Unable to play DirectX 12 native games (ROG MAXIMUS Z790 HERO)

[msalisbu]

Hello. This winter I built a new gaming PC (parts below) and since putting it together, every DX12 title will crash with some combination of an our of memory error, or unable to read a memory address. (E_OUTOFMEMORY PipelineDx12Graphics.cpp)

I've tested different video cards, clean installs of both windows 10 and 11, updated the bios, and many more troubleshooting steps with no working fix. Here's a list of things I've tried:

Reinstall windows 11
Install Windows 10
Update and roll back video drivers with DDU
SFC scan
Memtest
BIOS update (currently on 0816)
Generating new cryptography reg key for machine UUID
Tested with different video card
Disabled all game overlays
Modified pagefile settings (up to 100GB!)
Installed .Net framework/repair
Installed/repaired C++ redistributable
XMP enable/disable.
The system is not overclocked at all.

Specs:
Asus ROG Maximus Z790 Hero - BIOS 0816
Intel i9 139000K
64GB G.Skill Trident Z5 F5-6000J3238F16GX2-TZ5RK
WD_BLACK SN850X 2TB (Latest firmware)
RTX 3090 (531.29)

I'm starting to think the CPU or MB is a lemon... but if anyone has any idea or tips that would be great. Really at a loss.

[Jiaszzz_ROG]

Hello, @msalisbu 

We are currently confirming the situation you have encountered with relevant departments.
May I ask if you could share a video of a DirectX 12 game crashing?
Please try to install only 1 memory module in the DIMM_A2 slot to see if the problem still exists.

Thank you.

[Silent_Scone]

Although it might seem a little misleading, what others have said should point you on the right path.
These DX errors can occur when the CPU or RAM is unstable. If not done so already, you need to establish a baseline for stability with no overclocking, which includes the exclusion of any XMP or memory OC profiles.

As @Jiaszzz_ROG has mentioned, try with one module inserted in A2. You can test both modules individually using Memtest86 from the UEFI at complete UEFI defaults.

The Q-CODE readouts you've provided is the LED reporting the CPU package temp, these are not stop codes. These only appear at POST when the system is powered on.

Other avenues you can try are decreasing the GPU Core and Memory offset by -100 and see if the issue persists.

[Jiaszzz_ROG]

Hello, @msalisbu .

Additionally, please help us with the following steps for further testing:
- Install only one memory module and test in both DIMM_A2 and B2 slots.
If possible, please try other memory modules from the QVL list for cross-testing.
- Remove the graphics card and test using the onboard graphics output.

If the above steps are not able to resolve the issue, kindly provide us with more detailed information to help us reproduce and test the scenario, including:
- The brand and full model name of the GPU.
If it is an ASUS GPU, please use the driver from the official ASUS website to rule out any compatibility issues.
- The name of the DirectX 12 game and the frequency of the crashes.
If possible, please provide a video demonstrating the issue.
- Please assist with executing the 3D Mark Time Spy test and provide us with the evaluation results.

Thank you.