"PC Fails to Boot After Running Memtest86 – Suspected BIOS Corruption on Mother Board

[Asmodier]

Hello, due to sudden blue screen loops on Windows 11, I analyzed the minidump and suspected the issue was with the RAM. I then ran a test on my RAM using the latest version of Memtest86 on a USB flash drive. The test didn't find any errors (D.O.C.P. was turned off). Afterward, I selected the restart option from the program's interface to exit the software. However, when the computer restarted with the flash drive plugged in, I couldn’t get any display on the screen, and the computer failed to boot.


I tried troubleshooting by testing with one RAM stick, reseating the RAM, reseating the GPU, testing with different RAM and GPU, and resetting the CMOS on the motherboard, but nothing worked. On Reddit, I found only 3 users who had encountered this issue, usually in posts from over 10 years ago, and they hadn't found a solution. Eventually, I took my case to a computer technician, and they suggested that Memtest86 likely corrupted the BIOS on my motherboard.

(Note: During the Memtest86 test, my keyboard, mouse, ethernet cable and two different GPU outputs were connected to the PC.)

(Note 2: Yes my bios were updated except the latest beta version)


MY PC SPECS:

Processor: AMD Ryzen 5 5500 4.20 Ghz 6 Core 19MB AM4 7nm CPU

Mainboard: Asus Prime B450M-K II 4400mhz(OC) M.2 AM4 mATX Mainboard

SSD: Samsung 500GB 980 Evo NVMe Read 3100MB - Write 2600MB M.2

Installed memory (RAM): Corsair 2X8GB Vengeance Lpx 3200mhz CL16 DDR4

System type: 64 bit OS, x64 based processors

Graphics card: Gainward Geforce RTX 3060 Ghost 12GB GDDR6 192 Bit

[David (PassMark)]

Memtest86 doesn't write to BIOS. Generally this is impossible for 3rd party software (as it would be massive security issue if any old software could rewrite the BIOS firmware to brick machines or insert malware).

Firmware needs cryptographic signing and often uses the "UEFI Capsule Update" process. So any attempted changes are generally pointless on modern machines. There is hash verification, version checks, rollbacks, signing, etc....

There are millions of people using Memtest86 over the last 10 years. It isn't surprising that some people found their hardware was actually bad while using testing tools to test their suspected bad hardware.

Correlation is not causation.

[Asmodier]

Memtest86 doesn't write to BIOS. Generally this is impossible for 3rd party software (as it would be massive security issue if any old software could rewrite the BIOS firmware to brick machines or insert malware).

Firmware needs cryptographic signing and often uses the "UEFI Capsule Update" process. So any attempted changes are generally pointless on modern machines. There is hash verification, version checks, rollbacks, signing, etc....

There are millions of people using Memtest86 over the last 10 years. It isn't surprising that some people found their hardware was actually bad while using testing tools to test their suspected bad hardware.

Correlation is not causation.

Actually, I wrote to you for troubleshooting purposes and to inform you, but receiving an arrogant response didn’t surprise me. I wonder which piece of my hardware you believe is low-quality or faulty. As I mentioned above, I used your software because I suspected an issue with my RAM. It seems your software didn’t detect any problems with my RAM. I also stated that I removed each part of my computer one by one and tested it by replacing it with my friend’s components. This trial-and-error process also shows me that there is nothing "BAD (AS YOU MENTIONED)" with my hardware, doesn’t it?

So why would a PC that always boots suddenly fail to boot for the first time after using Memtest86? This raises curiosity as well... Even though I approached you for troubleshooting to consider the possibility of a massive security issue or a software conflict, was such an arrogant and dismissive response really appropriate? Furthermore, if I were the only one to experience this issue, how could I have come across other forums discussing this same problem without any resolution? This too is a question worth asking.

Yet, according to you, my hardware is in a "bad" state, and you thought a simple response would suffice—despite having no evidence to support this claim. Perhaps taking into account everything I mentioned and considering the possibility of a software conflict or security vulnerability could be helpful. Adding it to the "known issues" list for others who may experience the same issue in the future would likely be a better approach, wouldn’t it?

Additionally, I don’t think the issue is 100% caused by the Memtest86 software. However, as a developer, instead of simply saying, "Your hardware is bad," wouldn’t it be more productive to take this into consideration and investigate further? Doing so could help improve your software, don’t you think?

[David (PassMark)]

shows me that there is nothing BAD with my hardware

If nothing was bad why did it blue screen, fail to reboot and necessitate taking the machine to a computer technician?
Clearly there was something bad. And it was clearly bad before you downloaded MemTest86.

You didn't really present any evidence to back up your claim that Memtest86 was corrupting firmware. Or present any evidence that the firmware was actually corrupted. Machines can fail to boot for many reasons.

I wrote to you for troubleshooting purposes

There were no questions in your initial post, just an unfounded accusation we broke your machine.

This forum and post are public and the software is free, we'd have 100,000s of users complaining if there was BIOS corruption somehow occurring via software (despite the technical impossibility of it).

Your computer technician likely also has no idea what is required to make changes to firmware flash memory from code. Further if he knew what he was doing your ASUS board has a button to press to reset the Firmware called BIOS Flashback which would have fixed it (if corruption did really happen).

[Asmodier]

If nothing was bad why did it blue screen, fail to reboot and necessitate taking the machine to a computer technician?
Clearly there was something bad. And it was clearly bad before you downloaded MemTest86.

You didn't really present any evidence to back up your claim that Memtest86 was corrupting firmware. Or present any evidence that the firmware was actually corrupted.

This forum and post are public and the software is free, we'd have 100,000s of users complaining if there was BIOS corruption somehow occurring via software (despite the technical impossibility of it).

So, are you claiming that the sudden blue screen issue I encountered can only be caused by hardware? That it's impossible for a blue screen to result from a software-related issue, such as a Windows update or any other software error? My computer was booting perfectly fine until I ran Memtest86, as I’ve written above countless times. How coincidental could it be that a machine that always booted suddenly stopped booting right after using Memtest86?
It’s impossible for me to provide 100% proof that the firmware was corrupted. I wrote here because I saw 2-3 other people on different forums experiencing the exact same issue. So, does the fact that my computer failed to boot after Memtest86, and that the same issue occurred for three different people in the same way, mean that all of us had "bad" hardware? How does it happen that machines that were booting perfectly fine suddenly fail to boot after running Memtest86 in these cases as well?

We don’t call this a problem affecting hundreds of thousands of people; we call it a rare occurrence with a rare problem. But even rare issues like this can still be investigated, right? If you see a rare side effect on a drug prescription, do you think the response would be, "If this were a real problem, thousands of people would have complained"? Or do you think it should be researched further?

Clearly, since the issue I described has also happened to others in the exact same way, I made a logical assumption and theorized that this very rare issue might be caused by Memtest86. Of course, it’s impossible to provide 100% proof.

Also, I apologize for my accusatory tone; you’re right that starting the conversation in such a manner was inappropriate. However, several reasons and different scenarios all point to the same conclusion, showing that the same issue led to the same problem.

[David (PassMark)]

So, are you claiming that the sudden blue screen issue I encountered can only be caused by hardware?

If machine got BSOD and then doesn't boot to BIOS, then yes it is a hardware issue.

So, does the fact that my computer failed to boot after Memtest86, and that the same issue occurred for three different people in the same way, mean that all of us had "bad" hardware?

Yes, when looked at over a 10 year period.
In the same way that all people who eat McDonalds will die. It is a statistical fact.
If you have a 10 million people eating McDonalds, then a few of them will be dead in the McDonalds car park after their meal. Some things in life are just bad luck. A co-incidence.

What would prove it wasn't a co-incidence would be for example,
1) You press that button and it reloads the firmware, then the machine boots without errors and runs for some time without errors. I am guessing this didn't happen as you wouldn't be posting about this if a button press did actually fix it.
2) Then you run MemTest86 again and it immediately doesn't boot again. But pushing the button fixes it a 2nd time.
3) The same thing happens on at least one other similar machine to prove it isn't just a bad flash chip (Firmware is store on a flash memory chip on the motherboard).

Another way to prove it would be to dump the content of the BIOS flash chip before and after Memtest86 then compare them at a binary level for changes.

We would of course investigate this if we had multiple people claiming it was an issue with the current software. At some point it stops being a co-incidence.

Also note that older computers didn't have much protection for their firmware. It was easier to overwrite it and harder to restore it. Firmware is now very hardened against making changes.

When there are unwanted firmware changes (or limited corruption) a more typical outcome on a modern machine is a boot error like this.

[Asmodier]

If machine got BSOD and then doesn't boot to BIOS, then yes it is a hardware issue.

What would prove it wasn't a co-incidence would be for example,
1) You press that button and it reloads the firmware, then the machine boots without errors and runs for some time without errors. I am guessing this didn't happen as you wouldn't be posting about this if a button press did actually fix it.
2) Then you run MemTest86 again and it immediately doesn't boot again. But pushing the button fixes it a 2nd time.
3) The same thing happens on at least one other similar machine to prove it isn't just a bad flash chip (Firmware is store on a flash memory chip on the motherboard).

My computer was giving blue screens (BSOD), but it was booting without any issues. In fact, I could even load Windows and use it for a while before encountering another blue screen. So, as you can see, my PC could boot both into the BIOS and into Windows, but since it kept giving blue screens, I checked the minidump reports in safe mode.
After analyzing the reports, I noticed a few issues related to hex addresses. As a C++ developer with knowledge of memory management, I assumed these hex address issues might indicate a RAM problem. I posted my issue on several forums, created threads, and shared my minidump reports. After trying several software-based fixes, someone from a forum suggested I use Memtest86 to test my RAM and share the results. Since they shared the same suspicion I had, I decided to give it a try.

I downloaded the latest version of Memtest86 onto a flash drive and booted it on my PC. I ran the test, which took about 2 hours and 47 minutes. The test passed successfully with a result of 47/47, indicating no issues with my RAM. I was relieved and saved the HTML report.

I exited the software from the main menu by pressing "X," which gave me three options. I selected the leftmost option, which was "Restart." The PC restarted, but after that, it no longer booted—not even to the BIOS. It stayed stuck on a black screen.

I tried different RAM slots, booting with a single RAM stick, resetting the CMOS, and reseating the GPU and RAM, Changed with my friends components to check if its caused by my hardware but it wasnt realted to my hardware, none of these steps fixed the issue. I then had to take the PC to a technician for repair.

So, as I mentioned before, it seems like there was no issue with any of my hardware components. (AND LET ME REPEAT: MY COMPUTER WAS SUCCESSFULLY BOOTING TO THE BIOS BEFORE RUNNING MEMTEST86).

If you still believe this is a coincidence or insist it’s a hardware issue, fine. I won’t argue with you any further and will accept your position. However, if I find after the repair that the blue screen issue was software-related and manage to fix it, I won’t risk running Memtest86 again just to test my theory because reprogramming the BIOS was an expensive repair for me.

It’s clear you will continue to think there was no issue caused by your software. But if someone experiences the same problem in the future, I kindly ask you to remember this post, keep it in mind, and check the situation if it ever arises again.

Take care, and by the way, your software is genuinely very useful and well-designed. I hope you achieve even greater success with it in the future. Best of luck!

[David (PassMark)]

it seems like there was no issue with any of my hardware components

I then had to take the PC to a technician for repair

This, to me, is a contradiction.

How can this technician fix this via software when the machine doesn't even start to load any software.
If he thinks it is corrupted BIOS, just press the Flashback button. Should be fixed in 5minutes, max.

This post is public, people are welcome to post evidence of their corrupted firmware. We'll investigate any evidence provided.

[Asmodier]

This, to me, is a contradiction.

How can this technician fix this via software when the machine doesn't even start to load any software.
If he thinks it is corrupted BIOS, just press the Flashback button. Should be fixed in 5minutes, max.

If there is no BIOS firmware left, the computer cannot boot, so of course, it needs to be taken to a technician. Both for a hardware check (just to make sure) and to reinstall the BIOS firmware, because I don’t have an EEPROM USB (BIOS programmer) or any experience in flashing a motherboard BIOS this way. Also, how can I troubleshoot a computer that doesn’t boot on my own?

The idea of using the BIOS Flashback button didn’t occur to me because I thought it was only for updating the BIOS, which I’ve done many times before. I’m not even sure if the Flashback button can recover the BIOS firmware. I don’t know if the technician tried this. I haven’t gotten the computer back yet, and since the technician is a friend of my friend, my friend is handling the communication. If it turns out the technician knew about the Flashback button and tried it but it didn’t work, then the BIOS is truly corrupted, and the firmware will need to be reinstalled using a programmer.

However, if the technician knew about the Flashback button and didn’t even attempt it, then I’ve been scammed, and that would be entirely my fault.

Anyway, if the computer comes back with just the BIOS reinstalled and it continues to boot without issues, this will 100% prove that it wasn’t a hardware issue. I will update this matter once it’s resolved.

[David (PassMark)]

I don’t have an EEPROM USB (BIOS programmer

ROM, EPROM and EEPROM are not used anymore.
Modern firmware is in flash memory.
They don't come in sockets anymore either. Un-soldering the chip and reprogramming it correctly is very difficult. Experienced electronics engineer level. Your typical PC repair guy will have no idea about how to do this. Cheaper (compared to the time involved) to just replace the motherboard (or just push that button).

Also, how can I troubleshoot a computer that doesn’t boot on my own?

Replace the motherboard, then the CPU.

If it turns out the technician knew about the Flashback button and tried it but it didn’t work, then the BIOS is truly corrupted

The function of the button it to reload the firmware.
My understanding is that this can be done even if the existing firmware is totally wiped and machine doesn't turn on. See,
https://www.asus.com/support/faq/1038568/

---------------------------------