Stellar Blade
Same issue here on Arch Linux. I have tried Proton-GE 10.4 and also: https://github.com/arnxxau/proton-dualsense/releases/tag/9.0-4 from this thread: https://github.com/ValveSoftware/Proton/issues/5900 but no luck.
These builds should all include the patches needed to get haptics from the DualSense and do work in other games like The Last Of US Part 2 but no luck here. I even updated the firmware of my DualSense due to the game complaining about it but that didn't solve it either.
Must be something specific with this game.
+1 from my side as well; I’m on Fedora using Proton-GE 10.4, but the same goes for Proton Experimental.
In addition, when I suspend the PC while the game is running and then resume (good to quickly pick back up where one left of) the Controller is dead as far as the game is concerned - the controller is on and connected, but the game does not respond to it anymore. Closing the game and starting it back up, the controls works as expected again.
Seems like the game looses the Controller somewhere along suspend and wake.
A few months ago there were a dualsense-haptics-wip branch/beta for proton experimental which worked in Hogwart's Legacy.
@TheWyn is this when launching or when playing the game for a while? it's launching fine on my end with mesa 25.1.3 and git with my rx 6600, can you give more info about your setup? and does it work with other proton versions for you?
@TheWyn is this when launching or when playing the game for a while? it's launching fine on my end with mesa 25.1.3 and git with my rx 6600, can you give more info about your setup? and does it work with other proton versions for you?
Well that's funny, the new bleeding update fixed it, version experimental-bleeding-edge-10.0-210316-20250619-pe347be-wb79136-de668e3-v59e50c. It was happening just after the splash screen when the window opens.
I ran this by gemini a bit to see if it might be helpful in diagnosing the issue. The game's haptics start at the menu, simply going up and down it, so I looked for how that is setup in the logs, some example trace areas
I think this is when the source is setup, it appears towards the top of the log
038c:trace:xaudio2:XA2ECB_OnProcessingPassStart 0000000031DCCF30
038c:trace:xaudio2:XA2ECB_OnProcessingPassEnd 0000000031DCCF30
03e0:trace:xaudio2:IXAudio2Impl_CreateSourceVoice (0000000077B52D00)->(00000000511A4450, 000000005EB3FB50, 0x2, 2.000000, 00000000511A4458, 0000000000000000, 0000000000000000)
03e0:trace:xaudio2:IXAudio2Impl_CreateSourceVoice Created source voice: 0000000032B9E0B0
03e0:trace:xaudio2:XA2SRC_SubmitSourceBuffer 0000000032B9E0B0, 000000005EB3FB70, 0000000000000000
03e0:trace:xaudio2:XA2SRC_Start 0000000032B9E0B0, 0x0, 0x0
0410:trace:xaudio2:XA2ECB_OnProcessingPassStart 0000000077B52D00
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnBufferStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassEnd 0000000032B9E0B0
0410:trace:xaudio2:XA2ECB_OnProcessingPassEnd 0000000077B52D00
0410:trace:xaudio2:XA2ECB_OnProcessingPassStart 0000000077B52D00
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassEnd 0000000032B9E0B0
0410:trace:xaudio2:XA2ECB_OnProcessingPassEnd 0000000077B52D00
0410:trace:xaudio2:XA2ECB_OnProcessingPassStart 0000000077B52D00
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnBufferEnd 0000000032B9E0B0
0410:trace:xaudio2:XA2SRC_SubmitSourceBuffer 0000000032B9E0B0, 0000000059B5FCF0, 0000000000000000
0410:trace:xaudio2:XA2VCB_OnBufferStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnBufferStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassEnd 0000000032B9E0B0
0410:trace:xaudio2:XA2ECB_OnProcessingPassEnd 0000000077B52D00
Here's another voice setup (I'm guessing one is for the normal audio playback, the other for the haptics)
017c:trace:xaudio2:IXAudio2Impl_StartEngine (0000000031DCD930)->()
017c:trace:xaudio2:IXAudio2Impl_CreateSourceVoice (0000000031DCD930)->(00000000511A4450, 0000000000B9ECB8, 0x2, 2.000000, 00000000511A4458, 0000000000000000, 0000000000000000)
017c:trace:xaudio2:IXAudio2Impl_CreateSourceVoice Created source voice: 000000000BA6EFA0
Here's when I think playback was requested on the 2nd one, 000000000BA6EFA0 (when input occurred, as I see hid trace around it)
03cc:trace:xaudio2:XA2SRC_SubmitSourceBuffer 000000000BA6EFA0, 000000005C4FFE90, 0000000000000000
017c:trace:xaudio2:XA2SRC_Start 000000000BA6EFA0, 0x0, 0x0
038c:trace:xaudio2:XA2ECB_OnProcessingPassStart 0000000031DCCF30
038c:trace:xaudio2:XA2ECB_OnProcessingPassEnd 0000000031DCCF30
And here's when it occurred for 0000000032B9E0B0
410:trace:xaudio2:XA2ECB_OnProcessingPassStart 0000000077B52D00
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnBufferEnd 0000000032B9E0B0
0410:trace:xaudio2:XA2SRC_SubmitSourceBuffer 0000000032B9E0B0, 0000000059B5FCF0, 0000000000000000
0410:trace:xaudio2:XA2VCB_OnBufferStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnBufferStart 0000000032B9E0B0
0410:trace:xaudio2:XA2VCB_OnVoiceProcessingPassEnd 0000000032B9E0B0
0410:trace:xaudio2:XA2ECB_OnProcessingPassEnd 0000000077B52D00
I'm somewhat ignorant, it just seems like perhaps one of these should be directing to the actual controller. There's no errors or warnings in trace that look relevant to me, but I think (in my ignorance) that perhaps one of the above is potentially misdirected. I just don't know how to get much further on it with my level of knowledge. Hopefully this may be helpful to someone who knows more.
When testing the Game on Windows, there was one occurrence of the game loosing the Controller (Dualsense) and not reacting to it anymore unless the game was closed and opened again.
The controller was still connected and functional (tests in Steam work, Windows Controller Tests work)
This suggests that the controller losing issue might be game related and not necessarily due to Proton.
When testing the Game on Windows, there was one occurrence of the game loosing the Controller (Dualsense) and not reacting to it anymore unless the game was closed and opened again.
How do you reproduce this?
i tested it again with proton experimental, now the adaptive triggers work but i have zero vibration
i tested it again with proton experimental, now the adaptive triggers work but i have zero vibration
Vibrations (as regular rumble) only through Bluetooth. USB - no vibrations (game can't find gamepad audio device).
i tested it again with proton experimental, now the adaptive triggers work but i have zero vibration
Adaptive triggers worked before, the issue was just with the advanced haptics. I'm pretty sure it has something to do with how it's mapping the audio devices and will require some backend work on the specifics of how they went about implementing that in Stellar Blade. There are other games that work fine, so I think Stellar Blade probably did it a little bit differently. Sadly I don't know enough to really troubleshoot it, and I think it would probably require a significant amount of time on a developer to fix. So unless other new games have the issue, I doubt we'll see a fix anytime soon unless a good developer loves the game (and haptics) enough to contribute patches for it. I can't blame people for not putting that work in, I knew it was over my head.
Another issue is HDR gameplay - there's an error while saving an HDR photo in photo mode. Turn on HDR (DXVH_HDR=1 + turn on HDR in game + make sure photo mode has HDR enabled) and try to take a photo in photo mode - error while saving a photo.
Separate report for a launch crash on AMD/RADV.
This looks distinct from the original DualSense haptics issue in this thread.
I’m posting it here because I did not find a dedicated centralized Proton 11 beta feedback issue in this repo, and the strongest signal from my local logs is that this failing launch correlates with switching the game from GE-Proton10-34 to Proton 11.0 (Beta).
Compatibility Report
- Name of the game with compatibility issues: Stellar Blade
- Steam AppID of the game: 3489700
System Information
- GPU: AMD Radeon RX 7900 XTX (RADV NAVI31)
- Video driver version: Mesa RADV 26.0.4
- Kernel version: 6.19.11-cachyos-lto
- Proton version: Proton 11.0 (Beta) during the failing launch
- Launch options:
PROTON_LOG=1 gamemoderun %command%
Symptoms
- Game starts through Steam/pressure-vessel/Proton, opens the process tree, then exits about 8 seconds later without reaching a usable in-game state.
- The game writes Windows crash dumps under
StellarBlade/.crashdata/reports/for each failed launch. - Steam logs show the AppID as running briefly, then every tracked process exits.
Reproduction
- In Steam, set the compatibility tool to
Proton 11.0 (Beta). - Keep launch options set to
PROTON_LOG=1 gamemoderun %command%. - Launch Stellar Blade from Steam.
- Observe that the game starts and exits within a few seconds, leaving a new dump in
.crashdata/reports/.
Timeline / Verified local evidence
2026-04-22 07:47:17: Steam startsPROTON_LOG=1 gamemoderun ... Proton 11.0 ... SB.exe.2026-04-22 07:47:18: Steam logsgamemodeauto: dlopen failed - libgamemode.so: cannot open shared object file: No such file or directory.2026-04-22 07:47:20: Steam initializes breakpad / minidump handling.2026-04-22 07:47:26: Steam reportsGame Recording - game stopped [gameid=3489700]and removes all tracked processes for AppID3489700.- New crash dump written at the same time:
StellarBlade/.crashdata/reports/045e1a8a-951d-4f43-968f-b2c2e59b4801.dmp(~645 MB).
Important notes about local-only noise
- I do have two local GameMode-related problems on this machine:
gamemodeautocannot findlibgamemode.sobecause of a local NixOS packaging mistake.gamemodedalso fails some privileged helper actions because my shell is Nushell and it is not currently listed in/etc/shellson this machine.
- I do not think those fully explain the crash:
- the game still creates its own Windows crash dump on every failed run;
- there are no
amdgpu/ DRM reset / GPU fault lines in the kernel journal during the failure window; - the game exits as a user-space crash rather than a kernel GPU hang.
Potentially relevant regression detail
- Steam history on this machine shows the game was previously mapped to
GE-Proton10-34. - On April 22, 2026, the failing launch was specifically mapped to
Proton 11.0 (Beta). - Later the same morning I switched the compatibility tool back to
GE-Proton10-34, so this may be a Proton-version-specific regression rather than a generic game-startup failure.
Available local artifacts
- Steam logs:
~/.local/share/Steam/logs/gameprocess_log.txt~/.local/share/Steam/logs/compat_log.txt~/.local/share/Steam/logs/console-linux.txt
- Game crash dumps:
.../StellarBlade/.crashdata/reports/045e1a8a-951d-4f43-968f-b2c2e59b4801.dmp.../StellarBlade/.crashdata/reports/c0ce5328-cdbe-4ea2-9419-959e97c6bcfb.dmp.../StellarBlade/.crashdata/reports/f934dcb9-419d-4690-9549-677193df1aec.dmp
If useful, I can also provide the exact Steam log excerpts around the 07:47:17 to 07:47:26 launch window.
Log uploads
- Steam-side logs that capture the failing
Proton 11.0 (Beta)launch window: https://gist.github.com/caniko/7fe3827dc704d35ff782c2763573d1f1 - Note: the standalone
steam-3489700.logon this machine was overwritten later byGE-Proton10-34retests, so I am not attaching that file as representative of the failingProton 11.0 (Beta)run.
Why is my comment marked as low quality?
Hello @caniko, for transparency, it was marked low quality because you've decided to post what looks like an AI analysis without a Proton log. You've posted in a long form format what can be boiled down to: "Something unknown and undefined happened with the game and Proton 11." This isn't useful feedback when it takes a page of reading.
For transparency, I tried to give you as much information as possible, it is not one page. I would appreciate more inclusive communication principles from representatives of Valve.
The comment clearly isn't low quality, you just wanted more information. You can ask me for it; I can be cooperative. We want to make Proton 11 the best it can be.
Anyway here are the the logs, I went back and generated them for you, but please ask next time! AI or not.
steam-3489700.log
@caniko the problem with ai assisted reports (at least the ones I've seen posted for proton issues so far), is that they add verbosity and uncertainty where brevity and clarity is preferred, makes it hard to tell how much was experienced by the user and how much was tacked on because it thought it looked good, so they end up making the report harder to parse.
As for a good baseline for reports, there is a template available when you try to open a new issue, you can copy it and replace the relevant information.
Besides that, things you always to do is remove everything custom you've introduced so you can work with a clear base to help narrow things down. In your case, removing gamemoderun, also deleting the prefix which is something that should be tried when switching between proton versions (specially when third party versions are involved), and, specially, removing mods. I didn't see it mentioned in your comments, but you appear to be using optiscaler, you should also remove that and see if it's a factor.
At least on my end, with the same mesa and proton versions, but with a rdna4 card and unmodded game, it is running fine.
Same problem here. however there is a custom proton build which makes the full DualSense functionality working in "Stellar Blade": https://github.com/xzn/proton-ds5-haptic . btw the problem doesn't seem to affect only "Stellar Blade" but other games like "Death Stranding II" too. so maybe you could merge the patches in the linked custom proton into the official proton?
For whatever reason the game does not always start the Steam overlay which causes issues for my Steam controller since it requires Steam Input.
When that happens I also have issues turning off the game and make Steam realize the game is turned off.
I actually end up killing the Steam client instead because that's the only way to "unblock" it.
Problem is I can't reliably make it happen but it seems to happen regularly.
I'll try to get a log once it happens again.
EDIT: Got a log
steam-3489700.log
Additional information about my System:
steamruntimediag.txt
steamsysteminfo.txt
So far i've it has happened to me via:
Steam stable,
Steam beta,
Steam beta with RT3,
Proton Experimental and even third party Proton versions
Stellar Blade Crash Fix. (Fatal Error)
Issue transferred from https://github.com/ValveSoftware/Proton/issues/10083.
@Kuroum posted on 2026-08-21T20:37:38:
Compatibility Report
- Name of the game with compatibility issues: Stellar Blade
- Steam AppID of the game: 3489700
System Information
- GPU: NVIDIA GeForce RTX 5090
- Video driver version: NVIDIA 610.57.04
- Kernel version: 7.1.8-artix1-3
- Link to full system information report as Gist: N/A
- Proton version: Proton 11
I confirm:
- [x] that I haven't found an existing compatibility report for this game.
- [x] that I have checked whether there are updates for my system available.
Symptoms
Stellar Blade crashed immediately after startup under Proton 11.
The Proton log showed an access violation in xaudio2_9.dll:
Exception 0xc0000005
L"xaudio2_9.dll" + 000000000000D15A
The game also ships its own XAudio2 redistributable at:
Engine/Binaries/ThirdParty/Windows/XAudio2_9/x64/xaudio2_9redist.dll
Replacing the prefix's system32/xaudio2_9.dll with this xaudio2_9redist.dll allowed the game to start successfully.
The replacement caused a separate audio issue because my PipeWire PulseAudio server was not running. After starting pipewire, wireplumber, and pipewire-pulse, audio worked correctly.
The issue therefore appears to be related to Proton's xaudio2_9.dll rather than the game itself.
Reproduction
-
Install and launch Stellar Blade with Proton 11.
-
The game crashes during startup.
-
The Proton log shows an EXCEPTION_ACCESS_VIOLATION in xaudio2_9.dll.
-
Replace the following file in the Proton prefix:
compatdata/3489700/pfx/drive_c/windows/system32/xaudio2_9.dll
with:
common/StellarBlade/Engine/Binaries/ThirdParty/Windows/XAudio2_9/x64/xaudio2_9redist.dll
-
Launch the game again. The game starts successfully.
Proton versions
proton experimentalx6 2026-06ge-proton10-34x1 2026-04proton 11.0x1 2026-04
Launch options
PROTON_LOG=1x1 2026-04
Upstream links
- ValveSoftware/Proton/issues/10083x1 2026-08
- ValveSoftware/Proton/issues/5900x1 2025-06
DLLs
xaudio2_9.dllx1 2026-08xaudio2_9redist.dllx1 2026-08
Error codes
0xc0000005x1 2026-08
Compatibility Report
Stellar Blade (3489700)
System Information
I confirm:
Symptoms
Zero haptic feedback using a dual sense controller (this does work on other games in Linux, like Ghosts of Tsushima, and it does work perfectly in Windows, and this game has great haptics in general so I'd really like it to work like the other games do with native dual sense support).
Reproduction
Steam input disabled
Controller plugged in via front usb port
Game started (haptics on menu don't work)
Continue game (haptics while swinging sword and landing from air don't work)
steam-3489700.log
I added a custom udev rule someone mentioned in a forum to disable touchpad as a mouse (after I encountered this problem), that's the only customization I really have for this controller.
Some report haptics do work using bluetooth, although not the full normal haptics for dual sense, but for me after pairing with bluetooth the controller doesn't work. I have to use it in wired mode (which is preferred anyway).