MAG B850 TOMAHAWK MAX WIFI + Ryzen 7 9800X3D + RTX 3070 Ti - nvlddmkm Event ID 153 / GPU timeout

[alsi496]

Hello

I am experiencing intermittent GPU-related system instability on my MSI MAG B850 TOMAHAWK MAX WIFI motherboard.

My system configuration:

• Motherboard: MSI MAG B850 TOMAHAWK MAX WIFI (MS-7E62)
• CPU: AMD Ryzen 7 9800X3D
• GPU: Palit GeForce RTX 3070 Ti GamingPro
• PSU: DeepCool PQ850M 850W
• RAM: DDR5 32GB, currently running at JEDEC 4800 MT/s (EXPO disabled)
• BIOS versions tested:
  • Previous BIOS: E7E62AMSI.2A62
  • Current BIOS: 7E62v2A64

Problem description:

During gaming (mainly World of Warcraft, but also observed in other situations), the system occasionally experiences a GPU timeout:

• The image freezes for several seconds.
• Sometimes the screen briefly disappears and recovers.
• Sometimes the GPU driver recovers automatically and the system continues working.
• In more severe cases, the system requires a restart.

After recovery, Windows Event Viewer shows:

• Source: nvlddmkm
• Event ID: 153
• Message:
  • \Device\Video3
  • Error occurred on GPUID: 100

There are no WHEA errors.
CPU and memory stability tests pass.
GPU temperature is normal.
VRAM tests pass.

Troubleshooting already performed:

1. Clean NVIDIA driver installation using DDU.
2. Different NVIDIA driver versions tested.
3. EXPO disabled.
4. RAM tested at default JEDEC 4800 MT/s.
5. GPU tested in another PCIe slot:
   • Secondary PCIe slot
   • PCIe Gen3 mode
   • Power Limit reduced to 80%
6. Main PCIe slot tested with:
   • PCIe Gen4 forced manually instead of Auto.
   • ASPM disabled.
7. NVIDIA Control Panel:
   • Power management mode set to “Prefer maximum performance”.
8. CPU NB/SoC Voltage was tested manually at 1.25V.
9. GPU power connection verified:
   • Two separate PCIe power cables from PSU to GPU.

None of these changes completely eliminated the issue.

After updating BIOS to 7E62v2A64, I tested the system with completely default BIOS settings:

• PCIe settings Auto.
• EXPO disabled.
• RAM 4800 MT/s.

The system appears more stable so far, but I need to confirm long-term stability.

My questions:

1. Are there any known PCIe compatibility issues between MSI B850 motherboards and older PCIe 4.0 NVIDIA GPUs such as RTX 30 series?
2. Does BIOS 7E62v2A64 contain any PCIe/AGESA fixes related to GPU link stability, PCIe power states, or nvlddmkm recovery errors?
3. Are there recommended BIOS settings for this motherboard when using RTX 30 series GPUs?
4. Could this behavior be related to PCIe link training, ASPM, CPU I/O/SoC behavior, or AGESA firmware?
5. Are there any diagnostic logs or tests you recommend to determine whether the motherboard PCIe subsystem or the GPU is responsible?

I would appreciate any guidance or information about known issues with this platform combination.

Thank you.

 

[alsi496]

The error continues in the new BIOS.

 

[Svet]

Could be something wrong with the GPU itself or PSU / power cable related to the GPU.

Check power cables for the GPU
Try a different PSU
Test the GPU in another PC

 

[Svet]

During gaming (mainly World of Warcraft, but also observed in other situations), the system occasionally experiences a GPU timeout:

Which other situations did you mean?

About World of Warcraft:
You can try use DX11 instead of DX12 & retest.

• Motherboard: MSI MAG B850 TOMAHAWK MAX WIFI (MS-7E62)
• CPU: AMD Ryzen 7 9800X3D
• GPU: Palit GeForce RTX 3070 Ti GamingPro
• PSU: DeepCool PQ850M 850W
• RAM: DDR5 32GB, currently running at JEDEC 4800 MT/s (EXPO disabled)
• BIOS versions tested:
  • Previous BIOS: E7E62AMSI.2A62
  • Current BIOS: 7E62v2A64

OS? See: >>Posting Guide<<

1. Clean NVIDIA driver installation using DDU.
2. Different NVIDIA driver versions tested.
3. EXPO disabled.
4. RAM tested at default JEDEC 4800 MT/s.
5. GPU tested in another PCIe slot:
   • Secondary PCIe slot
   • PCIe Gen3 mode
   • Power Limit reduced to 80%
6. Main PCIe slot tested with:
   • PCIe Gen4 forced manually instead of Auto.
   • ASPM disabled.
7. NVIDIA Control Panel:
   • Power management mode set to “Prefer maximum performance”.
8. CPU NB/SoC Voltage was tested manually at 1.25V.
9. GPU power connection verified:
   • Two separate PCIe power cables from PSU to GPU.

Which GPU drivers you have tried or tested?

Also from OS:
Try disable HAGS (Hardware-accelerated GPU scheduling)
then retest.

• The image freezes for several seconds.
• Sometimes the screen briefly disappears and recovers.
• Sometimes the GPU driver recovers automatically and the system continues working.
• In more severe cases, the system requires a restart.

I assume you don't have any OC applied on the GPU right?

Put everything at stock speed and retest.
If it's already on stock:
Apply [-150MHz] to the GPU Core Clock with AfterBurner and retest.

 

[alsi496]

Could be something wrong with the GPU itself or PSU / power cable related to the GPU.

Check power cables for the GPU
Try a different PSU
Test the GPU in another PC

The problem started immediately after I replaced the motherboard and CPU. Before that, I was using an Intel Core i7-8700K with an ASUS motherboard, with the same GPU and PSU, and the system was completely stable.

I have also found quite a few reports on different forums from users with the same or very similar symptoms, especially on newer AM5/B850 systems, but unfortunately there does not seem to be a clear confirmed solution yet.

 

[Svet]

The problem started immediately after I replaced the motherboard and CPU. Before that, I was using an Intel Core i7-8700K with an ASUS motherboard, with the same GPU and PSU, and the system was completely stable.

This could be a coincidence. I would ignore it and proceed with what I already suggested you.

 

[alsi496]

Which other situations did you mean?

About World of Warcraft:
You can try use DX11 instead of DX12 & retest.



OS? See: >>Posting Guide<<



Which GPU drivers you have tried or tested?

Also from OS:
Try disable HAGS (Hardware-accelerated GPU scheduling)
then retest.



I assume you don't have any OC applied on the GPU right?

Put everything at stock speed and retest.
If it's already on stock:
Apply [-150MHz] to the GPU Core Clock with AfterBurner and retest.

Thanks for your suggestions. Here is my full system configuration and everything I have already tested.

System configuration:

• Motherboard: MSI MAG B850 TOMAHAWK MAX WIFI (MS-7E62)
• CPU: AMD Ryzen 7 9800X3D
• GPU: Palit GeForce RTX 3070 Ti GamingPro
• PSU: DeepCool PQ850M 850W
• RAM: 32 GB DDR5, currently running at JEDEC 4800 MT/s with EXPO disabled
• Current BIOS: 7E62v2A64
• Previous BIOS: E7E62AMSI.2A62
• GPU power connection: two separate PCIe power cables from the PSU to the GPU

The most important detail is that this problem started immediately after I replaced the motherboard and CPU.

Before that, I was using an Intel Core i7-8700K with an ASUS motherboard, with the same RTX 3070 Ti and the same DeepCool PQ850M PSU, and the system was stable.

The issue is not limited to World of Warcraft or DX12. It occurs randomly in different situations:

• World of Warcraft
• Dota 2
• different single-player games
• sometimes even while I am only using a browser, Word documents or presentations, with Dota 2 minimized in the background

The usual symptom is that the image freezes for several seconds. Sometimes I hear the Windows device disconnect/reconnect sound, then everything recovers and the system continues working normally.

In some cases the screen briefly goes black and recovers. In more severe cases I have had a complete freeze and had to restart the PC.

After these events, Windows Event Viewer often shows:

nvlddmkm — Event ID 153
\Device\Video3
Error occurred on GPUID: 100

There are no WHEA errors.

CPU, RAM, GPU and VRAM stress tests have completed without errors, and GPU temperatures are normal.

Regarding the GPU / PSU:

• There is no manual GPU overclock
• The GPU is connected using two separate PCIe power cables
• The same GPU and PSU worked normally in my previous Intel/ASUS system
• I tested the GPU in the secondary PCIe slot
• I tested the secondary slot at PCIe Gen3
• I tested GPU Power Limit at 80%
• I already tested -100 MHz GPU Core Clock in MSI Afterburner, but the issue still occurred
• I can now also test -150 MHz Core Clock as you suggested

For the -150 MHz test, I assume I should leave Memory Clock at stock and Power Limit at 100%, so that only the GPU Core Clock is changed.

Regarding Windows / drivers:

• HAGS has already been disabled for a long time, and the issue still occurred with HAGS OFF
• I performed a full NVIDIA driver cleanup using DDU
• I performed a clean NVIDIA driver installation afterwards
• I have also tested different NVIDIA driver versions

Regarding BIOS / PCIe troubleshooting, I have already tested:

• EXPO disabled
• RAM at JEDEC 4800 MT/s
• PCIe Gen4 manually forced instead of Auto
• ASPM disabled
• NVIDIA Control Panel → Prefer maximum performance
• CPU NB/SoC Voltage manually tested at 1.25 V

None of these changes completely eliminated the problem.

I have now updated the motherboard BIOS to 7E62v2A64.

After the update, I intentionally left the BIOS at default/Auto settings:

• PCIe settings: Auto
• ASPM: Auto
• CPU NB/SoC Voltage: Auto
• EXPO: disabled
• RAM: 4800 MT/s

So far, after installing BIOS 7E62v2A64, I have tested the system for several hours with World of Warcraft running together with Dota 2, Discord, Telegram and a browser, including multiple dungeon entries.

So far I have had no freeze and no new nvlddmkm Event ID 153 after the BIOS update.

However, because the problem is random, I need more testing time before I can confirm that the new BIOS actually fixed it.

I have also found several reports from users with the same or very similar symptoms on newer AM5/B850 systems, but unfortunately there does not seem to be one clearly confirmed solution.

Since the same GPU and PSU were stable before changing the CPU and motherboard, I am trying to determine whether this is related to the motherboard / BIOS / AGESA / PCIe platform, rather than immediately assuming that the GPU or PSU is defective.

 

[Svet]

So far, after installing BIOS 7E62v2A64, I have tested the system for several hours with World of Warcraft running together with Dota 2, Discord, Telegram and a browser, including multiple dungeon entries.

So far I have had no freeze and no new nvlddmkm Event ID 153 after the BIOS update.

Got it.

For the -150 MHz test, I assume I should leave Memory Clock at stock and Power Limit at 100%, so that only the GPU Core Clock is changed.

Yes, you can try -200MHz for memory too if needs or no change.

Since the same GPU and PSU were stable before changing the CPU and motherboard, I am trying to determine whether this is related to the motherboard / BIOS / AGESA / PCIe platform, rather than immediately assuming that the GPU or PSU is defective.

Maybe the GPU is no longer working correctly.
If problem persist, I would very the GPU in another system to ensure in present time the GPU is in fully working condition.
(or borrow a different GPU to try)

 

[tfbisqui156f02d4]

Disable all nvidia audio devices in device manager. Maybe it helps.

 

[alsi496]

Unfortunately, the BIOS update did not solve the issue.

I have just experienced another freeze again, and Event Viewer shows the same:

nvlddmkm — Event ID 153
\Device\Video3
Error occurred on GPUID: 100

The system recovered after the GPU driver reset.

I want to mention one more thing. While researching this issue, I found a lot of reports from different users with very similar symptoms:

• nvlddmkm Event ID 153
• black screen / temporary freeze
• GPU driver recovery
• no WHEA errors
• stress tests passing

What is interesting is that many users report this problem not only with RTX 30 series, but also with newer RTX 50 series cards (for example RTX 5070 Ti and RTX 5080).

In some cases, users even replaced the GPU, but the problem remained.

This makes me wonder if there could be some kind of hardware/software compatibility issue related to the MSI platform, BIOS/AGESA, PCIe subsystem, or GPU communication rather than a faulty GPU itself.

I understand that the GPU or PSU still need to be tested, and I will continue with those checks. However, the number of similar reports across different NVIDIA GPU generations makes me interested in whether MSI is aware of any such compatibility issues.

Have you seen similar cases with B850/X870 platforms and NVIDIA GPUs?

I will try the -150 MHz Core Clock test as suggested. However, I have already tested -100 MHz Core Clock before without any improvement, so I am not sure that this will resolve the issue, but I will test it anyway.

Got it.



Yes, you can try -200MHz for memory too if needs or no change.



Maybe the GPU is no longer working correctly.
If problem persist, I would very the GPU in another system to ensure in present time the GPU is in fully working condition.
(or borrow a different GPU to try)

 

[alsi496]

I will try this as well. I have not disabled the NVIDIA audio devices before. I will disable NVIDIA High Definition Audio and NVIDIA Virtual Audio Device in Device Manager and retest.

Disable all nvidia audio devices in device manager. Maybe it helps.

 

[jamarrb103]

I will try this as well. I have not disabled the NVIDIA audio devices before. I will disable NVIDIA High Definition Audio and NVIDIA Virtual Audio Device in Device Manager and retest.

I have the same board and same BIOS. 9700X + 9070XT.

Have you tried disabling the iGPU in the BIOS? Then clean up the AMD driver for your 9800X3D in Device Manager or however you see fit.

Hope it helps, GL!

 

[ericj159802e9]

Did you do a clean install of Windows when you switched motherboards?

 

[alsi496]

Did you do a clean install of Windows when you switched motherboards?

Yes, I did a completely clean reinstallation of Windows

I have the same board and same BIOS. 9700X + 9070XT.

Have you tried disabling the iGPU in the BIOS? Then clean up the AMD driver for your 9800X3D in Device Manager or however you see fit.

Hope it helps, GL!

I haven't tried forcing the iGPU off in the BIOS yet, thanks for the recommendation! Right now, I'm trying the steps suggested above. If that doesn't work, I'll definitely give this option a shot. Thanks again. Did you have the same issue?

 

[jamarrb103]

Yes, I did a completely clean reinstallation of Windows

I haven't tried forcing the iGPU off in the BIOS yet, thanks for the recommendation! Right now, I'm trying the steps suggested above. If that doesn't work, I'll definitely give this option a shot. Thanks again. Did you have the same issue?

I didn't have the same issue but when I built this PC last year in August I started with RAM tuning. First thing I did was disable iGPU, Wifi and bluetooth, but I see online it's an occurring issue with GPUs when iGPU is enabled. Not sure if it's OS and game having trouble detecting which GPU to use properly in the pipeline. Not an expert on it and haven't A/B tested, but I know that's the one difference from our BIOS set ups.

I hope it works out for you.

 

[ericj159802e9]

Do the tips in this article help in your case: https://github.com/RedDot-3ND7355/MPO-GPU-FIX ?

 

[electrostingz]

Hello,

Are you running the same frame rates as the hold hardware or is the GPU now pushing out a lot higher with that new CPU?
If so then perhaps it's running much hotter than before so have you checked the GPU temperatures?

Get hardware monitor here: https://www.cpuid.com/softwares/hwmonitor.html

GPU
Should stay close / under 80 DegC, anything lower for a high gaming load is good.

GPU Hotspot
90 DegC is good but this should peak at times, if you are mainly seeing peaks to 105 DegC it's a little hot
If the GPU hotspot touches 110 DegC it can shutdown the GPU Core.

GPU Memory
Peak temps of 90 DegC can occur for extremely high workloads but for this card and cooler it should be hoovering around 80 DegC MAX.

EG to show the numbers / areas to monitor.

If you find the temperatures are quite high the GPU may need the thermal paste changed.

 

[alsi496]

Hello,

Are you running the same frame rates as the hold hardware or is the GPU now pushing out a lot higher with that new CPU?
If so then perhaps it's running much hotter than before so have you checked the GPU temperatures?

Get hardware monitor here: https://www.cpuid.com/softwares/hwmonitor.html

GPU
Should stay close / under 80 DegC, anything lower for a high gaming load is good.

GPU Hotspot
90 DegC is good but this should peak at times, if you are mainly seeing peaks to 105 DegC it's a little hot
If the GPU hotspot touches 110 DegC it can shutdown the GPU Core.

GPU Memory
Peak temps of 90 DegC can occur for extremely high workloads but for this card and cooler it should be hoovering around 80 DegC MAX.

EG to show the numbers / areas to monitor.
View attachment 212965
If you find the temperatures are quite high the GPU may need the thermal paste changed.

Thanks. I have already monitored GPU temperatures during gaming and stress tests.
GPU temperature, hotspot and memory temperatures are all within normal ranges, and I have not observed overheating before the nvlddmkm Event ID 153 crashes.
The GPU also passes OCCT GPU/VRAM stress tests without temperature-related errors.
So at the moment overheating does not seem to be the cause, but I will continue logging temperatures with HWiNFO during testing.

 

[alsi496]

Thanks for the suggestion.
Right now I’m testing another diagnostic step first: I set the GPU Core Clock to -150 MHz and disabled the NVIDIA audio driver/device in Device Manager.

I’m going to play and test the system today with these changes.

If the nvlddmkm Event ID 153 / freeze happens again, my next step will be to disable the iGPU in BIOS and retest.

Thanks again for the idea — I’ll update the thread with the resul

I didn't have the same issue but when I built this PC last year in August I started with RAM tuning. First thing I did was disable iGPU, Wifi and bluetooth, but I see online it's an occurring issue with GPUs when iGPU is enabled. Not sure if it's OS and game having trouble detecting which GPU to use properly in the pipeline. Not an expert on it and haven't A/B tested, but I know that's the one difference from our BIOS set ups.

I hope it works out for you.

 

[electrostingz]

If the GPU is fine then your other questions would more likely need some consideration.

1. 1. Are there any known PCIe compatibility issues between MSI B850 motherboards and older PCIe 4.0 NVIDIA GPUs such as RTX 30 series?
2. 2. Does BIOS 7E62v2A64 contain any PCIe/AGESA fixes related to GPU link stability, PCIe power states, or nvlddmkm recovery errors?
3. 3. Could this behavior be related to PCIe link training, ASPM, CPU I/O/SoC behavior, or AGESA firmware?
4. 4. Are there any diagnostic logs or tests you recommend to determine whether the motherboard PCIe subsystem or the GPU is responsible?

1. The AMD AGESA firmware / board does have issues and you will see that the older BIOSes contain some notes.
2. AGESA 1.3.0.1d does contain changes to the Link training behaviour
3. Yes
4. Hwinfo check the PEX counters, all should show 0 excluding the recovery counter <this one changes when the lanes switch state but if the GPU remains under a constant load it shouldn't change much.

Consider downgrading the AGESA firmware, based on the release notes I would be inclined to go with BIOS 7E62v2A53