ARC Raiders Crashes: The Complete PC Master Fix Guide (DLSS, Black Screen, Freezes, Gear Loss)
๐ฏ Quick Answer
ARC Raiders is built on Unreal Engine 4 and pushes hard on modern rendering tech like DLSS 3.5 and DirectX 12. This combo, plus aggressive anti-cheat, creates a perfect storm for crashes that feel ran...
Context & Background
ARC Raiders is built on Unreal Engine 4 and pushes hard on modern rendering tech like DLSS 3.5 and DirectX 12. This combo, plus aggressive anti-cheat, creates a perfect storm for crashes that feel random but usually aren't. I've personally rebuilt my config a dozen times chasing these down. The core issue is almost always a conflict between the game's rendering pipeline, your GPU driver's state, and Windows' memory management. This guide consolidates every major crash variant into one place. You're not doing anything wrong; the game is just brittle with certain system states.
TL;DR: The 5-Minute Crash Blitz
If you're crashing and just want to try the big hitters fast, do these in order:
- Clean Install GPU Drivers: Use DDU in Safe Mode, install the latest stable (not beta) driver from NVIDIA/AMD/Intel.
- Clear Shader Caches: Run Windows Disk Cleanup, check "DirectX Shader Cache." Also delete the game's own shader cache in
%LOCALAPPDATA%\ARC\Saved\ShaderCache. - Disable All Overlays: Turn off Discord, Steam, Xbox Game Bar, NVIDIA GeForce Experience, and MSI Afterburner/RivaTuner overlays.
- Cap Your Frames: Use NVIDIA Control Panel or AMD Adrenalin to set a global frame cap 3-5 FPS below your monitor's refresh rate.
- Verify Game Files: In Steam, right-click ARC Raiders > Properties > Installed Files > "Verify integrity of game files."
If you're still crashing after that, your specific error is in the table below. Find it and jump to the deep fix.
Error Code & Symptom Reference
| What You See (Error/Symptom) | Most Likely Cause | Jump to Fix |
|---|---|---|
| Shadow artifacts (blocky, flickering) then crash | Corrupted DirectX Shader Cache interfering with DLSS | DLSS & Shadow Artifact Fix |
| Visual artifacts (texture corruption, distortion) then GPU crash | Unstable GPU memory (factory OC or manual) / Driver conflict | GPU Crash with Visual Artifacts |
| Full system black screen, audio cut, requires reboot | Power supply transient spike / Driver TDR failure | Black Screen Crash |
| Game freezes on mouse movement, GPU usage drops to 0% | HAGS conflict / Overlay input hook conflict | Mouse-Movement Freeze |
| Crash โ Reconnect loop โ "Connection to server lost" | Session state corruption / Firewall blocking reconnection | Reconnection Loop Crash |
| Crash results in permanent loss of equipped gear | Unclean client shutdown (any crash type) | Preventing Gear Loss |
Solutions by Impact: Start Here
These are universal fixes. Apply them first, regardless of your specific crash.
The Mandatory Foundation
- Clean GPU Driver Installation (Using DDU): This is non-negotiable.
- Download Display Driver Uninstaller (DDU).
- Download the latest Game Ready Driver (NVIDIA) or WHQL Driver (AMD) from the manufacturer's site. Do not use the Windows Update driver.
- Boot into Safe Mode (hold Shift while clicking Restart in Windows).
- Run DDU. Select your GPU type (NVIDIA/AMD/Intel) and click "Clean and restart."
- After reboot, install the driver you downloaded. Choose "Custom Installation" and check "Perform a clean install."
- Nuke the Shader Caches:
- Press
Win + R, typecleanmgr, and hit Enter. - Select your system drive (usually C:).
- Check "DirectX Shader Cache" and click OK. Delete it.
- Navigate to
%LOCALAPPDATA%\ARC\Saved\and delete the entireShaderCachefolder. The game will rebuild it.
- Murder All Overlays: Overlays are public enemy #1 for stability.
- Discord: Settings > Overlay > Disable "Enable in-game overlay."
- Steam: Settings > In-Game > Uncheck "Enable the Steam Overlay."
- Xbox Game Bar: Windows Settings > Gaming > Xbox Game Bar > Turn it Off.
- NVIDIA GeForce Experience: Open GFE, click the Settings cog, toggle off "In-Game Overlay."
- MSI Afterburner/RivaTuner: Exit them completely from the system tray.
In-Game Settings for Stability
- Upscaling: If you crash with DLSS, switch to FSR 3 temporarily. It's more stable in the current build. If you crash on AMD with FSR, try XeSS or TAAU.
- Frame Generation: Disable DLSS Frame Generation or AMD Fluid Motion Frames. This is a major crash source.
- Graphics API: Stick with DirectX 12. Do not use Vulkan if present; it's less stable.
- Frame Cap: Enable VSync in-game or use NVIDIA Reflex + Boost. Alternatively, cap frames to 60/120/144 in NVIDIA Control Panel or AMD Adrenalin globally.
Deep Dive Diagnostics: Fix Your Specific Crash
If the foundation fixes didn't stop your specific crash, target it here.
DLSS & Shadow Artifact Crash
When this applies: You see blocky, pixelated, or flickering shadows, especially when standing still. Crash follows within a minute. Faulting module in logs often mentions nvngx_dlss.dll.
- Roll Back or Update DLSS DLL: The game ships with a specific DLSS version. Sometimes updating it helps, sometimes rolling back does.
- Go to
\Steam\steamapps\common\ARC Raiders\ARC\Binaries\Win64\. - Look for
nvngx_dlss.dll. Rename it tonvngx_dlss.dll.backup. - Download a different version of
nvngx_dlss.dllfrom a trusted DLL repository. Try version 3.5.10 or 3.7.0. - Place the new DLL in the same Win64 folder. Launch the game.
- Disable Ray Traced Shadows: If the above fails, this is the culprit. In Graphics settings, set Shadow Quality to High (not Ultra). Ultra enables ray-traced shadows which conflict with DLSS in this engine.
- Disable Resizable BAR: If you have an NVIDIA GPU, go into NVIDIA Control Panel > Manage 3D Settings > Program Settings > Add ARC Raiders. Find "Resizable BAR" and set it to Off. This can cause memory mapping conflicts with DLSS 3.5.
GPU Crash with Visual Artifacts
When this applies: Textures stretch, colors freak out, polygons tear, then the game hard crashes or the driver restarts.
- Undo ALL Overclocks: This includes factory overclocks on "OC" model cards.
- Use MSI Afterburner. Set Core Clock and Memory Clock sliders to 0. Click Apply.
- Set Power Limit to 100% (not higher).
- For NVIDIA 30/40 series, this is critical. GDDR6X runs hot and can error at stock speeds. If this fixes it, your card's memory is unstable. Consider a slight underclock of -100MHz to Memory Clock for permanent stability.
- Increase GPU Timeout Detection (TDR) Delay: This gives the GPU more time to recover before Windows panics and crashes the game.
- Press
Win + R, typeregedit, hit Enter. - Navigate to
HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\GraphicsDrivers. - Right-click in the right pane > New > DWORD (32-bit) Value. Name it
TdrDelay. - Double-click it, set "Base" to Decimal, and set "Value data" to
8(seconds). Reboot.
- Monitor VRAM Temperature: Use HWiNFO64 and log sensors during gameplay. If your GDDR6X memory junction temp exceeds 100-105ยฐC, it's throttling and causing errors. Improve case airflow or consider repadding.
Black Screen Crash
When this applies: Screen goes completely black, audio dies, PC is unresponsive and needs a hard reboot. Common on high-end AMD 6000/7000 series and high-power NVIDIA cards.
- Address Power Transients: This is likely a power spike tripping your PSU's protection.
- Undervolt Your GPU: This is the most effective fix. In AMD Adrenalin, go to Performance > Tuning > Enable Custom > Advanced Control > Reduce "Max Voltage" by 5-10%. For NVIDIA, use Afterburner's Voltage-Frequency curve (Ctrl+F). A mild undervolt reduces peak power draw dramatically.
- Use Two Separate PCIe Power Cables: If you have a high-wattage GPU, do NOT use a single daisy-chained cable from the PSU. Use two separate cables.
- Limit FPS: A hard FPS cap is the simplest way to reduce peak power draw.
- Disable Windows Hardware-Accelerated GPU Scheduling (HAGS):
- Go to Windows Settings > System > Display > Graphics Settings.
- Turn Off "Hardware-accelerated GPU scheduling." Reboot.
- Test Your PSU: If you have a spare PSU, test with it. If not, and you're near your PSU's rated wattage (especially on 850W or less with a 4090/7900 XTX), this is highly suspect.
Mouse-Movement Freeze (0% GPU)
When this applies: Moving the mouse instantly freezes the game. GPU usage in monitoring software drops to 0%. Must force close via Task Manager.
- Disable HAGS (Again): This is the prime suspect. Follow the step above to turn it Off.
- Lower Mouse Polling Rate: High polling rates (1000Hz+) can cause engine deadlocks.
- Use your mouse software (Logitech G Hub, Razer Synapse, etc.) to set the polling rate to 500Hz. Test. If stable, you can try 1000Hz again, but 500Hz is often the fix.
- Run Game as Administrator: Right-click the ARC Raiders shortcut or .exe > Properties > Compatibility > Check "Run this program as an administrator." This can prevent input hook permission conflicts with anti-cheat.
Reconnection Loop Crash
When this applies: Game crashes, then on restart you get an infinite loop of "Reconnecting..." and "Connection to the server was lost." All gear from that session is gone.
- Flush DNS & Reset Network: This clears local network caches that can confuse the reconnect handshake.
- Open Command Prompt as Admin. Run these commands:
ipconfig /flushdns
netsh winsock reset
netsh int ip reset
- Reboot your PC and router.
- Configure Firewall: Ensure ARC Raiders and Easy Anti-Cheat have full permissions.
- Press
Win + R, typefirewall.cpl, hit Enter. - Click "Allow an app or feature through Windows Defender Firewall."
- Click "Change settings," find ARC Raiders and Easy Anti-Cheat (EAC), ensure both Private and Public boxes are checked. If they're missing, add them manually by browsing to the game's
Binaries\Win64folder.
- This is often a server-side bug. If the above fails, you've done all you can. The session is corrupted on the server. Report it through official channels.
Preventing Gear Loss from Crashes
This applies to ANY crash that causes an unclean exit. The server only preserves your gear if it receives a proper "player disconnected" signal.
- The Only Mitigation: If you feel a crash coming (artifacts, stuttering), do not Alt+F4 or kill the process. Let the game crash on its own. Sometimes this gives the client a fraction of a second to send the disconnect signal. It's not reliable, but it's the only "trick."
- Don't Equire Your Best Gear During Unstable Play: If you're testing fixes or know your setup is crash-prone, run with gear you're willing to lose.
Platform-Specific & Driver Notes
- NVIDIA Drivers: Version 551.86 or later is recommended for ARC Raiders. Avoid the 5xx.xx branch. If on a 40-series card, disable Frame Generation as a primary troubleshooting step.
- AMD Drivers: Use the latest WHQL optional driver. In Adrenalin, set "Graphics Profile" to "Standard" (not Gaming or Esports). Disable "AMD SmartAccess Video" if present.
- Intel Arc Drivers: Must be the absolute latest. Stability improves with each driver release.
- Easy Anti-Cheat: Repair it. Go to the game's install folder, find the
EasyAntiCheatfolder, runEasyAntiCheat_Setup.exeand select "Repair."
Escalation Path: When Nothing Works
If you've executed every relevant fix above and still crash, the problem is deeper.
- Analyze the Crash Log:
- Navigate to
%LOCALAPPDATA%\ARC\Saved\Crashes. - Open the most recent folder, then the
Report.werfile in Notepad. - Search for
FaultingModuleName. This tells you which .dll caused the crash (e.g.,nvwgf2umx.dll= NVIDIA driver,d3d12.dll= DirectX 12). - This points you to the culprit: GPU driver, DirectX, or a specific game module.
- Test RAM Stability: Use TestMem5 with the
Extreme1@anta777profile or OCCT Memory Test for 1 hour. A single error means unstable RAM, which will crash UE4 games.
- Fresh Windows User Profile: Create a new local Windows user account. Log into it, install only the GPU driver and Steam/ARC Raiders. Test. If it works, your main user profile is corrupted.
- Ultimate Nuclear Option: If you have a second drive, perform a clean Windows installation on it. Install only drivers, Steam, and ARC Raiders. This isolates software conflict 100%. If it crashes here, it's a hardware problem.
Your last resort is hardware: PSU under-specification, GPU memory instability (RMA), or motherboard PCIe lane instability. Use the diagnostics above to gather evidence before contacting support or considering replacements.