PC Processors

[CrispyCrunch]

Mainboard: MSI x570 Unify
Mainboard-BIOS: 7C35vA82 (Beta version)
CPU: Ryzen 5900x
RAM: Crucial Ballistix BL2K32G36C16U4B 3600 MHz, 64GB (32GB x2)
Drive: M.2 Samsung 970 Evo+ 1TB SSD
Graphics: SAPPHIRE Nitro+ Radeon RX 5700 XT
PSU: be quiet straight power 11 750w Platinum
OS: Win 10 Pro (64bit) - all updates installed
Chipset driver: 2.9.28.509 (released 2020-11-09)

I first assembled the PC with a Ryzen 3800x a week ago because it was unclear if and when I would get the Ryzen 5900x I ordered. Worked with the included AMD Prism Wrath CPU cooler for one week without any problems.

- Today I installed a Ryzen 5900x and a Scythe Fuma 2 CPU cooler.
- After 20 min the first crash/restart with the following entries in the Event Viewer: WHEA-Logger ID 18 and critical error Kernel-Power ID 41.
- Happens irregularly again and again, sometimes after minutes, sometimes longer: Windows freezes for a few seconds and then the PC reboots. Doesn't matter if load or not.
- CPU temperature between 30 and 40 °C
- Updated to BIOS and chipset driver mentioned above: Problem still exists
- XMP Profile disabled (RAM on 2600 MHz): problem still exists
- CMOS Reset: Problem still exists

Either there is a compatibility problem of something with the CPU, or the CPU is defective?
What to do? Really frustrating.

[Jazzinplayer · Accepted solution]

Im having a similar issue, x570 aorus and 5600x. Have same errors on windows. 

Disable CBP and PBO and run it at default settings (3.7 ghz and xmp on). That works for me. 

[Saaam-chan]

I did that a few weeks ago...

Reinstalled windows and everything and and and.... still getting those black screens..

5600x here.

Got newest drivers, newest BIOS

[B_Junior]

@flyxytk I dont think it helps with errors directly, but you can try it ofcourse, it cant be worse Also Windows will not only automatically detect the model of your CPU, but will also download and install the newest drivers for your CPU once you begin the update process, thats why I reinstall W10 once again without internet connection and install chipset driver before connecting to internet. I didnt actually investigate further...but my stability of cpu is better atm. Yesterday I run chess analysis for 4h and I use all cores, which means that processor was at 100% load all the time and I had no issues. I  still didn¸¸'t receive any restarts...I also found out what triggers higher temps of cpu in idle state. In my case this was software (G.SKILL Trident Z Lighting Control v1.00.22). After uninstalling it, my temps in idle are again very good. 

I still believe in most cases errors (restarts) are bios - driver related and in some small % are ofcourse reason bad chips.

I read some other forums on same topic and many said that after RMA and new processor still have crashes and whea events, so lets wait if AMD will say something about...!?

We are waiting...

[NanoExplorer]

Hey there,

    I'd like to provide an update on my RMA. I have received a replacement CPU and installed it. I posted a while ago that I would see crashes most consistently (~every 10-15 minutes) when I ran a Unigine benchmark such as Heaven or Superposition. I started Superposition 45 minutes ago, and so far no crashes.

I'll post again once I have more data and certainty.

[Dave798]

5800X I am no longer crashing/freezing, I found the culprit in my system, it was neither a CPU or RAM error that was causing my system to crash. AIDA64 was what was causing my faults, this was found by trial and error and took me 3 weeks to find it out, I would advise people with freezing/crashing, WHEA errors first disable all monitoring software ie HWINFO, AIDA64 etc etc. I am no way saying this will fix your problems but it definitely fixed mine for the time being (fingers crossed)

[B_Junior]

@Dave798 , 

"5800X I am no longer crashing/freezing, I found the culprit in my system, it was neither a CPU or RAM error that was causing my system to crash. AIDA64 was what was causing my faults, this was found by trial and error and took me 3 weeks to find it out, I would advise people with freezing/crashing, WHEA errors first disable all monitoring software ie HWINFO, AIDA64 etc etc. I am no way saying this will fix your problems but it definitely fixed mine for the time being (fingers crossed)"

Now, when you mention software was perhaps the reason in your case, I remembered when I build new AMD Pc, I  also installed HWINFO software for monitoring. After reinstalling W10 , I didnt install it back. Maybe, just maybe this can be 1 of the reasons?

Also I am waiting for new BIOS version, which will give us better performance improvement for CPU (they say), so we ll see...fingers crossed , yes

[flyxytk]

@B_JuniorBut that would be idk.. weird? because on my side the crash occurs on Davinci Resolve most of the time and I have no kind of monitoring software opened. I just want my PC to function normally...

[B_Junior]

@flyxytk 

I guess there are many  different potencial reasons for crashing...we are just here to give some info, details about our experience. If there would be solution for this, AMD will already post it.  Something changed in my situation, since I didnt get any whea for almost 2 weeks now, and I just share my changes which I made, on this forum.

[root1308]

Last week I upgraded my system to Ryzen 9 5900X and a XFX Radeon RX 6900 XT, the Motherboard is still the same: MSI MPG 570 Gaming Plus. Unfortunately, I guess, there are some bugs with the hardware monitore software. Before I installed the new graphic cards, I replaces the CPU from Ryzen 7 2700X to 5900X and watched, what my system is doing... after a hour, clicking something in the Firefox browser, system is completely freezing. Via 7taskbar tweaker I set clock to showing the seconds at windows taskbar. Seconds are freezed to. Changed to the eventvwr, I saw critical messages Like kernel power and whea logger.

I thought, maybe something is wrong with windows or BIOS. So after install the the graphics card, I installed windows completely new and upated the UEFI/BIOS.

Actually I use AIDA64 with the sensor panel on a thirds little screen to monitor my hardware... After reinstall windows and restore my sensor panel via backup, everything works fine. But after a couple of hours, same problem with freezing system.

After than I decide to uninstall AIDA64 complete just for fun - no more issues from Thursday to Sunday.

Today (Sunday) I want to optimise and checking some temperatures on CPU, and decide to install HWinfo. No good idea. After installing and running the program, all was fine, but while I opening live monitoring, system crashed immediately with the famous symtoms.

Because of this reason, I point out, that something is going wrong with the hardware-monitor software... on both unindepented constellation with monitor software, my system crashes. And I suspect, it deals something with the hardware access from the software to the CPU sensors or something else.

For the next time, I won't use monitor software and hope, this will be solved in the next weeks via updates.

[NanoExplorer]

Hey Root1308!

   Sorry you're experiencing these crashes. I'd say it's a glimmer of hope that it worked fine for a few days when aida and hwinfo were off. If you didn't already you might try uninstalling HWInfo and replacing it with the latest Beta version of HWInfo, since there was a bug in an older stable version of HWInfo that was triggering this error, and this problem is supposed to be fixed in the beta versions. (ref: https://www.hwinfo.com/forum/threads/is-hwinfo-causing-the-dreaded-whea-logger-event-id-xx-cache-hie... )

An update on my experience: it has now been 6 days since I received the warranty replacement CPU, and I have not had any crashes. I have done several of the activities that would have crashed my old CPU, so it's probably safe to say that there at least was a hardware defect in the old CPU

Raziels_Lament noted that their replacement 5900x was from batch BG 2104PGS and has worked fine. Just for completeness, my replacement 5900x was also from BG2104PGS. My old CPU was BG 2047PGS

[marathonman]

I got one of the early CPUs BG 2946SUS and it was just evil - in my case it would crash randomly during games usually after 3--60 minutes or so.

I spent way too much money and time over a couple of months resisting the idea the CPU was defective:  new RAM, new PSU and about a million permutations of BIOS versions / settings, none of which made a difference.  The one thing that would stabilize the system was to disable core performance boost, but I couldn't live with that.

When I bit the bullet and finally RMA'ed the CPU, AMD were - to their credit - pretty efficient and sent a replacement within 2 weeks - maybe it was10 days.  The new one is BG 2103SUS and it works perfectly fine.

Looking back on it, I'd be less patient next time.  If you have an unstable 5000 series CPU from 2020, my recommendation would be to just send it back and be done with it

[TVicars]

I will add one more to the list on a 2 week old build of a 5600x

Mainboard: MSI x570 Tomahawk
Mainboard-BIOS: 7C84v161(Beta version) (Went to that from 7C84v15)
CPU: Ryzen 5600x
Cooler: Wraith Prism RGB (from a 3800X, but I don't have that CPU to test)
RAM: G.Skill Ripjaws - F4-3200C16-16GVK (2x16GB)
Drive: Boot M.2 XPG Gammix s50 Lite 1TB (also 2 SATA drives, and a WD Black M2 500GB)
Graphics: Gigabyte 1070GTX
PSU: Seasonic X-850
OS: Win 10 Pro (64bit) - all updates installed
Chipset driver: 2.13.27.501 (Just went to this today as it was newer than MSI's version)

Disabled PCO,CBP, Disabled C-State switching, set power to typical voltage, swapped out: Ram (even tried different mem slots), M2 Boot Drive, Vid)

Ran Windows Driver Verifier (only once did it do a Memdump, and it was a WHEA error, the rest of the time it didn't even catch it)

All tests run stable AIDA64, MEMTEST, PRIME95, all run multiple times for multiple hours. Uninstalled AIDA64 to no effect (one thread said that was the issue)

Under stress testing didn't push more that 60c. (Good cooling, and ambient room temp is 16c)

Errors are:

Log Name: System
Source: Microsoft-Windows-WHEA-Logger
Date: 2021-03-07 8:17:58 PM
Event ID: 18
Task Category: None
Level: Error
Keywords:
User: LOCAL SERVICE
Description:
A fatal hardware error has occurred.

Reported by component: Processor Core
Error Source: Machine Check Exception
Error Type: Cache Hierarchy Error
Processor APIC ID: 9

Along with

Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 2021-03-07 8:17:50 PM
Event ID: 41
Task Category: (63)
Level: Critical
Keywords: (70368744177664),(2)
User: SYSTEM
Computer: Main
Description:
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

[ivan-filippov]

To anybody with crashes errors:

Try to run this script
https://www.overclock.net/threads/single-core-prime95-test-script-for-zen-3-curve-offset-tuning.1777...
on default CPU settings - if you get crashes to bsod\errors in console log - most probaly your CPU is not good enough on stock settings and requires additional voltage to work. 

[TVicars]

Update on my thread:

I ended up buying another motherboard / CPU. When I placed the new (5800) CPU in the older motherboard it worked perfectly. I then put the CPU with the errors into the new motherboard (now an ASUS vs the original MSI), and got the same errors.

I finally decided to go through the hassle of an RMA. I am in Canada, and I had to pay to ship it to Florida, (in the US). 

It took a few weeks to get it back, but the new CPU worked without any problems. No issues and no crashing. 

[Blackdove77]

Saw mention of 2020 CPU's. This issue doesn't just come up with those. I have a 2102 which is experiencing the same and awaiting RMA approval.

[root1308]

Hi NanoExplorer,

many thanks for your response. I guess, the best way is to talk to my local dealer for a replacement about my existing CPU.

Unfortunately, currently no spare CPU are in stock. But it's a global problem

I hope I will get a new one and can fully enjoy the power of 12 cores...

Greetings from Germany

[3Dfx]

Now reaching system builders... the widespread issue continues and AMD is so silent...

https://www.pugetsystems.com/labs/support-hardware/AMD-Machine-Check-Exception---X570-Chipsets-2080/

[zukmeister]

I just signed up for an RMA on mine today.  It reset 3 times in the middle of the night last night during idle.  This shouldn't be happening and it's only a matter of time before it resets when I'm doing something really important.  

What sucks is I don't have another cpu to throw in there while I wait for a new one so I'll have to go back to my trusty 8th gen intel laptop that hasn't crashed ever.  Shame on you AMD 

[zukmeister]

how long of a timeframe am I looking at for turn around on 5900x?  Couple of weeks?

[root1308]

Hi NanoExplorer,

today I disassamble the 5900X and saw the number BG 2105 SUS.

So, I have no idea, whats the number indicates.... 2105 --> Manufactured on Week 5, 2021?

Is this serious good?

Greetings, root

[FirefoxNS]

Hola,

Despues de haber publicado una posible solucion con mi Ryzen 3600X

Re: Ryzen 5900x: System constantly crashing/restarting WHEA-Logger ID 18,19 and Graphics

Antes de hacer dichos pasos ejecute este codigo, espero entiendan la solucion posteada, esta solucion la tome de un foro de DELL para procesadores Intel:

https://www.dell.com/community/Optiplex-Desktops/WHEA-Logger-Processor-Core-Internal-Parity-Error-on...

Ejecute este codigo con PowerShell como Administrador:

Link referencia: https://support.microsoft.com/en-us/topic/windows-server-guidance-to-protect-against-speculative-exe...

AMD processors only: Enable the full mitigation for CVE-2017-5715 (Spectre Variant 2)

By default, user-to-kernel protection for CVE-2017-5715 is disabled for AMD CPUs. Customers must enable the mitigation to receive additional protections for CVE-2017-5715.  For more information, see FAQ #15 in ADV180002.

Enable user-to-kernel protection on AMD processors along with other protections for CVE 2017-5715:

reg add "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management" /v FeatureSettingsOverride /t REG_DWORD /d 64 /f

reg add "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management" /v FeatureSettingsOverrideMask /t REG_DWORD /d 3 /f

If the Hyper-V feature is installed, add the following registry setting:

reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Virtualization" /v MinVmVersionForCpuBasedMitigations /t REG_SZ /d "1.0" /f

If this is a Hyper-V host and the firmware updates have been applied: Fully shut down all Virtual Machines. This enables the firmware-related mitigation to be applied on the host before the VMs are started. Therefore, the VMs are also updated when they're restarted.

Restart the computer for the changes to take effect.

AMD processors only: Enable the full mitigation for CVE-2017-5715 (Spectre Variant 2) and CVE 2018-3639 (Speculative Store Bypass)

By default, user-to-kernel protection for CVE-2017-5715 is disabled for AMD processors. Customers must enable the mitigation to receive additional protections for CVE-2017-5715.  For more information, see FAQ #15 in ADV180002.

Enable user-to-kernel protection on AMD processors along with other protections for CVE 2017-5715 and protections for CVE-2018-3639 (Speculative Store Bypass):

reg add "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management" /v FeatureSettingsOverride /t REG_DWORD /d 72 /f

reg add "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management" /v FeatureSettingsOverrideMask /t REG_DWORD /d 3 /f

If the Hyper-V feature is installed, add the following registry setting:

reg add "HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Virtualization" /v MinVmVersionForCpuBasedMitigations /t REG_SZ /d "1.0" /f

If this is a Hyper-V host and the firmware updates have been applied: Fully shut down all Virtual Machines. This enables the firmware-related mitigation to be applied on the host before the VMs are started. Therefore, the VMs are also updated when they're restarted.

Restart the computer for the changes to take effect

Espero les funcione.

[ivan-filippov]

"today I disassamble the 5900X and saw the number BG 2105 SUS. So, I have no idea, whats the number indicates.... 2105 --> Manufactured on Week 5, 2021?"

Yep, BG2105 = the 5th week of 2021

[zukmeister]

I found an old picture of my processor before I installed it... I asked for an RMA today

Processor says BG 2049PGS

I imagine this is a late 2020 processor... ones with reported issues right?

[Numbernine]

Update #3! (Original Post, Update #1, Update #2).

It's been a while since I posted - it took me a little longer than planned to ship the CPU back to AMD (c/o ModusLink) in Florida. The day my RMA was approved we got a bad snowstorm here and didn't feel like venturing to Fedex in miserable weather! Anyway, I digress.

My replacement 5900X arrived today (batch 2104 for those interested - previous was 2051) and I finished installing it about an hour ago. So far, things look very promising! I began by resetting the BIOS settings to default, then enabling an XMP profile. I've ensured that the chipset drivers are up-to-date in Windows 10, and even used Ryzen Master to apply some overclocking to the CPU. Everything seems perfectly stable so far; no crashes under load (e.g. while running Furmark, OCCT, or Prime95) and - more importantly - no crashes while running at idle or near-idle. All of this is using the same hardware configuration that I mentioned in my first post, and without any of the other BIOS workarounds that are frequently touted to "solve" the WHEA problem either (e.g. disabling C-States, disabling CPB/PBO, changing voltages, etc.).

Given the 'spastic' nature of the problem, I will need to run this configuration for a while before I would feel comfortable saying that it's solved, but so far I'm very optimistic; at the time when I returned the CPU, running Windows for more than 20 minutes without encountering the WHEA error was difficult. RMAing the CPU really does seem like it fixed the problem. Note that I do not think this should be the first solution anyone should try - there are a lot of potential causes of WHEA errors and/or Machine Check Exceptions. It's painful, but you really should try swapping all the other hardware you can before reaching the conclusion of, "yes, the CPU is faulty."

---

Now, if anyone from AMD is reading this: let's talk about your RMA process.

Firstly, it's pretty sad that you don't offer an "Advance RMA" process, which is to say a process where I provide you a credit card number, you post a temporary charge for the cost of the CPU, send me the new CPU, I return the old CPU, and you remove the temporary charge when you get the old one back (or permanently charge me if you don't get it!). I'm fortunate enough that I had another CPU I could use during the in-between period, but for those without a backup CPU, they're totally SOL. I've done this process with other hardware manufacturers in the past - why not you? It's 2021 - we have the technology.

Secondly, the shipping back to Florida at my expense is kind of a kick in the junk. I only owned the CPU for a month and then have to pay an additional $80 CAD for a warranty claim on an already expensive CPU? It's seriously making me reconsider AMD CPUs going forward. I'm happy to have a working CPU now, but... yeesh.

To your credit: the CSRs that I dealt with via email where very friendly and helpful, and the time between "defective CPU received in Florida" and "notification of shipped replacement CPU" was in the order of a few hours, which is excellent, and the two-day shipping back to me was very much appreciated.

Overall, the RMA process is okay but has definite room for improvement.

---

I'll post back here again in a few days with a (hopefully) final update on the health and behaviour of the replacement CPU!

[Coachz]

With NEW defective CPU the mg SHOULD pay all shipping to replace it ethically.  Fact is,  demand is high enough they can screw the customer,  it is that simple,

[colesdav]

AMD should pay the RMA shipping costs for faulty CPU/GPU.
They are currently making massive profits. 
Not surprising if they can sell junk CPU for > MSRP price with no RMA for months and the user pays for the return postage...

[rumple]

They replaced my CPU via RMA. And yes, they can provide a prepaid return label. And yes, I've even seen them advance ship a replacement, but it's extremely rare.  The old CPU was an older batch, lower binned part. The new CPU runs 7% faster and 10% colder.  The old FCLK limit was 3733 and the new FCLK limit is somewhere around 3933.  No crashing under load, but it's important to make sure that your all your BIOS settings are correct.  The curve optimizer can't undervolt any core too much.  THe CPU LLC can't be too low.  I'm running SOC 1.13, CCD 1.13, IOD 1.13, VDDP 1.0  and that got it done for me along with CPU LLC 3.  Also note that when just ONE of my RAM settings was too tight, I had errors, so I had to relax a timing setting to finally resolve the last error.   Good luck.  In my case, yes the RMA did the trick.  

[lyme]

"Firstly, it's pretty sad that you don't offer an "Advance RMA" process, which is to say a process where I provide you a credit card number, you post a temporary charge for the cost of the CPU, send me the new CPU, I return the old CPU, and you remove the temporary charge when you get the old one back (or permanently charge me if you don't get it!). I'm fortunate enough that I had another CPU I could use during the in-between period, but for those without a backup CPU, they're totally SOL. I've done this process with other hardware manufacturers in the past - why not you? It's 2021"

I agree with you, and that the shipping charge sucks.. The crappy thing is that if this were an option, there would be people doing Advance RMA on non-existent CPUs so that they don't have to struggle with retail. It is the sad state of the world.

[zukmeister]

They wouldn't... before AMD would RMA my processor they wanted me to prove that I have the processor.  They wanted a picture of it with matching serial numbers installed in my motherboard socket.  AMD is just far behind the times and they suck as it relates to warranties.  30 days for a replacement RMA is unacceptable for tech items 

[flyxytk]

Soo guys after more than one week testing I can finally say: The new Installation of Windows fixed my crash issues completely. So if you didnt try reinstalling windows and manually install drivers do it.

1. Take a Windows Install USB Stick or a CD. If you dont have one create one with the windows media creation tool.

2. Download all necessary drivers (Chipset, Graphics, Sound, Ethernet etc.) to a USB Stick or your Drive (Only if you going to keep the data)

3. Unplug all internet connections

4. Start your PC and get into the boot menu. There choose your USB stick or CD where the Windows installer is on.

5. Choose install Windows in the installer, then accept all agreements and if you want to keep you data on the C Disk dont delete the Partition. Instead install the new Windows over the old if possible. (The drive with the greater space is most likely the right one, search in the internt if you didnt understand what I said).

6. Let Windows install

7.  When Windows is installed complete the setup

8. Search you C Drive in the explorer

9. Go into the folder Windows.old > Users > (your Username) > Downloads and find all the Drivers. If you have downloaded them on a USB drive plug that in and find the drivers on there.

10. Install all drivers you have downloaded earlier

11. Restart your system if the driver installer says it

12. Connect the internet connection

13. Try if the Blackscreen error is gone

Hope I helped ^^

[TexasAVI03]

Look in your event viewer and see if you have a "Critical" Kernel-Power:Event 41 error 10-20 seconds before the WHEA:Event 18 error.

I think Windows Pro 64-bit is not setup for handling the idle cores falling asleep and go below .9 Watts as seen in HWiNFO64-7715.

I have had a my 5900X stable at 5.1MHz using PBO2 Curve on multiple cores but if I stop opening & closing programs (moving the mouse or typing on the keyboard) and/or walk away I will come to a log in screen.  I will check Event Viewer & the Kernel-Power is ALWAYS before the WHEA.  I have never had a BSOD.

All of the Windows complaints about this are usually Laptop & Battery/Power Issues.  I have all of those options changed as well.

AMD is this a possibility? Can we change something on our end?

[TexasAVI03]

Look in your event viewer and see if you have a "Critical" Kernel-Power:Event 41 error 10-20 seconds before the WHEA:Event 18 error.

I think Windows Pro 64-bit is not setup for handling the idle cores falling asleep and go below .9 Watts as seen in HWiNFO64-7715.  or the opposite it is not setup for the extreme wattage change from frequencies over the 4.9MHz limit when using the PBO2 Curve to create more stable Boost Limits.

I have had a my 5900X stable at 5.1MHz using PBO2 Curve on multiple cores but if I stop opening & closing programs (moving the mouse or typing on the keyboard) and/or walk away I will come back to a log in screen.  I will check Event Viewer & the Kernel-Power is ALWAYS before the WHEA.  I have never had a BSOD.

All of the Windows complaints about this are usually Laptop & Battery/Power Issues.  I have all of those options changed as well.

AMD is this a possibility? Can we change something on our end?

[TexasAVI03]

I have continued to read posts on Microsoft of other Windows related thread based upon "Kernel-Power Event ID 41" WHICH IS THE PROBLEM.  It happens 10-20 seconds before the WHEA Error Event.

This is from 2011, Microsoft and/or board manufactures base BIOS has never resolved the issue with an AMD processor (possibly also plus discrete GPU, power supply, but common denominator is AMD) issue: https://answers.microsoft.com/en-us/windows/forum/windows_7-performance/no-blue-screen-kernel-power-...  

Most of these posts relate to an old issue with original JAVA in the original UEFI of the BIOS that manage these boards today.  Stories about contacting AMD 10 plus years ago sound just like we do with a different list of parts in their build.

Then others change their video card.  But I have used a RTX3070 (2 - MISI & Aorus) and Radeon RX 6900 XT (Red Devil) so I can attest that it can happen with both chipsets.

Lastly, changing to a stronger more stable power supply has helped.  I have done this upgrade between 750W to 1000W, between Gold & Platinum.

I have gone days and 18 hours of use without the error.  Then the next morning early it can happen while I am idle reading a webpage (I.E. it just happened, autosave worked) 

However, upping the voltages operating voltages on the board via BIOS seems to work for most.  That is what I am still playing with, and the reason for my stability back to a instant reboot this morning.  I am trying to maintain a PBO2 Curve that gets me to 5.0 to 5.1 PBO2 Boost.

[TexasAVI03]

This is a known issue...https://docs.microsoft.com/en-us/windows/client-management/troubleshoot-event-id-41-restart

Typically, the symptoms described in this scenario indicate a hardware problem. To help isolate the problem, do the following:

• Disable overclocking. If the computer has overclocking enabled, disable it. Verify that the issue occurs when the system runs at the correct speed.
• Check the memory. Use a memory checker to determine the memory health and configuration. Verify that all memory chips run at the same speed and that every chip is configured correctly in the system.
• Check the power supply. Verify that the power supply has enough wattage to appropriately handle the installed devices. If you added memory, installed a newer processor, installed additional drives, or added external devices, such devices can require more energy than the current power supply can provide consistently. If the computer logged Event ID 41 because the power to the computer was interrupted, consider obtaining an uninterruptible power supply (UPS) such as a battery backup power supply.
• Check for overheating. Examine the internal temperature of the hardware and check for any overheating components.

[TexasAVI03]

This was posted on an AMD thread about Kernel-Power Event 41: https://community.amd.com/t5/red-team-discussions/kernel-power-41-rx5700xt/td-p/159128

I do not support or profit form the software listed on this link, but the Windows setup changes should help: https://windowsreport.com/kernel-power-41-error-windows-10/2/

Changing the Automatic Restarts is a good change.

[TrapoSAMA]

Hi!

is correct! i have kernel-power:ID41 before whea ERROR.

i have a powe supply ANTEC hpc 1000w platinum, the 4 memory stick are equal (2 pack 2x8GB) buy the same day.

I try with PBO Curve this time.

[TexasAVI03]

Ok, good deal so like me you feel your equipment (PSU) is ample & above PAR.

And, I also have 4 RAM sticks, B-die,1 Single Channel, so I am running them at 1.45V

Let's try turning off DRAM Power Standby, my current test while fully PBO2 Overboosted via Curve to 5100MHz on Main Cores

Are you comfortable in your BIOS?  I have an ASUS B550-E board with latest BIOS so I can only relate to that.

[TrapoSAMA]

hi!

my system is:

Mainboard MSI MPG X570 Gaming Pro Carbon Wifi Rev 1.0 Bios 1.C1
Ryzen 9 5900x
Termal Pad Grizzly Carbonaut
Sapphire Pulse 5600XT 6GB OC
32GB RAM DDR4 4x8GB 3600Mhz CL16 BL2K8G36C16U4B Ballistix Crucial Black P/N BL8G36C16U4B.M8FE1
Sabrent Rocket 1TB NVMe 4.0
Sabrent Rocket Q 2TB NVMe 3.0
Antec 1000W HCP Platinum
Seagate Ironwolf 8TB 256MB cache 7200rpm
Mouse Razer Mamba Elite
Keyboard Logitech G213 Prodigy Spanish
Cooler Be Quiet Dark Rock Pro 4
2x 140mm Noctua NF-P14s Redux 1700rpm
3x 120mm Noctua NF-P12 Redux 1700rpm
Case CoolerMaster Masterbox CM694

let me check in this bios the similar option.

[TrapoSAMA]

hola!

ningun efecto en mi caso

Mainboard MSI MPG X570 Gaming Pro Carbon Wifi Rev 1.0 Bios 1.C1
Ryzen 9 5900x

Sapphire Pulse 5600XT 6GB OC
32GB RAM DDR4 4x8GB 3600Mhz CL16 BL2K8G36C16U4B Ballistix Crucial Black

Saludos.

[flyxytk]

@B_Junior Its just I dont want to reset my PC. Like im happy with the performance of my system (Ryzen 5 3600, RX 5700XT, 16G Gskill Aegis, Asrock B450m Pro4) but you know these crashes often occur in Davinci Resolve (sometimes MS Edge or Minecraft) and if I am editing a project and the pc suddendly crash it is a really demotivating feeling. I dont loose my projects thanks to Live save but these crashes are demotivating af.

I really like the AMD products but this is not acceptable.

[flyxytk]

@B_JuniorI have now reinstalled windows. The Ethernet was disconnected until I installed all necessary drivers, then connected my system to the internet. Now I'll test my pc and we'll see if the blackscreen of death whea error is gone.