Good morning,
I am experiencing some issues with my board (128GB model). In both Windows and Fedora Linux I am getting crashes and hard lock ups in AI and gaming. BIOS is at version 3.04, tried Frameworks driver pack (windows), latest AMD drivers, multiple kernels in Linux.
Looking for some advice. I see some people have rolled back their BIOS to 3.03. Any other thoughts would be greatly appreciated.
Kyle
Any more details on the said crashes ?
In Windows some of the kids games have just locked hard. Needed to long press the power button to re-boot the system. Event viewer has no clues.
Loading LLMs Fedora Linux under LM Studio has gotten less reliable. LLMs frequently get stuck close to the end of the load and now have caused kernel panics with large ones. If you know what logs I could pull to help identify the issue that’d be great.
Thank you!
Kyle
What games, did you got a hard lock on? And what timeframe, like in the first 10 minutes or after sometime, and thus it happen only in windows.
What is your current driver versions for the gpu/chipset & was the driver a upgrade install or a fresh install by using AMD Cleanup Utility for windows?
I’m testing myself with windows only, so can’t help with linux, but that info would be help full for community at large.
Ps, When a game locks thus the game sound still keep going?
In the last week I’ve had three hard locks in Windows. Also on Bios 3.04, 128GB model.
Twice while playing Battlefield 6, probably ~10 minutes into booting. No sound, screen frozen, can’t get into task manager.
I also had 1 hard lock using LM Studio while loading a model, also no ability to interrupt and so had to hard power off.
I also boot with Fedora on this machine and not had an issue yet but that could be coincedence rather than a rule.
Hi,
Currently on:
Chipset: 8.02.18.557
Graphics: 26.2.2
I have also tried with Framework’s package. Had the same result. Sound stops as well. No mouse input. Fans keep running hard.
Windows 11: 25H2
Had Insurgency Sandstorm and Battlefield 1 lock up hard around 5-15 minutes in. But, I was able to run Furmark in GL and Vulkan for more than 20 mins using far more GFX % without crashing.
Kyle
I did do some testing, and got stuck in windows safe mode boot loop, when I tried to use (AMD Cleanup Utility) because Windows Hallo PIN didnt work in safe mode, and I didn’t get any Sign-In option to swich to password.
After a hour with AI, only by restarting while pressing down shift to get in to windows recorery than Troubleshoot > Advanced and open *Command Prompt and use bcdedit /deletevalue {default} safeboot press enter. If it successful, then use exit and select continue what will restart the system.
Not even system restore got me out of the safe boot loop, because you cant login with a pin.
I need to disable my PIN before, using AMD Cleanup Utility or make a extra local admin account with password only.
*(If bitlocker is set you need your bitlocker pass string to open Command Promp in Windows recovery, make sure to get it before trying to use safe mode by using this: manage-bde -protectors -get C:in command promp in admin mode in normal windows when your still logged in, or get it from your MS Account if your lucky as a last resort. Or you might have to reinstall Windows!! So get your bitlocker pass string added to your password manager so you got it save for future use)
Just some notes for me/community.
Now, I did do some testing before all of this safe mode crap.
@Kyle_Morrison & @Guest584 There is a unstable boost going on for the first 9 minutes. I fix this by reseting game profiles in AMD Adrenalin Edition client. Go to Gaming tab and at the leftside on the bar with Games & Graphics you will find reload icon use it and select Reset all Experiences that fixed that behaviour for me, even if I put a game on HYPR-RX profile.
Next for testing/default, make sure to set under Globel Gaming Experience part of the Graphics tab to Default and make sure everything is OFF with the sliders than in your game, go in the game settings menu, make sure to disable AMD FSR Upscaling and AMD FSR Frame Generation options and restart your game after to make sure those are off in your game.
Also let me know if any other games you got do work without problems? may be not based on unreal or frostbite engines. Just have look at changelogs for the drivers > here as there’re some issues.
Don’t enable anything on the game profile while testing, and always keep Globel Gaming Experience on Default with everying OFF with the slides. If your game is stable you can try on the games tab to change something per game.
If that’s not working, I recommend contacting support. And if you get to a solution, please to put it here, so not everyone has to open a ticket. 
Been having the same kind of crashing with Marathon, NAKWON, and Wreckfest 2. Either the system would hard freeze or a BSOD with ‘DPC_WATCHDOG_VIOLATION’
AMD Software: Adrenalin - 26.2.2
VRAM set to 16gb for the 64gb Desktop
AMD Chipset 8.02.18.557
WhoCrashed
Event Viewer
The computer has rebooted from a bugcheck. The bugcheck was: 0x00000133 (0x0000000000000000, 0x0000000000000500, 0x0000000000000500, 0xfffff807ee3c43b8). A dump was saved in: C:\WINDOWS\Minidump\031926-21109-01.dmp. Report Id: 45c5888e-8c37-4824-954b-d888a375830c.
Going to resort back to using the drivers provided directly from Framework for the time being until I notice a new AMD update come out. Just going to use DDU as AMD Cleanup just always wants to run in safemode but Windows 11 sucks as last time it had that ‘no pin available’ issue and not going to be stuck in that softloop again.
I’m still on 26.2.2, I didn’t have much time to do some game testing yet, as my elderly mother is not doing well at the moment. And that is mostly been eating up my time, to even make use of my framework desktop.
I only got one lock up with sound still going, on 3dmark Night Raid test. I was planing to try GPU driver clean up, but got a PIN softloop, atleast I learned something, how fix that. Anyway let us know how 25.10.2 is going for when it comes to stability? I will be testing different games like Elite Dangerous, No Man’s Sky & 7 Days to Die. I’m not much a shooter type, plus I’m coming from Xbox.
As for the AMD Cleanup Tool, you can still run it in normal windows. You just click on NO, if it asks you to restart in safe mode, or you create extra local account with password for it.
I would not let that tool put me in safe mode again, as it edits bcdedit, just start up safe mode manually by pressing shift key while you reset than Troubleshoot > Startup Options! Make sure to get your bitlocker key string, before you do that, if you using ms account. You will need that to even get in to Startup Options.
Also AMD just has released: Adrenalin Edition 26.3.1 (change logs). I will wait a bit to test that one, may be there will be 26.3.2 released before the end of this month.
I went back to framework drivers and had a crash then upgraded drivers via AMD Adrenalin and it crashed again. I’m not much of a windows power user so any suggestions I will try and follow, I’m tempted to just try and re-install windows and start again. It’s just a gaming partition for me and the fact I haven’t had the same in Fedora makes me think I’ve just got some weird software interaction.
Now I was a windows power user with windows 10, but 11 is abit of a mess right now.
But let us know if windows install works, as your supposed to install windows and then update it first before installing the driver pack. I’m not sure if I did it that way, plus there is still much to try before I do that. And event viewer is also looking good, no critical error’s, when I got gpu system hang I don’t get any logs of the event so that it not helping.
I did do a small test yesterday, while I was recovering my pc elite dangerous account, with the game elite dangerous. And it looked good, with v sync/ 60 fps limit, but when I disabled it, and opend the galaxy map the frames when to 400+ to 10 to 100 and the system locked up for me again. It reminded me of the early xbox series x days, as for me the sound still keeps going, after a minute or 2 that sounds starts to stutter more and more.
Looks like 26.2.2 is no good, I think it’s a spiking issue with the this driver. I will have to test it more, by putting the game in windows mode, and running hwinfo with some graph open on temps and gpu clocks so I can see what is going on. But first I will do some memory testing, when I have time, before getting back to this issue. As I would like Framework to update their OEM drivers, and that will take time anyway, if they even test windows gaming at all.
I’m also still on 3.0.3 bios, but I don’t think it a bios issue, plus I would like to wait for 3.0.5 full release. My understanding the beta is set somewhere in early April, so may be end April or early May for full release if we are lucky.
@Nix1 Did going back to the driver pack driver 25.10 fix your gaming crashes? I only did do some 3dmark tests, but for that driver it was stable for me. It also 5 to 20% slower than 26.2.2. I don’t remember the gpu clocks in the OEM driver, but I do believe it was 200 to 400 MHz slower then 2600 MHz
@Guest584 ps, Do you game on Lixus as well? And that is stable, if you do game? What driver version is that using?
For me to switch to Linux is the last option, as I’m a Linux noob. Plus I don’t like to run a dual boot on my 1 ssd with windows, as windows is known to break that. So it’s one or the other for the moment.
Small update: since using the AMD cleanup tool and then installing the latest AMD drivers I have not had any game freezes in Windows 11. Installed version is 26.3.1 for the graphics and 8.02.18.557 for chipset.
Haven’t gone back into Linux just yet. Thinking I’ll wait for the new Ubuntu LTS and start again there.
Kyle
I’ve hadn’t been able to be exactly sure since I forgot to also uninstall/stop the AMD Install Manager from auto-updating and it updated me to a newer one then the one that I was crashing on rather frequently. Version 26.3.1 seems to have been a lot better than 26.2.2 in regards to these hard system freezing that I’ve gotten, probably be another week or two before I post another update regarding that prior hard crashing I kept getting on multiple games, if it still happens or if it was just that driver version being really bad.
I didn’t have the time, to even turn on my framework desktop since last post. I’ve been too busy to play with it. But there is some progress of frameworks side, a new bios in beta and they are working on a new drive pack update.
I do hoop they atleast do some game testing on they’re own, with default driver they will update for their driver pack with.
As for me, I will use AMD cleanup tool when it comes to trying out a new driver for gpu 8060s.
When I have time to really test en play with my fw desktop build.
I just need to keep in mind some windows issues like no pin support with safemode, and go manually in safemode. Don’t let that AMD cleanup tool do it.
If anyone has system freezing in windows, please add those to this topic with what game you got system freezing on! And what driver you are using, with your update path: (From the first time you install the driver pack, until now, unless you used AMD Cleaner Tool)
Would be that I had another system hard freeze just playing Roblox with a friend.
Same generic DPC_WATCHDOG_VIOLATION
Going to go through my Control Panel > System and Security > System > Scroll in the box for Device specifications, related links > Advanced system settings > Startup and Recovery Settings > and switch from Small memory dump to Automatic memory dump to try to get a better dump file to read with WinDbg as it’s also not showing much information.
WhoCrashed only able to say “The DPC watchdog detected a prolonged run time at an IRQL of DISPATCH_LEVEL or above.”
BlueScreenView only able to grab that ntoskrnl.exe was the fault,
ntoskrnl.exe ntoskrnl.exe+29c95c fffff80499860000 fffff8049acb0000 0x01450000 0x4473d78d 5/23/2006 9:48:29 PM Microsoft® Windows® Operating System NT Kernel & System 10.0.26100.8117 (WinBuild.160101.0800) Microsoft Corporation C:\WINDOWS\system32\ntoskrnl.exe
AMD Installer updated me to
Adrenalin Edition - 26.3.1
Chipset 8.02.18.557
Framework BIOS tool showing I’m still on the latest,
LFSP0.03.04
Had a new hard crash, this time, DRIVER_POWER_STATE_FAILURE
WhoCrashed
Crash dump file: C:\WINDOWS\FULLMEMORY.DMP (Full memory dump)
Bugcheck code: 0x9F(0x3, 0xFFFF888D22AF0060, 0xFFFFF800ADB99000, 0xFFFF888D2B426C10)
Bugcheck name: DRIVER_POWER_STATE_FAILURE
Driver or module in which error occurred: fffff8011a4fc38c : 0000000000000000 fffff800adb993f0 ffff888d4e80f498 fffff800adb99189 : nt.sys (fffff8011a4fc38c : 0000000000000000 fffff800adb993f0 ffff888d4e80f498 fffff800adb99189 : nt!KeStallWhileFrozen+0xb81d)
Bug check description: This bug check indicates that the driver is in an inconsistent or invalid power state.
Analysis: A device object has been blocking an IRP for too long a time. This is likely caused by a hardware problem, but there is a possibility that this is caused by a misbehaving driver.
This bugcheck indicates that a timeout has occurred. This may be caused by a hardware failure such as a thermal issue or a bug in a driver for a hardware device.
Read this article on thermal issuesGoogle query: fffff801
1a4fc38c : 0000000000000000 fffff800adb993f0 ffff888d4e80f498 fffff800`adb99189 : nt DRIVER_POWER_STATE_FAILURE
Was trying to play Tom Clancy’s Rainbow Six Siege where it crashed upon loading into the game. The first time loading, the game froze for a few seconds and gave an AMD Timeout Screen, had reported it, second time, was the hard crash above.
After the restart, Windows Update was probably working on an update, but my AMD drivers looks like they were entirely removed, the GPU not even appearing in the task manager and trying to use the Adrenalin tool just says version mismatch…
Just went ahead and tried to reinstall AMD again through their website tool, then the irony..
Unsure if one of the recent Windows Update has caused this, but a screenshot of my Update history incase it’s related to one of these updates for others to compare.
Here would be a WinDbg !analyze -v of the full memory dump,
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
DRIVER_POWER_STATE_FAILURE (9f)
A driver has failed to complete a power IRP within a specific time.
Arguments:
Arg1: 0000000000000003, A device object has been blocking an IRP for too long a time
Arg2: ffff888d22af0060, Physical Device Object of the stack
Arg3: fffff800adb99000, nt!TRIAGE_9F_POWER on Win7 and higher, otherwise the Functional Device Object of the stack
Arg4: ffff888d2b426c10, The blocked IRP
Debugging Details:
------------------
Implicit thread is now ffff888d`186ea040
KEY_VALUES_STRING: 1
Key : Analysis.CPU.mSec
Value: 1734
Key : Analysis.Elapsed.mSec
Value: 7937
Key : Analysis.IO.Other.Mb
Value: 11
Key : Analysis.IO.Read.Mb
Value: 1
Key : Analysis.IO.Write.Mb
Value: 42
Key : Analysis.Init.CPU.mSec
Value: 921
Key : Analysis.Init.Elapsed.mSec
Value: 26570
Key : Analysis.Memory.CommitPeak.Mb
Value: 135
Key : Analysis.Version.DbgEng
Value: 10.0.29547.1002
Key : Analysis.Version.Description
Value: 10.2602.27.2 amd64fre
Key : Analysis.Version.Ext
Value: 1.2602.27.2
Key : Bugcheck.Code.KiBugCheckData
Value: 0x9f
Key : Bugcheck.Code.LegacyAPI
Value: 0x9f
Key : Bugcheck.Code.TargetModel
Value: 0x9f
Key : Dump.Attributes.AsUlong
Value: 0x21800
Key : Dump.Attributes.DiagDataWrittenToHeader
Value: 1
Key : Dump.Attributes.ErrorCode
Value: 0x0
Key : Dump.Attributes.LastLine
Value: Dump completed successfully.
Key : Dump.Attributes.ProgressPercentage
Value: 100
Key : Failure.Bucket
Value: 0x9F_VRF_3_DXG_POWER_IRP_TIMEOUT_IMAGE_ACPI.sys
Key : Failure.Hash
Value: {4450b202-fb03-b620-ae4a-e3d5e61fa358}
Key : Hypervisor.Enlightenments.ValueHex
Value: 0x7497cf94
Key : Hypervisor.Flags.AnyHypervisorPresent
Value: 1
Key : Hypervisor.Flags.ApicEnlightened
Value: 1
Key : Hypervisor.Flags.ApicVirtualizationAvailable
Value: 0
Key : Hypervisor.Flags.AsyncMemoryHint
Value: 0
Key : Hypervisor.Flags.CoreSchedulerRequested
Value: 0
Key : Hypervisor.Flags.CpuManager
Value: 1
Key : Hypervisor.Flags.DeprecateAutoEoi
Value: 0
Key : Hypervisor.Flags.DynamicCpuDisabled
Value: 1
Key : Hypervisor.Flags.Epf
Value: 0
Key : Hypervisor.Flags.ExtendedProcessorMasks
Value: 1
Key : Hypervisor.Flags.HardwareMbecAvailable
Value: 1
Key : Hypervisor.Flags.MaxBankNumber
Value: 0
Key : Hypervisor.Flags.MemoryZeroingControl
Value: 1
Key : Hypervisor.Flags.NoExtendedRangeFlush
Value: 0
Key : Hypervisor.Flags.NoNonArchCoreSharing
Value: 1
Key : Hypervisor.Flags.Phase0InitDone
Value: 1
Key : Hypervisor.Flags.PowerSchedulerQos
Value: 0
Key : Hypervisor.Flags.RootScheduler
Value: 0
Key : Hypervisor.Flags.SynicAvailable
Value: 1
Key : Hypervisor.Flags.UseQpcBias
Value: 0
Key : Hypervisor.Flags.Value
Value: 38670575
Key : Hypervisor.Flags.ValueHex
Value: 0x24e10ef
Key : Hypervisor.Flags.VpAssistPage
Value: 1
Key : Hypervisor.Flags.VsmAvailable
Value: 1
Key : Hypervisor.RootFlags.AccessStats
Value: 1
Key : Hypervisor.RootFlags.CrashdumpEnlightened
Value: 1
Key : Hypervisor.RootFlags.CreateVirtualProcessor
Value: 1
Key : Hypervisor.RootFlags.DisableHyperthreading
Value: 0
Key : Hypervisor.RootFlags.HostTimelineSync
Value: 1
Key : Hypervisor.RootFlags.HypervisorDebuggingEnabled
Value: 0
Key : Hypervisor.RootFlags.IsHyperV
Value: 1
Key : Hypervisor.RootFlags.LivedumpEnlightened
Value: 1
Key : Hypervisor.RootFlags.MapDeviceInterrupt
Value: 1
Key : Hypervisor.RootFlags.MceEnlightened
Value: 1
Key : Hypervisor.RootFlags.Nested
Value: 0
Key : Hypervisor.RootFlags.StartLogicalProcessor
Value: 1
Key : Hypervisor.RootFlags.Value
Value: 1015
Key : Hypervisor.RootFlags.ValueHex
Value: 0x3f7
Key : SecureKernel.HalpHvciEnabled
Value: 1
Key : Stack.Pointer
Value: NMI
Key : WER.OS.Branch
Value: ge_release
Key : WER.OS.Version
Value: 10.0.26100.1
Key : WER.System.BIOSRevision
Value: 3.4.0.0
BUGCHECK_CODE: 9f
BUGCHECK_P1: 3
BUGCHECK_P2: ffff888d22af0060
BUGCHECK_P3: fffff800adb99000
BUGCHECK_P4: ffff888d2b426c10
FILE_IN_CAB: FULLMEMORY.DMP
TAG_NOT_DEFINED_202b: *** Unknown TAG in analysis list 202b
DUMP_FILE_ATTRIBUTES: 0x21800
FAULTING_THREAD: ffff888d186ea040
DRVPOWERSTATE_SUBCODE: 3
ADDITIONAL_DEBUG_TEXT: DXG Power IRP timeout.
IRP_ADDRESS: ffff888d2b426c10
DEVICE_OBJECT: ffff888d186ca040
DRIVER_OBJECT: ffff888d18ed3da0
IMAGE_NAME: ACPI.sys
MODULE_NAME: ACPI
FAULTING_MODULE: fffff800ac230000 ACPI
BLACKBOXACPI: 1 (!blackboxacpi)
BLACKBOXBSD: 1 (!blackboxbsd)
BLACKBOXNTFS: 1 (!blackboxntfs)
BLACKBOXPNP: 1 (!blackboxpnp)
BLACKBOXWINLOGON: 1 (!blackboxwinlogon) (!blackboxwinlogonnotify)
PROCESS_NAME: System
DPC_STACK_BASE: FFFFF800ADBA1FB0
STACK_TEXT:
fffff185`1e01efa0 fffff801`1a258204 : ffffdd01`c8ea2180 ffffdd01`c8ea2198 00000000`00000000 ffff888d`18819040 : nt!KiSwapContext+0x76
fffff185`1e01f0e0 fffff801`1a221eed : 00000000`00000000 00000000`00000000 fffff185`1e01f270 fffff801`1a18c369 : nt!KiSwapThread+0x6d4
fffff185`1e01f170 fffff801`1a2200e9 : 00000000`00000000 00000000`00000000 ffffe776`00000000 00000000`00000000 : nt!KiCommitThreadWait+0x39d
fffff185`1e01f200 fffff800`abd4923b : fffff185`1e01f318 ffff888d`00000000 00000000`00000000 ffff888d`2b75c000 : nt!KeWaitForSingleObject+0x859
fffff185`1e01f2e0 fffff800`abd48f4c : fffff185`1e01f300 00000000`00000004 ffffe552`d1b9c1ec 00000000`00000000 : dxgkrnl!DpiFdoHandleDevicePower+0x2cb
fffff185`1e01f370 fffff800`abd28990 : ffff888d`2b426c10 ffff888d`2b426c10 ffff888d`2b75c180 ffff888d`2b75c030 : dxgkrnl!DpiFdoDispatchPower+0x1c
fffff185`1e01f3a0 fffff801`1a41a557 : fffff185`00000001 ffff888d`4e80f420 ffff888d`4e80f400 ffff888d`00000000 : dxgkrnl!DpiDispatchPower+0xe0
fffff185`1e01f4c0 fffff801`1a3b782a : 00000000`00000000 ffff888d`186ea040 fffff801`1a41a180 00000000`00000000 : nt!PopIrpWorker+0x3d7
fffff185`1e01f570 fffff801`1a5db324 : ffffdd01`c8c64180 ffff888d`186ea040 fffff801`1a3b77d0 00000000`00000000 : nt!PspSystemThreadStartup+0x5a
fffff185`1e01f5c0 00000000`00000000 : fffff185`1e020000 fffff185`1e019000 00000000`00000000 00000000`00000000 : nt!KiStartSystemThread+0x34
STACK_COMMAND: .process /r /p 0xffff888d186b4040; .thread /r /p 0xffff888d186ea040 ; kb
FAILURE_BUCKET_ID: 0x9F_VRF_3_DXG_POWER_IRP_TIMEOUT_IMAGE_ACPI.sys
OS_VERSION: 10.0.26100.1
BUILDLAB_STR: ge_release
OSPLATFORM_TYPE: x64
OSNAME: Windows 10
FAILURE_ID_HASH: {4450b202-fb03-b620-ae4a-e3d5e61fa358}
Followup: MachineOwner
---------
Going to probably be risky, but since this looks to be a ACPI issue, the latest beta bios may help, probably going to give this a try, even though the update info for this bios update doesn’t show anything regarding ACPI, it’s still best to be on the latest to report further bugs.
Hey guys! I’m also having the same issue. Framework Desktop 128GB.
Crashing in Rainbow Six Siege, a few minutes into menu or 30 minutes of being ingame. Computer just freezes (no sound (sometimes has sound but distorted), screen freezes) and I have to hard reboot by holding the power button.
Tested in other games: The Witness, REPO, PEAK.
All will eventually lead to the crash. The longest I had played for any game without crashing was 30 minutes.
Updating the BIOS from 3.03 to 3.04 did not help.
AMD Adrenalin version 26.3.1
AMD Chipset version 8.02.18.557
Did a clean install of windows and the Framework AMD driver bundle. Nothing helps.
Any updates from anyone here?
============================
Going to try the 3.05 beta version next —> Crashed too.
I’m said to say I didn’t get around to test myself.
But I do see some useful information here, from everyone. That got me thinking about limiting FPS to see if the game can stabilize on console level frame rates (30/60fps) as I did see a fps spike before every freezes.
Also I think @ollie96 has more what I’m seeing with my system. As for @Nix1 If you did not do already put everything to system memory and test your memory to make sure it good.
Also and this is just a hunch, but try to set gpu memory with a max of 24gb in the bios, instead via windows/amd driver. As it can be a windows/driver issue with memory handling.
These are you some of my thoughts on this matter.
Any solutions to this issue found? I seem to be getting daily freezing when trying to play games and running discord with camera on. Some high end games like cyberpunk are near immediate, but some lower end games like counter strike are still leading tot he same issue?
I am also using the 64gb Ryzen AI Max+ 395 series.
Still crashing/freezing here. Sometimes it happens the moment I get into the main menu for Rainbow Six Siege. It always happens soon after I get into a match.
Also happening on Roboquest.
It doesn’t happen on League of Legends.
I haven’t tested out video editing and rendering yet but I kinda of don’t want to at this point.
It seems to me that whenever something intensive happens on the FW Desktop, it would lead to the freeze/crash.
I’d say this is a pretty big issue given how many of us here on this post have similar experiences. Really hoping this gets resolved soon.