BSOD - 0x124_0_AuthenticAMD_MEMORY__UNKNOWN_FATAL_IMAGE_AuthenticAMD.sys - 1.5 year now!!!!

[ieronymous]

Hi

A quick CV like presentation and straight to the problem. I am an electronic Engineer / IT manager - Admin / Network - Server Architect. I study a lot about every technical aspect that interests me and is needed for my work, either if that will be upon the Network, H/W, server, OSes, S/W field. I ve spent 2/3 of my life in these damn machines, trying to install - configure / tune - maintain - monitor them and it is the first time I can t get a solution. The frustration is extremely high. The fact alone that I still have my pc intact with no wounds from axe or a sword, surpasses me.

The problem started out of a sudden 1 and a half year ago, when my pc was comprised by a 3900x cpu / Noctua D15 cooler/ B550 (Asus Gaming F) mobo / 16Gb DDR4 ram G.Skill Trident 3600 Mhz / 1080 ti Gigabyte Aorus Extreme gpu / Corsair HX850i Platinum psu / Phanteks Evolv X ATX case / Samsung evo 970 m.2 pci-e Gen 3 (I think) / 2 other 2Tb disks WD Red for temporary data / backup / IIyama 34" 1440p 144hz Freesync monitor All that upon a UPS Nova AVR1500VA.
I don t play a lot now (due to my work), but there are a couple of games I try to find time, like Dota2 and mostly RTS games or 3rd person action / beat em up .. etc. The problem was a sudden restart or a bsod in Win 10 back then (latest updates / latest drivers). I installed Windbg and starting collecting the minidump files. The problem in most situations was 0x124_0_AuthenticAMD_MEMORY__UNKNOWN_FATAL_IMAGE_AuthenticAMD.sys. With time, upon troubleshooting this situation, I changed cpu, mobo (2 of them), ram (3 different brands -G.Skill/Hyper X, Patriot), gpu, OS hard disk and in order to keep it simple ..... I changed the whole system!!! which bring us to today where my system consists of
5900x cpu / Noctua D15 cooler/ B550 (Asrock Taichi Razer edition. I also have the none Razer edition) mobo / 32Gb DDR4 ram HyperX 3600 Mhz / 3080 Asus TUF gpu / Seasonic Vertex 1200 Platinum psu / Phanteks Evolv X ATX case / Seagate Firecuda 530 m.2 pci-e Gen 4 / 2 other 2Tb disks WD Red for temporary data / backup / IIyama 34" 1440p 144hz Freesync monitor.

The issue still exists!!!!!!!! I can t think of anything else.
-lowered hz of the monitor
-disabled freesync
-changed all kind of cables and different versions of them (display port, hdmi , power cables)
-change power plug where I connect the pc
-tried and still I am at stock ram hz
-bios update and degrade (between p2.30 and 2.20 the others two which are newer are alphas)
-keep anything stock
-disabled onboard 2.5g lan / wan and blue tooth and bought a usb - ethernet adapter
-changed monitor with a friend of mine (happened to have the same model)
-unplugged the fans of the tower from the motherboard (I was and still do using an adapter 1-3 in order to control them easily - yet I unplugged them)
-checked temps. They are ok (probably I ll upload a photo) both on cpu / gpu and on stress tests the pc is rock solid. (used OCCT mostly but other graphical stress tests too)
-yesterday I re-installed win10 from win11 I had and after frtesh install with only drivers (from vendor) and steam, tried to play dota and had bsod and restart.
-..... probably did other things too I can t recall right now as well (like an exorcism).

... bottom line is, I had enough of this, I m too old to stress my-self any further and this would be the first time in my life probably I m asking for help.

Here is a pic from core temp and gpu-z while I was in game (Dota2).
sensors
As you can see the temps are absolutely fine.

Here are the minidump files from both Win 10 and Win 11

minidump files

Shared with Dropbox

www.dropbox.com

Win 10 are the recent ones, but the latest from win 11 are also valid, since yesterday late night I re installed Win 10,

I need to upload a speccy export as well, later in the day.

Thank you in advance and I apologize for the long post, but I wanted to be as analytic as I could.

 

[ubuysa]

Welcome to the forum! I certainly appreciate your frustration with this.

The Win10 and Win11 dumps are pretty much identical, and that of course suggests that Windows is not the issue and that this is a hardware problem. However, you've replaced all the hardware over time, so that's also an unlikely cause. What's left are the third-party drivers and the third party apps.

It might help if you upload the kernel dump you have, it's the file C:\Windows\Memory.dmp and it will be large. If you can please upload that to Dropbox with a link here. I'm not saying that the answer will be in the kernel dump but minidumps only contain a subset of the system state, so the kernel dump might have some extra clues.

Does it only BSOD when gaming, or can you make it BSOD in other ways?
Does it BSOD as soon as you start the game, or after playing for some time?
Do you get any advance clues that it's about to BSOD?
Does it always BSOD or does it sometime crash without a BSOD?

You've mentioned drivers a couple of times, but it's important we know where ALL the drivers came from...

• Did you run Windows Update after installing Windows - and across reboots - until no more updates were found?
• Have you used the graphics card vendor's website to download drivers?
• Have you used the motherboard vendor's website to download drivers?
• Have you used hardware device vendor's websites to download drivers?
• Have you ever used a driver install tool (like DriverEasy) to find drivers?

When you reinstall Windows, what other apps do you install afterwards?
What external hardware devices do you have plugged in?

There might be some value in uploading the System and Application logs to Dropbox with a link here. Here's how to do that...

1. Enter the command eventvwr into the Run command box. The Event Viewer will open.
2. Locate the Windows Logs folder in the left hand pane and expand it by clicking on the arrow (>) to the left of it.
3. Right-click on the Application entry and select 'Save all events as...'. Choose a folder anywhere that suits you and a filename of 'Application' (an .evtx suffix will be added automatically).
4. Right-click on the System entry and select 'Save all events as...'. Choose a folder anywhere that suits you and a filename of 'System' (an .evtx suffix will be added automatically).
5. Either zip the Application.evtx and System.evtx files together and upload the zip file, or upload the .evtx files as you choose.

If the problem isn't where you're looking then it must be somewhere else.

 

[ieronymous]

Does it only BSOD when gaming, or can you make it BSOD in other ways?
Does it BSOD as soon as you start the game, or after playing for some time?
Do you get any advance clues that it's about to BSOD?
Does it always BSOD or does it sometime crash without a BSOD?

First time when it appeared (win 10 with the old system), could be even when I was browsing.
No, but the time it happens, varies a lot. It could be after 2 mins or after 18mins
Nothing in advance, it freezes along with the sound bsod and reboots.
Sometimes restarts as well. (.... maybe motherboard overvolts the cpu?)

• Did you run Windows Update after installing Windows - and across reboots - until no more updates were found?
• Have you used the graphics card vendor's website to download drivers?
• Have you used the motherboard vendor's website to download drivers?
• Have you used hardware device vendor's websites to download drivers?
• Have you ever used a driver install tool (like DriverEasy) to find drivers?

-Yes I always do.
-Mostly, and I install with DDU and nvclean install in order to avoid installing telemetry and other stuff. But I have install the drivers as they are directly from the vendor(not the last couple of time though).
-Both (for instance now Im on Win10 with vendor's)
-Only with gpu and motherboard
-No. I have better tools but I havent used them in a while now, since I m trying to cover every possible scenario.
On the other hand, we re talking about a clean installation now.

When you reinstall Windows, what other apps do you install afterwards?
What external hardware devices do you have plugged in?

- VIvaldi browser / Adobe acrobat reader dc / Revo Unistaller or BCU Unistaller / VLC / Office 2021 (specific apps, but now I don t have it installed) / Anydesk or Rustdesk / 4k download video / Paragon Partition Manager / Dropbox app / Fast stone image viewer / Photoshop. Usually I install way more programs due to the nature of my work but for now I kept it simple to only needed ones.
-A razer keyboard and a simple mouse (If I recall I have them changed as well). Oh and the TP-Link adapter for internet since there is a well known issue with the killer ethernet E3100 (had many errors in event viewer with this)

I have to be home, for all the other stuff you mentioned.

Thank you for your quick response and your effort to help here.

PS Not a new member here, but it is the first time I m posting due to the unresolved issues I m dealing with.

 

[ubuysa]

Ah, welcome back then!

Please don't use any third-party tools for downloading drivers, stick to Windows Update and the motherboard/device vendor's websites (I appreciate that you haven't used any on this system).

I'm interested that an earlier Windows system would BSOD when not gaming. If you can spare the time it would be good to see whether you can make this system BSOD without gaming. I'm asking that, because if you can make it BSOD outside a game then we can try starting Windows in Safe Mode and see whether you can make it BSOD in Safe Mode (obviously you can't game in Safe Mode). You probably know that in Safe Mode almost no third-party drivers are loaded, so this is a good way of seeing whether the issue is (was) a third-party driver or whether it's something else.

I see that it has sometimes just restarts as well and doesn't BSOD? I think it would be wise then to disable Windows automatic restart, that will give you a chance to see any error messages that are displayed when it just crashes. To do that, enter the command sysdm.cpl in the Run command box. The System Properties dialog will open. Click the Advanced tab, then click the bottom Settings button (in Start-up and Recovery). Uncheck the box 'Automatically restart'. Also be sure that the 'Overwrite any existing file' IS checked for the dump.

It would be useful to see the log entries.

BTW. Have you considered that this might be an environmental issue? I dealt with a similar problem on another forum a couple of months ago which we also think may be environmental. That user is moving house soon and will report back after the move.

The UPS should protect you from anything electrically noisy, but it' still possible that this could be some other device in your home causing noise on the power supply (or even radiated through the air). If you can, try the PC plugged into a different socket, preferably on a different power ring, and without the UPS. The gold-standard test, if you can do it, would be to take the PC somewhere else entirely and see whether it BSODs in a different location. I know that sounds extreme, but stranger things have happened, and we do need to 'think outside the box'..

 

[ieronymous]

There might be some value in uploading the System and Application logs to Dropbox with a link here.

As promised, here are the extra files you want.
https://www.dropbox.com/sh/q4tktpswewdxw9j/AABFJxaQOFrglvK1QXSXDIgDa?dl=0
From a debug on memory.dump noticed the message was some kind more specific
FAILURE_BUCKET_ID: 0x124_0_AuthenticAMD_PROCESSOR_BUS_L1_SRC_IRD_I_NOTIMEOUT_IMAGE_AuthenticAMD.sys
That _Image has to do with the communication of the cpu -> gpu via direct ram?

New edit:
The ram kit currently running is a dual kit 32GB (2X16GB) HyperX Predator DDR4 HX436C17PB3AK2/32 which crashes at both 2400 and xmp 3600 profile.
I ve tried my old dual kit 16GB (2X8GB) G.SKILL TridentZ Neo F43600C18D-16GTZN which crashes at both 2400 and xmp 3600 profile. Notice that Z Neo are more AMD friendly than other kits of the same brand.
Also tried my NAS's ram 32GB (2X16GB) Patriot PSP432G3200KH1 which also crashed.
Finally I had laying around from a customer a dual stick 8Gb (2X4GB) of Hyper X Fury HX426C15FBK2/8 which had no different effect.

..... now I borrowed from work a 64Gb (2 x32Gb) Hyper X KF436C18BB/32 at 3600 but currently running it at 2400 default and game didn t crash at the 30 min duration of it (dota2).
This stick isn t even close to the part numbers QVL list of the mobo suggests while others crashed were.

Is it maybe the amount of Gbs the dimms have? Then again, how come it played fine all the years before?

New edit 2: too soon to talk. Started a second game and let the char at idle. It restarted less than a minute.

New edit 3: Just tried a different gpu Gigabyte 3070Ti and restarted like a bullet within 2 minutes.

 

[ieronymous]

Ok Ive got something to work with here.
Ive overlayed core temp while I was trying to play the game, and noticed spikes on voltage (VID) at 1.475. That seemed high and disabled on bios Core performance boost (Advanced CBS->common options->Core performance boost) and 40 mins after game still on.

I ll keep an eye on that, leaving like this, but my issue now is the clocks don t boost at all and remain all at stock which is 3.6Ghz.

Maybe the whole problem all along is the wrong (high) voltage the mobo gives to the cpu?

 

[ieronymous]

Hi again

It appeared to be the cpu causing the problem. RMA it and after a couple of days got back a new 5900x and since then no problems. I can even run the ram at it's XMP profile without any issues.

Thanks all participants for the help.

PS1 I ve written a 6 pages email to the shop I had purchased the processor from, complaining of what I went through all that time (providing receipts of all the new parts I ve purchased that period, thinking that every other part except the cpu was the problem ) which helped RMA it in only 2 days and replace it with a new model and extra warrantee.

PS2 The shop's tech guy mentioned that in their environment, they couldn t even finish the windows installation which never happened to me. Was it true or they just wanted to get on with my problem without any further delays, I don t know (since I am a good customer of theirs)