it ran for like 20 seconds and then crashed to desktop, tried again and again, and crashed faster each time. lowered the quality to low and it ran for a few minutes until I closed it.

*I play lots of intense video games on max settings like helldivers2, cyberpunk 2077, Baldurs gate 3.
Never experienced any kind of crashes while playing these games.


What could it be?

My PC specs:

CPU: Intel Core i7-13700F
CPU cooler: Arctic Freezer i35
Motherboard: PRIME Z790-P WIFI (LGA1700)
Ram: Corsair Vengeance RGB 2x16GB DDR5 5200MHz
SSD: TS1TMTE220S
HDD: WDC WD20EZBX-00AYRA0
GPU: NVIDIA GeForce RTX 3060
PSU: Antec CUPRUM STRIKE CSK 750W 80 Plus Bronze 120mm Fan PSU CSK750H
Chassis: DF700 FLUX mid-tower
OS: Windows 10 pro 64-bit
Monitor:27 (68.58cm) UltraGear® FHD IPS 1ms 240Hz HDR Monitor with G-SYNC® Compatibility /// LG TV 32 inch (not sure of model)

All of the hardware is 1 year old, and recently cleaned all of it from dust (after crashes started)

 

[bug]

This thread may give you some pointers: https://answers.microsoft.com/en-us...e/73055535-78e0-417e-9b64-d9c91bc7997b?page=2

Basically test the RAM (extended test) and SSD. Unfortunately, you cheaped out and didn't get an IGP, so you can't try to take the GPU out of the equation.
Since it seems to crash when it senses 3D load, it could be that the system cannot handle the power spike. You can try monitoring voltages and stuff.

 

[P4-630]

Unfortunately, you cheaped out and didn't get an IGP

Right... For these reasons, I always say , buy a CPU with iGPU....

 

[bug]

Right... For these reasons, I always say , buy a CPU with iGPU....

It's not always a requirement, the OP may have an older GPU around (I know I do), but for the price difference, it's mostly a no-brainer.

 

[idanlevayev]

This thread may give you some pointers: https://answers.microsoft.com/en-us...e/73055535-78e0-417e-9b64-d9c91bc7997b?page=2

Basically test the RAM (extended test) and SSD. Unfortunately, you cheaped out and didn't get an IGP, so you can't try to take the GPU out of the equation.
Since it seems to crash when it senses 3D load, it could be that the system cannot handle the power spike. You can try monitoring voltages and stuff.

Hey, thanks for the answer,
How do I do the extended test on RAM and SSD, I am afraid I never knew of the IGPU thing, and how can I monitor voltages?

These 3 things you mentioned is something I did not try yet.
And I think I might have an older GPU, gtx 1050, if it still works.

Maybe I should turn off the intel turbo boost technology in the bios? if I understand it correctly it uses some kind of OC when needed, I am just not sure if I should, and what would it mean turning it off.

 

[bug]

Hey, thanks for the answer,
How do I do the extended test on RAM and SSD, I am afraid I never knew of the IGPU thing, and how can I monitor voltages?

These 3 things you mentioned is something I did not try yet.
And I think I might have an older GPU, gtx 1050, if it still works.

Maybe I should turn off the intel turbo boost technology in the bios? if I understand it correctly it uses some kind of OC when needed, I am just not sure if I should, and what would it mean turning it off.

The Windows memory test you ran has two modes: quick and extended. If you already did the extended test, you're good.
For the SSD, there's SMART. Install the software from the manufacturer, it will tell you about drive health, number of errors (a small number is nothing to worry about) and general drive health.

You can monitor the CPU and GPU via CPU-Z and GPU-Z, both have the ability to create logs, should you need them.

And if everything checks out, try swapping out the video card. If the 1050 still works, it doesn't necessarily mean the 3060 is faulty, it may be that your PSU doesn't have enough juice to feed it.

Just go through the above steps, let's try to fond out the culprit. Then we can start thinking about a fix.

 

[QuietBob]

CPU cooler: Arctic Freezer i35
PSU: Antec CUPRUM STRIKE CSK 750W 80 Plus Bronze 120mm Fan PSU CSK750H

I'd start by replacing these.
What are your temperatures and CPU clock like when playing those problematic games? The 13700F can dissipate up to 220 W when running heavily multithreaded tasks. The i35, on the other hand, is an average direct-touch cooler, which may not be capable of keeping the CPU temperature adequate. This could also happen in games, especially if you play with an uncapped frame rate.

But the more pressing issue is your power supply. It's a low quality Tier E (avoid) unit, which I wouldn't trust to run even a backup rig. Try getting something from Tier A or B, there are a number of PSUs there which are great value for money.

 

[bug]

I'd start by replacing these.
What are your temperatures and CPU clock like when playing those problematic games? The 13700F can dissipate up to 220 W when running heavily multithreaded tasks. The i35, on the other hand, is an average direct-touch cooler, which may not be capable of keeping the CPU temperature adequate. This could also happen in games, especially if you play with an uncapped frame rate.

That's BS. I run an i34, it's perfectly adequate for a mid-range CPU. It's perfectly adequate for a high-end CPU, too, it will just not boost as high.
Here's a review that says it can handle 220W (though not much beyond that): https://play3r.net/reviews/cooling/arctic-freezer-i35-a-rgb-review/

But the more pressing issue is your power supply. It's a low quality Tier E (avoid) unit, which I wouldn't trust to run even a backup rig. Try getting something from Tier A or B, there are a number of PSUs there which are great value for money.

Which is why I have already told the OP to monitor voltages and stuff. 750W is enough for the rig the OP runs, but the PSUs can always go bad... Still, it's unlikely, because when PSUs go bad, you're dealing mostly with reboots or shutdowns.

 

[RJARRRPCGP]

Looks like unstable RAM! Looks like RAM corruption.

 

[idanlevayev]

Hello guys, thank you all for the help, I have a question.
I was starting to do the steps you sent up there, and I opened the case, once I opened the case, a cable from the PSU just sticked out, the case probably like lightly pushed it,
And I don't know if it was the problem because when I saw it sticking out of the case range I ran Heaven benchmark few times on ultra and it never crashed, not even once, tho I did not check the other problematic games where it always does crash, can it be the problem?

+I have screen recordings of the voltages monitoring of the CPU and GPU when running the heaven benchmark on low and ultra (before I opened the case), do you want to see it and tell me if you see the problem? because I dont see anything.

I was thinking it might be a cooling issue, but the temp were always fine on every game, not going over 70. and averaging on 56-60

 

[bug]

Yes, a loose cable would definitely cause the kind of instability you were experiencing.
Inspect the cable closely, if it was loose for some time, it may have melted around the connectors. Replace it if it did.

 

[idanlevayev]

I have uploaded a few photos, it looks fine
And the last photo is how it looks like the cable is sticking out when the case is open.

 

[pavle]

Please, also re-test the games, now that the cable has been re-seated (and inspect it).

Edit: it looks alright to me.

 

[idanlevayev]

But if it really is the problem, how come that the heaven benchmark could run on low without crashing and on higher quality it always crashes, (now it does not)
+it crashes on a specific game which is not the most demanding game I am playing.. :\

 

[BoggledBeagle]

Your CPU may have degraded. Just change voltage to automatic with offset mode and add 50mV to see if it will stabilise the PC.

 

[idanlevayev]

Your CPU may have degraded. Just change voltage to automatic with offset mode and add 50mV to see if it will stabilise the PC.

What do you mean by degraded? how can it happen?

 

[BoggledBeagle]

Degraded - it needs more voltage to run properly.

It may be weak chip, or subjected to too high power draw, too high temperature, too high voltage, etc.

This is the easiest troubleshooting thing to try. If you suddently start getting an instability with Intel CPU, this it the most obvious reason.

BTW did you get this chip as new, or does it have previous owner/owners?

 

[QuietBob]

I was thinking it might be a cooling issue, but the temp were always fine on every game, not going over 70. and averaging on 56-60

Are these CPU temperatures? Could you run the multicore CPU test from Cinebench 2024 and let it finish while monitoring the CPU temp in the background? What's your peak temperature? Is the PC stable during a full pass of this benchmark?

 

[idanlevayev]

Are these CPU temperatures? Could you run the multicore CPU test from Cinebench 2024 and let it finish while monitoring the CPU temp in the background? What's your peak temperature? Is the PC stable during a full pass of this benchmark?

Hey I used Intel Processor Diagnostic Tool to test the cpu and it says it had no errors,
Tried the cinibench, I tried running the multi-core and after a few seconds, it just stops and freezes or just crashes to desktop.

Please, also re-test the games, now that the cable has been re-seated (and inspect it).

Edit: it looks alright to me.

Just launched the game, and got a black screen and PC rebooted, not sure what it might be.

Are these CPU temperatures? Could you run the multicore CPU test from Cinebench 2024 and let it finish while monitoring the CPU temp in the background? What's your peak temperature? Is the PC stable during a full pass of this benchmark?

Hey just did the cinibench again after the PC rebooted,

I am pretty sure that is a problem.

 

[BoggledBeagle]

The CPU will hit 100°C when you try to push more heat in the cooler than it can handle. Your cooler can comfortably handle just 160W . You have to limit the power draw of the CPU to 160W. The CPU should be able to deal with such situation without crashing.

If you have been running the CPU at 100°C it migh have degraded because of that.

Could you PLEASE try to increase the CPU voltage?

 

[idanlevayev]

The CPU will hit 100°C when you try to push more heat in the cooler than it can handle. Your cooler can comfortably handle just 160W . You have to limit the power draw of the CPU to 160W. The CPU should be able to deal with such situation without crashing.

If you have been running the CPU at 100°C it migh have degraded because of that.

Could you PLEASE try to increase the CPU voltage?

Hey, thanks for the answer, how do I do it, and how does it affect performance id you don't mind me asking

 

[QuietBob]

Hey just did the cinibench again after the PC rebooted, 100C I am pretty sure that is a problem.

Yeah, 100 C within one minute of launching the benchmark is a sign of thermal throttling, which could introduce instability. I'd try limiting maximum allowed CPU temperature in the BIOS first.

Go to AI Tweaker Menu, then Digi+ VRM and Internal CPU Power Management. Set Maximum CPU Core Temperature to 90 C and re-run Cinebench. Does it complete the 10-minute run without freezing/crashing now?

 

[BoggledBeagle]

Hey, thanks for the answer, how do I do it, and how does it affect performance id you don't mind me asking

You are not meant to run it with positive voltage offset. If you figure out that this is the cause for your stability problem, you need to return the CPU as defective (it cannot run stable at stock settings) and get a new CPU or refund.

And the next time please make sure that your CPU is not overheating and you have set the power limits properly to what you cooling solution can handle.

BTW there is nothing wrong with your cooler, the CPU still has plenty performance even with 160W power limit!

 

[idanlevayev]

I think it might be the problem, on the problematic game it hits 90+ and even 100, how do I limit it, and how do you know its supposed to be 160W?

I am trying to find online on how to do it, I am getting more confused.

 

[BoggledBeagle]

You need to set both power limits in BIOS to the same number. 160W was my educated guess, you need to test if the cooler can handle that. I think it could.

Page 38 of the BIOS manual for your board (english version).