What causes my PC to crash and is there a dump file available?
What causes my PC to crash and is there a dump file available?
Hello.
Please review the dump file and let me know what might be causing the crashes.
Hello, and thank you for joining the forum! Could you provide more details about the crash? What were you doing when it occurred? Was it linked to a specific application? Is this issue consistent or sporadic? Additionally, having the complete specifications of your computer, including power supply information, would be very useful.
Regarding the diagnostic data, it's generally safer to rely on multiple dumps rather than just one. Since you mentioned several crashes, please share the other dump files as well. It’s not necessary to upload the kernel dump (memory.dmp); the minidump files located in C:\Windows\Minidump should suffice for most troubleshooting needs.
This particular crash occurred because of a graphics issue. We observed the Windows graphics Timeout Detection and Recovery system (TDR) attempting to restart the graphics driver and card due to a hang or stall. The TDR reset failed to restore the graphics subsystem, resulting in the BSOD.
The root cause is likely either an outdated or faulty graphics driver or a defective graphics card. First, ensure you're using the most recent driver available. Although you didn't specify your graphics card model, the driver installed on 15th August 2025 is quite old and probably needs updating. Use DDU to completely remove the existing driver before proceeding.
Also, it appears you have the Nvidia audio driver installed, which can sometimes conflict with the Realtek audio driver you have. If you don't require the Nvidia audio driver, simply deselect it during installation (opt for a custom install).
If the BSOD persists after updating the graphics driver, download and run SysnativeBSODCollectionApp from the Desktop. Upload the generated zip file to a cloud service with a link, as this tool compiles all available troubleshooting data to simplify diagnosis. It does not store any personal information and is trusted by many reputable Windows forums.
Feel free to inspect the contents of the zip file before uploading, though avoid modifying or deleting any files. For detailed descriptions of each file, refer to the provided link.
I have a thread here with details about the situation.
Question - MSI MAG B550 Tomahawk GPU issue?
Hello tech enthusiasts, I’m facing a problem with my MSI MAG B550 Tomahawk. The client has an MB and a Gigabyte RTX 3070Ti after about two years. It started losing picture, then partially recovered, but now it’s unstable. After some troubleshooting, I managed to get it working again three weeks ago. I’ve tried a few fixes, but the issue persists. I recently acquired the GPU and have no idea what’s happening. The computer works for a couple of weeks, then crashes with BSODs and repeatedly tries to start up, preventing access to Windows.
System Specs:
- CPU: Ryzen 7 3700X
- GPU: Gigabyte RTX 3070Ti Gaming OC 8GB
- Motherboard: MSI MAG B550 Tomahawk
- RAM: HyperX Predator DDR4 32GB 3600MHz CL17 DIMM (2x16GB)
- SSD/HDD: WD SN750 1TB Black
- PSU: Evolveo FX750 80+Bronze (over two years old)
- Chassis: MSI Gungnir 100R
- OS: Windows 10 Home
In the Event Viewer, I found errors like:
"Computer was restarted from process control error: 0x00000116..."
I don’t have deep technical expertise to resolve this. Any advice would be greatly appreciated.
It's not advisable to engage in discussions across multiple forums simultaneously. I don't regularly visit the Motherboards forum, so it's unlikely many users there are here. We can't truly gauge what others recommend on other platforms. You should choose one forum to focus on and pause the others. This might seem risky, but we currently lack any concrete evidence of a problematic motherboard.
Mobo
support page
- Have you refreshed the BIOS after receiving the board? Never, or not at all? Refer to the manual for update procedures.
Memory appears compatible with the board – likely
HX436C17PB3K2/32
DDR4
2400
3600
SK hynix C
1.35
DUAL
16
√ | √ | √
Could you identify the memory model in
cpu-z? There are many results at the variety level, but it’s not clear what you’re looking for.
You might want a menu if you repeatedly fail to boot and manage to reach safe mode+networking – the exact steps aren’t remembered. Avoid frequent trips to the system settings just to learn it by heart.
If you can’t even enter safe mode, it may be necessary to reinstall winders.
From safe mode or the command prompt, you can run commands like DISM and sfc /scan.
But honestly, these often don’t work unless you have a restore point or a system image backup – something you’d only have if you’re familiar with troubleshooting.
Here’s a resource to help you understand some of these steps:
win 10 clean install guide
You should be aware that Windows 10 support is ending?
It’s possible the issue you face could be linked to the very next day, which might just be a coincidence.
While there’s a willingness to assist beginners, it seems the chance to rescue them from technical difficulties isn’t always strong.
You’ll need detailed notes on your Windows setup, license key, and backup plans for data. Consider having a Linux environment via a live USB to access help or downloads in a more user-friendly way.
To be honest, it probably started with a problem related to the client’s GPU. If it isn’t working, it could be a contributing factor. If it was working before, you won’t know for sure.
What a situation – it should not have been an issue unless it was.
There might be unconventional solutions, but substituting parts usually isn’t free or worth the cost. Once the damage is done, the advice to prevent further issues is better than treating the symptoms.
PSU:
Evolveo FX750 80+Bronze (at least two years old)
Not listed on the Spl’s
Tier list. What we know is it isn’t Tier A. It could be a contributing factor – such as frequent BIOS issues, not updating to version 11, managing Windows keys with an MS account or backed-up data, and experiencing GPU sag.
Witcherses: sometimes things are more than one thing, and now there are four.
I am aware, but noone was responding anymore. I thought the problem was motherboard, but I cannot be sure. Can be faulty gpu. Or so many other things.
We got windows error so we went info safe mode and performed sfc /scannow - it repaired some files now I am waiting for response if it starts up normaly into windows.
Right now GPU was extremely hot and he had to turn the computer off.
If we get into windows I will try your suggestion
I changed the BIOS recently, and I believe this hasn't happened before. Drivers were updated on Saturday and yesterday when attempting to download a new one, it triggered a BSOD and Windows wouldn't start. A clean reinstall of Windows is definitely an option, but I would prefer to perform a fresh installation of Windows 11 at this stage.
I created a bootable USB with Ubuntu a few years ago, and it functioned properly back then. Interestingly, the computer has been operational for several weeks before another issue arose. It occasionally works for a while, then another problem appears. If I could reproduce the issue, it would be great, but I can't.
This highlights an important aspect of the issue. It points to a potential hardware-related problem. I've reviewed the logs and noticed a checksum discrepancy error... The system indicates that the checksum of the watchdog.sys driver from Microsoft doesn't align with the one in your dump. This is commonly associated with RAM concerns.
It might be wise to first address any RAM overclocking. The details in post #5 mention a clock speed of 3600 MHz, which is an overclock. However, the official certification for that RAM is 2400MHz. While it may function at higher speeds, AMD doesn't guarantee stability beyond their certified limits. You should test the system with the RAM running at its standard 2400MHz to see if it remains stable. If it does, consider increasing it to 3200MHz and reassess.
Additionally, performing a Memtest86 check on your RAM after removing any overclocking is recommended. Use the free tool from downloads, such as imageUSB.exe, and create a bootable USB drive. Run the tests on another machine if possible, since you can't fully rely on your current hardware at this stage. After four rounds of 13 tests, restart Memtest86 and repeat the process again. Even minor errors should be caught during these checks.
Its true that I haven't considered RAM as an issue.
I will check what is turned on, I know that there is some built-in overclock for components for Intel (if I recall corectly) - if there is amd option I will turned that off or find the overclock to turn it off.
Thank you for your feedback, i will get back to you as soon as I will have more information