[Crash][Stability] Silent process termination within 2 minutes on RX 9070 since Patch v42.30 (Hardware Specific Regression). No crash reporter, logs, or Event Viewer entries.
Please select what you are reporting on:
Unreal Editor for Fortnite
What Type of Bug are you experiencing?
Stability
Steps to Reproduce
Launch Fortnite via UEFN or Epic Games Launcher.
Idle in the lobby or enter any match/island.
Wait approximately 1 to 2 minutes.
Expected Result
The game client should remain open, stable, and active indefinitely—exactly as it did on this identical hardware and driver version prior to Patch v42.30.
Observed Result
The game silently terminates instantly to the desktop within 60–120 seconds. No Unreal Crash Reporter appears, no logs are generated in AppData/Saved/Logs, and no Application/System errors are logged in Windows Event Viewer.
NOT ACCOUNT-BASED: Same account works flawlessly on PS5.
NOT SYSTEM CORRUPTION: SFC/DISM clean, BIOS updated, full game/launcher/EAC clean reinstalls did nothing.
HARDWARE SPECIFIC PROOF: Physically swapped out the RX 9070 for an older RTX 3050 in this exact system. The game ran perfectly for 20+ minutes. Re-inserting the RX 9070 immediately brings back the 2-minute silent crash.
NOT A TDR TIMEOUT: Increasing the Windows TDR registry delay yields the exact same silent process termination timeframe with no driver recovery alerts.
RENDERING MODE PROOF: Tested across both DirectX 12 and standard Performance Mode at the lowest graphical scalability settings with a strict 30 FPS cap. The silent crash still occurs at the exact same ~2-minute mark regardless of selected rendering mode or graphic workload.
Platform(s)
PC / Windows 11 (AMD Radeon RX 9070 specific regression)
Additional Notes
ADDITIONAL TROUBLESHOOTING PERFORMED (ALL FAILED TO RESOLVE):
Disabled AMD Adrenalin Overlay and all software optimizations (Radeon Boost, Anti-Lag, etc.) for Fortnite specifically.
Fresh installed Epic Games Launcher & Fortnite (twice).
Cleared local AppData and shader caches.
Repaired Easy Anti-Cheat (EAC) binaries.
Performed SFC /scannow and DISM system repairs.
Updated motherboard BIOS to the latest version.
Passed full CPU stability stress tests (ruling out undervolting/PBO instability which commonly causes silent UE engine drops).
CONCLUSION:
This issue is a consistent regression introduced in the v42.30 patch cycle that triggers a hard process kill exclusively when the RX 9070 is active in the system. Hoping this data helps the PC stability team look into what changed for this hardware configuration in the latest build.
i have the same problem with the rx 9070 and the r7 5700x, i haven’t overclocked it, and all games works fine, it’s only fortnite. it’s shutdown 20 seconds after loading
Hi, quick clarification — I’m new here and honestly thought UEFN stood for “Unreal Engine Fortnite” rather than “Unreal Editor for Fortnite,” so I picked that option thinking it was the general Fortnite tag. My crash is in the game client (launched via Epic Games Launcher), not the editor. I don’t have UEFN installed, so I can’t provide those logs. I’ve attached my game client logs from %localappdata%\FortniteGame\Saved\Logs instead. Sorry for the mix-up!
First off — the in-game crash reporter suggestion. I get it, but the game CTDs within 30 seconds of launch. There’s no UI to interact with at that point, the process is already gone. Sure, in my case I could have just turned on my PlayStation 5 and done it there, but after a long day of trying every fix I could think of, that’s not exactly the first thing that comes to mind. And for anyone who doesn’t have a second platform to fall back on — where would they go?
Adrenalin overlay and all per-game optimizations (Radeon Boost, Anti-Lag, etc.) already disabled for Fortnite — mentioned in my original post. No third-party overlays or monitoring tools.
Installed UEFN to test — it doesn’t even launch. No entry in Event Viewer (Application or System), so the process is being terminated at a level below Windows Error Reporting.
Yeah I wish it was that easy but sadly it isn’t. Already tested — both via in-game settings (Performance Mode / lowest scalability / 30 FPS cap) and via the -d3d11 launch option. Same result either way: silent CTD within ~30s.
Tried it, still silent CTD within ~30s. So it’s not a Performance Mode vs. full DX12 issue — it crashes in both. The RX 9070 runs everything else fine, so something in the v42.30 patch just doesn’t get along with it.
That folder contains crash dumps that sometimes get generated for these kinds of errors. If you have any named FortniteClient-Win64-Shipping, could you share that with me
You’re welcome to share the crash dump privately if you wish - the link will start a direct message to me
Separately the team has told me to have you try again as some changes have been made
No Fortnite-related files in the crashDumps. Latest update still CTDs within ~30s. Retried -sm6 and -d3d11 on the latest update as well, no difference.
Update:
I tried a few more things on the latest patch today:
Disabled Hardware-Accelerated GPU Scheduling (HAGS), which didn’t fix it (reenabled it afterwards).
Tried offsetting the GPU clock by -200MHz, which didn’t work either (undid that as well).
Jumped from Adrenalin 26.10.1 back to 26.6.1, and then updated to 26.10.2—still CTDs within ~30 seconds.
I ran the logs through an automated analyzer to dig deeper into the crash, and it flagged a hard memory access violation right at process termination:
Error:exit status 0xC0000005 (Access Violation)
Decimal Code:3221225477
It looks like the client gets past EAC_EOS, starts loading, and then immediately hits a bad memory read/write address on the RX 9070 architecture. Hope this helps the team pin down the issue!
