What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
“Unreal Engine is exiting due to D3D device being lost” is a symptom, not a diagnosis. Unreal has lost access to its Direct3D graphics device, often because Windows reset the graphics driver. The cause may be exhausted GPU memory, a GPU timeout, a driver or engine bug, operating-system instability, or hardware trouble. Work through the low-risk checks first; changing Windows’ timeout settings is a narrow workaround, not a universal fix.
This guidance applies to Unreal Engine 4 and 5 on Windows. The steps can also help players whose Unreal Engine games display related errors such as DXGI_ERROR_DEVICE_REMOVED or DXGI_ERROR_DEVICE_HUNG.
What the D3D device-lost error means
Unreal sends graphics work to the GPU through Direct3D, commonly called D3D. Windows monitors whether the GPU and its driver are responding. If an operation appears stuck or the driver stops responding, Windows may reset the graphics driver through Timeout Detection and Recovery (TDR). Unreal can then find that its former D3D device is no longer usable and exit.
Epic describes Windows’ default GPU timeout as approximately two seconds, but that does not mean every crash has the same cause or follows an identical timing threshold. A device-lost message does not, by itself, mean the physical graphics card has failed. Epic lists possible causes including GPU out-of-memory conditions, long-running GPU work, engine or driver bugs, operating-system issues, and hardware faults. Epic’s GPU crash guide explains these causes and the diagnostic options.
#1 Best Overall
- Powered by Radeon RX 9070 XT
- WINDFORCE Cooling System
- Hawk Fan
- Server-grade Thermal Conductive Gel
- RGB Lighting
Record the error and when it happens
Before changing settings, note the full message, any error code, and what the computer was doing immediately before the crash. Messages can include DXGI_ERROR_DEVICE_REMOVED, DXGI_ERROR_DEVICE_HUNG, or a GPU-crash message ending in “D3D Hung.” The stage matters: a failure during a large scene render suggests a different investigation than one that happens immediately at launch.
- Does it happen on launch, while opening a project, during Play in Editor, or while compiling shaders?
- Does it follow one asset, scene, lighting bake, render, or packaging step?
- Does it occur in one project or game, or in several unrelated GPU-heavy applications?
- Did it begin after a driver, engine, Windows, or project change?
Try the low-risk fixes first
- Restart Windows. After a driver reset, restart the computer rather than repeatedly relaunching Unreal. This gives the driver and graphics state a clean start.
- Return hardware settings to stock. Temporarily turn off GPU and VRAM overclocks, undervolts, custom power or fan profiles, and vendor performance modes. Close tuning utilities such as MSI Afterburner while testing. If the crash stops, that is useful evidence of instability, not proof that the GPU is defective.
- Update or roll back the graphics driver. Use the GPU maker’s own download page: NVIDIA, AMD, or Intel. If the problem started directly after an update, test the previous known-good driver; the newest version is not automatically the best for every project and system.
- Close competing GPU software. Shut down other games and 3D applications, recording or streaming tools, overlays, hardware-accelerated browser windows, and monitoring utilities. These can use GPU memory or interact with rendering.
- On a laptop, check GPU selection and power. Confirm that Unreal or the game is assigned to the intended high-performance GPU in Windows Graphics settings or the GPU vendor’s control panel. Test while connected to AC power if the crash occurs only on battery. Do not disable the integrated GPU as a routine fix.
Check whether GPU memory or workload is the trigger
While the project is running, open Windows Task Manager → Performance → GPU. Observe dedicated and shared GPU memory, especially just before the crash. If dedicated memory is near its available limit, close other GPU-heavy applications and reduce the workload to see whether the failure moves or disappears. System memory pressure and shared GPU memory can complicate the picture, so do not assume that adding RAM alone will solve a dedicated VRAM shortage.
- Lower texture or mesh resolution and reduce scene complexity.
- Reduce the editor viewport resolution or screen percentage; close extra viewports.
- Use culling to reduce the objects being drawn.
- As an isolation test, temporarily disable demanding features such as Lumen, Nanite, virtual shadow maps, ray tracing, or path tracing.
These changes help identify memory pressure or an expensive rendering path. A feature that is disabled only to stop a test crash is not necessarily the underlying fix.
Compare DX11 and DX12
Testing a different rendering hardware interface (RHI) can reveal whether the crash is associated with a particular API, feature path, driver interaction, or project configuration. For an editor installation where the executable is named UnrealEditor.exe, the test commands typically look like this; adjust the executable and project paths for your installation:
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsRank #2
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Powered by GeForce RTX 5070 Ti
- Integrated with 16GB GDDR7 256bit memory interface
- PCIe 5.0
- WINDFORCE cooling system
UnrealEditor.exe "C:PathProjectProject.uproject" -dx11
UnrealEditor.exe "C:PathProjectProject.uproject" -dx12If the editor cannot launch, add -dx11 or -dx12 after the closing quotation mark in the Unreal Editor shortcut’s Target field. Test one argument at a time.
- If only one RHI crashes, compare its logs and the features enabled in that mode; the result points to a useful diagnostic difference, not proof that the API is globally broken.
- DX11 may be a workable compatibility test, but it does not provide every DX12-only feature. Some modern rendering features, including hardware ray tracing, require DX12.
- Neither API is universally more stable across all GPUs, drivers, engine versions, and projects. Epic recommends RHI switching as an investigation technique, not a permanent fix for every case.
Isolate a project, asset, or rendering feature
Make one change at a time so you can tell what affects the crash. Toggle a suspect rendering pass, hide recently added objects, or disable a heavy Niagara system. Test the same steps in a blank project and, if available, an older backup of the affected project. If a particular asset or recent change consistently brings the failure back, investigate its materials, shaders, effects, or rendering path rather than changing system-wide settings first.
If the crash began after an engine or driver change and appears limited to one project, cached shader or derived data could be involved. Back up the project before clearing any project cache. Cache locations and safe cleanup practices vary by Unreal version and project setup; deleting all Unreal cache data is not a necessary first step.
Ray-tracing workloads
Epic notes that hardware ray tracing can be costly and may contribute to TDR events. For supported ray-tracing global illumination and reflection paths, these settings can split work into tiles when their values are greater than zero:
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Rank #3
- AI Performance: 767 AI TOPS
- OC mode: 2632 MHz (OC mode)/ 2602 MHz (Default mode)
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Axial-tech fan design features a smaller fan hub that facilitates longer blades and a barrier ring that increases downward air pressure
- A 2.5-slot design maximizes compatibility and cooling efficiency for superior performance in small chassis
r.RayTracing.GlobalIllumination.RenderTileSize
r.RayTracing.Reflections.RenderTileSizeThese settings are version- and feature-dependent, not a general switch for every ray-tracing path. First test by disabling the relevant feature or lowering the workload; use the commands only when they apply to your project.
Collect useful GPU crash evidence
For a reproducible developer crash, run Epic’s diagnostic arguments separately, not in the same test:
-gpucrashdebugging
-d3ddebugEpic says -gpucrashdebugging collects GPU progress and state, while -d3ddebug provides Direct3D pipeline information. Resulting logs are saved in the project’s Saved/Logs directory. The same guide lists advanced investigation options:
-onethread
-forcerhibypass
r.RDG.Debug=1
r.RDG.ImmediateMode=1These can help investigate threading, timing, or render-graph behavior, but they change execution conditions. In particular, Epic warns that immediate mode can produce more meaningful call stacks while also creating misleading clues. Treat results as diagnostic evidence, not proof in isolation. See Epic’s GPU crash debugging guidance.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Rank #4
- Powered by the NVIDIA Blackwell architecture and DLSS 4. System Requirements: Minimum 850W PSU with 16-pin 12V-2x6 (12VHPWR) connector required. Verify before purchasing.
- Military-grade components deliver rock-solid power and longer lifespan for ultimate durability. Compatibility: 348mm (13.7") length, 3.6 slots, 4.3 lbs. Confirm case clearance and slot spacing. GPU bracket included.
- Protective PCB coating helps protect against short circuits caused by moisture, dust, or debris
- 3.6-slot design with massive fin array optimized for airflow from three Axial-tech fans
- Phase-change GPU thermal pad helps ensure optimal thermal performance and longevity, outlasting traditional thermal paste for graphics cards under heavy loads
When a TDR registry change may be relevant
Consider the timeout workaround only after the safer tests, and only when the failure plausibly involves a long GPU operation—for example, a heavy render that causes a brief system freeze or visible driver reset, and which becomes less likely when you lower the workload. A crash at every launch, in an empty project, or across unrelated apps at low GPU usage is less convincingly explained by a timeout alone.
Warning: Editing the Windows registry carries risk. Create a system restore point or other backup and record the existing values before making changes. Epic’s documented workaround sets both values to 60 decimal in the GraphicsDrivers key:
- Open Windows search, find Run, and enter
regedit. - Navigate to
ComputerHKEY_LOCAL_MACHINESYSTEMCurrentControlSetControlGraphicsDrivers. - In the
GraphicsDriversfolder itself, create or edit a DWORD (32-bit) Value namedTdrDelay. Choose Decimal as the base and enter60. - Create or edit a second DWORD (32-bit) Value named
TdrDdiDelay, also with base Decimal and value60. - Close Registry Editor and restart Windows.
This tells Windows to wait longer before deciding that the GPU process has taken too long; it may let a valid but expensive operation finish. It does not repair a bad driver, defective render pass, VRAM shortage, overheating, unstable overclock, failing hardware, or other cause of device removal. To undo the change, restore the previous values you recorded, or delete the values if you created them, then restart Windows. Follow Epic’s TDR instructions if you need the documented procedure.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Use the crash pattern to choose the next test
| Observed pattern | Possible direction | Next test |
|---|---|---|
| Crash after loading a very large scene | VRAM pressure or expensive GPU work | Monitor GPU memory; reduce textures, viewport resolution, and scene complexity. |
| Crash only with ray tracing or path tracing | Timeout, driver interaction, or feature-specific engine issue | Disable the feature, reduce render load, and compare supported RHIs. |
| Crash only in DX12 or only in DX11 | API-specific feature, driver, or project path | Test the other RHI if project features permit and compare logs. |
| Started after a driver update | Driver regression or damaged installation | Clean-install the driver or test a known-good earlier version. |
| Appears in multiple Unreal games or other GPU-heavy apps | System driver, Windows, power, thermals, or hardware | Return hardware to stock; check temperatures and test another GPU workload. |
| Limited to one project, asset, or scene | Project, shader, material, effect, or render-pass problem | Test a blank project and revert or hide recent changes. |
| Only happens on a laptop when unplugged | Power profile, hybrid-GPU selection, or reduced thermal/power budget | Test on AC and verify that the intended GPU is selected. |
| Occurs only after prolonged load | Heat, power delivery, overclock, or driver instability | Check temperatures, stock settings, and system logs. |
| Occurs immediately on startup | Driver/API initialization, wrong GPU, unsupported configuration, or damaged installation | Verify GPU selection, test another RHI, and consider repairing the affected installation. |
These patterns are clues rather than guarantees. Epic’s historical issue UE-42280 covered varied “D3D device being lost” failures in older UE4 versions and was marked fixed in 2017; Epic noted that related cases continued under other tracking. It is not a diagnosis for every current UE5 crash.
Best Value
- Next-Gen Intel Arc Graphics: Powered by Intel Arc A580 GPU with Intel Xe HPG microarchitecture, featuring 384 XMX engines for enhanced AI acceleration and content creation.
- High-Performance Memory: 8GB GDDR6 on a 256-bit interface running at 16 Gbps, delivering excellent bandwidth for 1440p gaming and creative workloads.
- Factory Overclocked: Engine clock set at 2000 MHz out of the box, providing optimized performance for smooth gameplay and multimedia tasks.
- Advanced Dual-Fan Cooling: Features a dual-fan design with striped axial fans and an ultra-fit heatpipe for efficient thermal management. 0dB Silent Cooling stops fans completely at low temperatures for silent operation.
- Durable Construction: Includes a stylish metal backplate for enhanced PCB rigidity and a premium aesthetic, backed by ASRock's Super Alloy components for long-term reliability.
Know when to investigate the wider PC
If several unrelated GPU applications crash, or you see artifacts, black screens, repeated display-driver resets, or instability under sustained load, stop treating the problem as Unreal-only. Check GPU and VRAM temperatures, power connections and supply, laptop cooling, and any factory or user overclock. Multiple monitors, HDR, high refresh rates, overlays, and hardware-accelerated desktop apps can also alter GPU memory and compositing load; test with unnecessary displays and overlays disabled, but do not treat that as a guaranteed Unreal fix.
Update Windows through its normal update channel rather than seeking a separate DirectX download. Avoid random DLL sites or replacing system graphics DLLs. Reinstalling Unreal is more relevant if one engine installation fails, its files may be corrupted, binaries were modified, or the launcher’s verification or repair finds a problem; it is low-value if the same crash occurs across several projects and games.
When to contact Epic or the hardware maker
Escalate when crashes persist at stock settings across driver versions, affect multiple applications, produce artifacts or repeated resets, or coincide with abnormal temperatures. For an Unreal Engine support request, include:
- Unreal Engine and Windows versions
- GPU model and driver version, plus CPU and installed RAM
- The full error text and code, and the RHI in use
- Steps that reliably reproduce the crash and whether it follows a project or asset
- Relevant project logs from
Saved/Logs - Results from DX11/DX12 comparisons, lower-load tests, and stock hardware settings
Unreal developers can use Epic’s Unreal Engine support resources. For suspected driver or hardware faults, contact the GPU manufacturer or computer maker with the same reproduction details and any relevant system logs.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
