I've reproduced this bug several times. When I create a new "Game - Player Controller" project, the project begins to load in the Editor window but then freezes. Even if I let it sit (for hours) the project does not finish loading. Clicking on the project brings up the Windows prompt to close the app or give it time to respond. Then the project crashes on a GPU failure. I am running an Nvidia GeForce RTX 4060. After the first 2 crashes, I updated my drivers but the problem persists. I am currently running Geforce Game Ready Driver Version 617.42. The applicable logs are below. The Crash Dump identifies the possible shader at issue. This does not happen when I open a new "Game - Empty" project.
Game Log:
2026/10/13 22:23:37.8566 [engine/RenderSystem] Layer GalaxyOverlayVkLayer uses API version 1.1 which is older than the application specified API version of 1.3. May cause issues.
2026/10/13 22:23:37.8573 [engine/RenderSystem] Layer GalaxyOverlayVkLayer_VERBOSE uses API version 1.1 which is older than the application specified API version of 1.3. May cause issues.
2026/10/13 22:23:37.8573 [engine/RenderSystem] Layer GalaxyOverlayVkLayer_DEBUG uses API version 1.1 which is older than the application specified API version of 1.3. May cause issues.
2026/10/13 22:23:37.8573 [engine/RenderSystem] Layer VK_LAYER_EOS_Overlay uses API version 1.2 which is older than the application specified API version of 1.3. May cause issues.
2026/10/13 22:23:37.9391 [engine/RenderSystem] Vulkan Physical Device: NVIDIA GeForce RTX 4060
...
2026/10/13 22:23:59.1956 [Generic] Bootstrap Init Done
2026/10/13 22:23:59.2037 [Generic] Bootstrap Networking...
2026/10/13 22:24:00.3772 [engine/ResourceSystem] Error loading resource file "box_test.vmdl_c" (Error: ERROR_FILEOPEN: File not found) - from scenes/reward-unbox.scene
2026/10/13 22:24:00.4052 [engine/ResourceSystem] Error loading resource file "particles/break/break.cardboard.vpcf_c" (Error: ERROR_FILEOPEN: File not found) - from models/citizen_props/balloonregular01.vmdl
2026/10/13 22:24:00.4213 [engine/ResourceSystem] Error loading resource file "materials/default/default_mask_tga_3a4f4d17.generated.vtex_c" (Error: ERROR_FILEOPEN: File not found) - from materials/generic/trash_a.vmat
2026/10/13 22:24:00.4213 [engine/ResourceSystem] Error loading resource file "models/props/metal_wheely_bin/metal_wheely_bin_compost.vmat_c" (Error: ERROR_FILEOPEN: File not found) - from models/sbox_props/metal_wheely_bin/metal_wheely_bin_body_gib.vmdl
2026/10/13 22:24:00.5676 [engine/ResourceSystem] Error loading resource file "sounds/tools/sfm/denyundo.vsnd_c" (Error: ERROR_FILEOPEN: File not found) - from sounds/player_use_fail.sound
2026/10/13 22:24:03.3289 [engine/RenderSystem] Wrote NVIDIA Aftermath GPU Crash Dump to F:\SteamLibrary\steamapps\common\sbox\aftermath_dumps\core_5.nv-gpudmp
2026/10/13 22:24:03.3289 [engine/RenderSystem] Wrote NVIDIA Aftermath GPU Crash Dump JSON to F:\SteamLibrary\steamapps\common\sbox\aftermath_dumps\core_5.nv-gpudmp.json
2026/10/13 22:24:03.3416 [engine/RenderSystem] ERROR! VK call failed! result = VK_ERROR_DEVICE_LOST (-4) ( vkQueueWaitIdle( VulkanQueue() ) )
2026/10/13 22:24:03.3416 [engine/RenderSystem] ERROR! VK call failed! result = VK_ERROR_DEVICE_LOST (-4) ( vkQueueSubmit2( pQueue, 1, &submitInfo, pFence->Get() ) )
2026/10/13 22:24:03.3416 [engine/RenderSystem] CONCLUSION: the GPU faulted on a memory address (see above). If no resource is named for it,
2026/10/13 22:24:03.3416 [engine/RenderSystem] relaunch with -gpu_diagnostics so address binding can attribute it.
2026/10/13 22:24:03.3416 [engine/RenderSystem] GPU CRASH DIAGNOSTICS:
2026/10/13 22:24:03.3416 [engine/RenderSystem] Failing call: vkQueueSubmit2( pQueue, 1, &submitInfo, pFence->Get() ) -> VK_ERROR_DEVICE_LOST (-4)
2026/10/13 22:24:03.3416 [engine/RenderSystem] Frame: 215, queue: graphics
2026/10/13 22:24:03.3416 [engine/RenderSystem] Adapter: NVIDIA GeForce RTX 4060 (vendor 0x10DE device 0x2882)
2026/10/13 22:24:03.3416 [engine/RenderSystem] Driver: NVIDIA 617.42 (raw 0x9A4A8000), Vulkan 1.4.351
2026/10/13 22:24:03.3416 [engine/RenderSystem] Memory heap 0 (device local): 1195 MiB used of 7189 MiB budget, 7957 MiB heap size
2026/10/13 22:24:03.3416 [engine/RenderSystem] Memory heap 1 (host): 583 MiB used of 48297 MiB budget, 32724 MiB heap size
2026/10/13 22:24:03.3416 [engine/RenderSystem] Memory heap 2 (device local): 14 MiB used of 200 MiB budget, 214 MiB heap size
2026/10/13 22:24:03.3416 [engine/RenderSystem] Diagnostics: device_fault=yes breadcrumbs=no address_binding=no nv_checkpoints=no
2026/10/13 22:24:03.3416 [engine/RenderSystem] [device_faults] (2 addresses, 0 vendor infos)
2026/10/13 22:24:03.3416 [engine/RenderSystem] Address: 0x0000001600E81640 (range 0x0000001600E81640 - 0x0000001600E8164F, precision 0x10) : VK_DEVICE_FAULT_ADDRESS_TYPE_INSTRUCTION_POINTER_FAULT_EXT
2026/10/13 22:24:03.3416 [engine/RenderSystem] no address attribution: VK_EXT_device_address_binding_report not enabled
2026/10/13 22:24:03.3416 [engine/RenderSystem] Address: 0x000000001804C000 (range 0x000000001804C000 - 0x000000001804CFFF, precision 0x1000) : VK_DEVICE_FAULT_ADDRESS_TYPE_READ_INVALID_EXT
2026/10/13 22:24:03.3416 [engine/RenderSystem] no address attribution: VK_EXT_device_address_binding_report not enabled
2026/10/13 22:24:03.3416 [engine/RenderSystem] Vendor binary: 439660 bytes written to F:\SteamLibrary\steamapps\common\sbox\gpu_crash_dumps\device_fault_5.bin
2026/10/13 22:24:03.3416 [engine/RenderSystem] Potentially faulting shaders:
2026/10/13 22:24:03.3416 [engine/RenderSystem] [0] shaders/morph_composite.shader_ps
2026/10/13 22:24:03.3416 [engine/RenderSystem] CONCLUSION: the GPU faulted on a memory address (see above). If no resource is named for it,
2026/10/13 22:24:03.3416 [engine/RenderSystem] relaunch with -gpu_diagnostics so address binding can attribute it.
2026/10/13 22:24:06.1208 [engine/ToolsStallMonitor] Stall detected.
Crash Dump:
[
{
"Base info": {
"Application name": "sbox-dev.exe",
"Dump date": "Tue Oct 13 17:44:54 2026",
"Graphics API": "Vulkan",
"PID": 15088
}
},
{
"Display driver info": {
"Version": "617.14"
}
},
{
"OS info": {
"Build number": 26200,
"Service pack major version": 0,
"Service pack minor version": 0,
"Windows major version": 10,
"Windows minor version": 0,
"Windows product type": 1
}
},
{
"GPU info": {
"Adapter LUID": 73005,
"Adapter name": "AD107-A"
}
},
{
"Page fault info": {
"Access Type": "Read",
"Client": "Graphics Processing Cluster",
"Engine": "Graphics",
"Fault Type": "Failed to translate the virtual address.",
"GPU virtual address": 402964480
}
},
{
"Device info": {
"Adapter reset occurred": false,
"Device state": "Error_DMA_PageFault",
"Engine reset occurred": true
}
},
{
"Shader infos": {
"Info": {
"Shader hash": 7518473093209836524,
"Shader name": "fragment_02 [shaders_morph_composite.shader_ps]",
"Shader size": 1024,
"Shader type": "Fragment"
}
}
},
{
"ApplicationName": "core"
},
{
"ApplicationVersion": "2"
},
{
"Active Warps": [
{
"GPU PC Address": "fragment_02 [shaders_morph_composite.shader_ps] @ 0x000001c0",
"Shader mapping": null,
"Warp count": 15
}
]
},
{
"Faulted Warps": [
{
"Fault Description": "A shader instruction caused an MMU fault when accessing memory.\nThis can be caused by shader bugs and binding setup issues, or possibly by a shader compiler bug or shader microcode corruption.",
"Fault Name": "MMU Fault Error",
"Shader GPU PC Address": "fragment_02 [shaders_morph_composite.shader_ps] @ 0x000000c0",
"Shader mapping": null
}
]
},
{
"Fingerprint": {
"Hash": 17173432755012887482,
"Quality": 20
}
}
]
edited:
Did someone just ninja fix this or something? I repro'ed this thing 5 separate times. I was doing it again to make sure I didn't miss anything in the report and now it works. I haven't made any system changes.
edited:
I take it back, it's still happening.
