Results 1 to 2 of 2

Thread: Bluetooth DualShock3 has incorrect mappings in 15.10

  1. #1
    Join Date
    Nov 2015
    Beans
    2

    Bluetooth DualShock3 has incorrect mappings in 15.10

    I've tested this scenario with two dual shock controllers on two different 15.10 (wily) installs. The same issue appears with both controllers on both systems.

    Using a wired connection, the DS3 controllers behaves mostly as expected.

    * The lights flash when connected, before the controller is activated.
    * The lights turn off and the controller is activated when the PS button is pressed.
    * The PS button opens big picture mode in Steam.
    * Games detect the controller
    * Buttons are mapped as expected (same positions as an xbox controller).
    * The device path is a standard value: /dev/input/js0
    * jstest-gtk shows the device name as: Sony PLAYSTATION(R)3 Controller
    * jstest-gtk can be used to verify the axis and buttons events / mappings (Select = 0, R3 = 2, PS = 16, X = 14)

    After pairing the controller via the `System Settings -> Bluetooth -> +`:

    * DIFFERENT LIGHTS: The controller port light correctly indicates the controller mapping (light 1 -> js0, etc)
    * SAME PATH: The device path is the same: /dev/input/js0
    * DIFFERENT NAME: jstest-gtk shows the device name as: PLAYSTATION(R)3 Controller
    * jstest-gtk visual feedback shows the same buttons being triggered as a wired connection. (Select = 0, R3 = 2, PS = 16, X = 14)
    * The PS button DOES NOT open big picture mode in Steam.
    * Buttons are mapped wrong in-game: Select acts as X; R3 acts as Square

    The Bluetooth connection correctly detects the controller as a joystick, and the mappings for the controller appear the same in jstest. In game, however, the bluetooth connected controller is mapped completely differently from the same controller over a wired connection.

    Any idea how to fix this?

  2. #2
    Join Date
    Nov 2015
    Beans
    2

    Re: Bluetooth DualShock3 has incorrect mappings in 15.10

    I used jscal-store to write the mappings file to /var/lib/joystick/joystick.state.

    joystick.state.txt

    This verifies my observations about the device name, and that the joystick mappings are the same. I'm still not clear on why the effective mappings of the controller are different.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •