Build 69913: Repeated ERROR #132/#135 crashes on RTX 5070 Ti — D3D11 and D3D12
Summary
Since playing WoW Forever Beta build 69913, I am experiencing repeated client crashes shortly after entering the world.
I have reproduced the problem using both D3D12 and D3D11, with no addons loaded.
The crashes are not all identical. So far I have captured:
- ERROR #132 ACCESS_VIOLATION in WowB.exe under D3D12
- ERROR #135 ASSERTSAFE in
GxuFontMain.cpp(434)under D3D12 - ERROR #132 ACCESS_VIOLATION inside NVIDIA’s
nvgpucomp64.dllunder D3D11
The D3D11 crash occurred only ~45 seconds after entering the world.
Environment
Client: WoW Classic / Forever Beta 1.60.1, build 69913
Realm: Classic Beta PvE
OS: Windows 11 Pro x64
CPU: AMD Ryzen 7 7700X
GPU: NVIDIA GeForce RTX 5070 Ti, 16 GB
RAM: 32 GB
NVIDIA Driver: 32.0.16.1692
Driver Date: 2026-09-04
Graphics APIs tested: D3D12 and D3D11
No addons were loaded during these crashes and the crash reports show 0 Lua errors.
Reproduction
- Launch WoW Forever Beta build 69913.
- Log into a character.
- Enter the world.
- Play normally.
- Client crashes shortly afterward.
The crashes have occurred within approximately 1–8 minutes of launching/entering the world.
Changing from D3D12 to D3D11 did not resolve the issue.
Crash #1 — D3D12 / ERROR #132
The first crash was:
ERROR #132
ACCESS_VIOLATION
The exception occurred inside WowB.exe.
The client had only been running for approximately 5 minutes.
No addons were loaded and there were no Lua errors.
Graphics API was D3D12.
Crash #2 — D3D12 / ERROR #135
The second crash was different and appears potentially more useful:
ERROR #135
ASSERTSAFE(depthMode == 0 && depthTestMode == GxuDepthTestMode::Disabled && !multisample, ...)
Source:
Engine\Source\Gxu\GxuFontMain.cpp(434)
All addresses in the assertion stack shown in the crash report were inside WowB.exe.
This crash occurred after approximately 8 minutes in the client.
The graphics diagnostics also recorded:
GxApi: D3D12
GxDeviceLostCount: 1
Again, no addons were loaded and there were 0 Lua errors.
Crash #3 — D3D11 / NVIDIA GPU compiler
After the D3D12 crashes, I changed the Graphics API to D3D11 as a troubleshooting step.
The crash report confirms:
GxApi: D3D11
ShaderModel: dx_5_0
The game then crashed approximately 45 seconds after entering the world.
This was another:
ERROR #132
ACCESS_VIOLATION
However, unlike the first crash, the exception occurred inside NVIDIA’s GPU compiler:
nvgpucomp64.dll
The top of the stack is entirely nvgpucomp64.dll, followed by multiple calls through:
nvwgf2umx.dll
The exception was:
ACCESS_VIOLATION - The instruction at "0x00007ffb028c6916" referenced memory at "0x0000000000000048". The memory could not be "read".
This occurred using the RTX 5070 Ti with NVIDIA driver 32.0.16.1692.
Troubleshooting performed
- Removed/renamed the WoW beta Cache folder.
- Tested with no addons loaded.
- Confirmed 0 Lua errors.
- Tested D3D12.
- Tested D3D11.
- Verified adequate free disk/system memory.
- Streaming diagnostics show 0 streaming errors.
- D3D11 still crashes.
- D3D12 still crashes.
Changing graphics APIs therefore does not appear to avoid the problem.
Additional observations
The crashes are occurring on the same beta client:
1.60.1.69913
WowProject: Camelot
Config: Beta
The executable reports that it was built:
Sep 17 2026 17:42:13
The failures are not completely identical, but all three occur very shortly after launching/entering the world.
Of particular interest is that one D3D12 crash triggers an internal Gxu assertion in GxuFontMain.cpp, while the D3D11 crash subsequently occurs inside NVIDIA’s GPU shader/compiler stack.
Given that the issue reproduces with both D3D11 and D3D12 and without addons, this appears potentially related to the rendering/shader path in build 69913 or its interaction with the current NVIDIA driver.
I have retained the complete .txt crash reports and can provide them if useful.
I considered doing a fresh install of a previous NVIDIA driver dates from July 28th but I think I’ll just wait for this to hopefully be patched so I can play