Linux Port – Feedback, Known Issues, and FAQ

I've the same issue with my G29 and my Logitech Attack 3 (Joystick which I use as a handbrake).
Even when I assign steering just to the X-Axis of my wheel, tilting the joystick in it's X-Axis also steers the car. This makes the Joystick as a Handbrake basically unusable, as it's very hard to ONLY tilt it in one direction and not even a slight amount to the left/right.

 

uh... I'm on Ubuntu 20.04 and when I open that file it says that there is no application to open SharedLib files. Any help? I'm a linux noob

 

Latest Proton Experimental has this on the changelog:

• Improve force feedback compatibility for BeamNG and Forza Horizon 5

Any idea what this fixes and for what version of the game? I guess the DX11 instead of the Windows Vulkan version, but I don't know.

 

I'm having the same issue with my Fanatec CSL Pedals (USB Adapter), and my Logitech DFGT. Have you had any progress with this issue?

 

sadly, not
If I have, I'll post another reply to this thread

 

thx for linux support!

have a Segmentation fault on start map:
[ +0.000035] Code: 00 5b c3 c3 90 66 90 f3 0f 1e fa 48 83 ec 18 64 48 8b 04 25 28 00 00 00 48 89 44 24 08 31 c0 48 8b 05 41 a8 c5 01 48 89 3c 24 <48> 8b b0 b0 00 00 00 48 3b b0 b8 00 00 00 74 2d 48
89 3e f0 83 47
[Feb19 01:13] BeamNG.drive.x6[10486]: segfault at 145 ip 00007efe0fe24f7e sp 00007ffe74c6b1b0 error 4 in libnvidia-glcore.so.525.89.02[7efe0e87f000+2505000] likely on CPU 12 (core 24, socket 0)
[ +0.000032] Code: 2c 8f 48 8d 3c b2 41 89 ef 48 89 7c 24 48 4d 89 ce 0f 1f 00 48 8b 4c 24 08 41 8d 57 ff 48 8d 14 52 48 8d 34 91 8b 16 c1 ea 05 <41> 3b 96 04 01 00 00 0f 83 0d 01 00 00 48 c1 e2 06
49 03 96 08 01

 

what distro do you use and what GPU do you have?

 

gentoo model name : 12th Gen Intel(R) Core(TM) i7-12700KF

 

you use the closed source nvidia driver or nouveau?

 

x11-drivers/nvidia-drivers-525.89.02

steam proton looks full well
dump
0x0.st / HsXo.dmp

ps: I had already something similar(bug glibc-2.34 patch 803950 dot bugs.gentoo dot org/attachment.cgi?id=757176) with browsers maybe there is an option to run --no-sandbox

 

I think I found a solution:

It seems you can edit the function getDeviceInfo() at arount line 410 in BeamNG.drive/lua/ge/extensions/core/input/bindings.lua
From

Code:

local function getDeviceInfo(device)
  -- ask T3D information about the provided devname
  local guid = WinInput.getProductGUID(device)
  local productName = WinInput.getProductName(device)
  local pidvid = WinInput.getVendorIDProductID(device)
  --local battery = WinInput.getBatteryLevel(device)
  return guid, productName, pidvid, battery
end

To:

Code:

local function getDeviceInfo(device)
  -- ask T3D information about the provided devname
  local guid = WinInput.getProductGUID(device)
  local productName = WinInput.getProductName(device)
  local pidvid = WinInput.getProductGUID(device)
  --local battery = WinInput.getBatteryLevel(device)
  return guid, productName, pidvid, battery
end

This way the system uses the entire GUID for mapping instead of just the first 8 chars

I was able to bind Button0 of both shifter and wheel to different actions as well as the same action.

As seen below, the game does not complain about duplicate bindings anymore.

 

DAMN! Thanks a lot!

Now the game just has to get a LOT more stable, and I can finally enjoy BeamNG.drive how it's meant to be!

(since I think the last update, when setting the texture quality to 'high', I get a black/white mess)

System:
Garuda Linux, 6.1.12-zen1-1-zen
mesa 22.3.4-1
BeamNG.drive 0.27.1
KDE 5.27.0, Wayland
i5-12600k
Gigabyte Z590 Gaming DDR4
MSI RX 6700 XT

when verifying the gamefiles with the windows version through wine, the bindings.lua is the only file that didn't verify correctly (as expected)

 

I am having the same issue on Arch Linux with a RTX 2060 Super and Nvidia driver version 530.30.02 If I use 520.56.06 or below then the game works well. Anything above that doesn't work. I have the issue whether I use proton or I use the native build. Usually when I start the game and the window opens, I get a segfault after " 4.19594|D|createDevice| Using default audio device: SteelSeries Arctis 7 Game", sometimes I get through and a few seconds after I load a map I get the segfault:

Code:

17427 segmentation fault (core dumped)  ./BeamNG.drive.x64

dmesg:

Code:

[  127.996448] BeamNG.drive.x6[3695]: segfault at 2dc0 ip 00007f6335b52399 sp 00007ffc006626b0 error 4 in libnvidia-glcore.so.530.30.02[7f6334501000+25e4000] likely on CPU 14 (core 6, socket 0)
[  127.996458] Code: 48 89 7c 24 58 44 89 ff 4d 89 e7 41 89 fc e9 e4 00 00 00 66 0f 1f 44 00 00 48 c1 e2 06 49 03 97 10 01 00 00 0f 84 ed 00 00 00 <8b> 3a 8d 77 fa 83 e6 fd 74 0c 81 ff 18 25 a0 3b 0f 85 d7 00 00 00

 

I've noticed something weird when using my steering wheel. Now I don't know whether I just don't understand how the feature works or bug:
When setting the update rate limit to any setting above 15 I get an output in dmesg like this

Code:

tmff2: output queue full

, but only when the frame rate is not limited.

The output is caused by my wheels driver because it was overwhelmed with inputs. Leaving the setting at automatic causes the same output.

When limiting the frame rate to 80, I can set the update rate limit as high as 2000 without any output in dmesg.
Anything above 80 will result in the same output.

What exactly does the update rate limit do? the wheel update rate seems to be tied to the framerate rather than the update rate limit setting

 

seems to work for me

(Ultra preset)

 

I'll try later on my Laptop (RTX 3080, nvidia-open driver)