Difference between revisions of "Talk:Manual:Config:Advanced:Bindings:Joystick"

From VsWiki
Jump to: navigation, search
(Stuck a header in.)
(SideWinder (and other) joysticks under Linux: Signed it.)
Line 8: Line 8:
 
As for buttons, things are a little more normal: Buttons 0 through 7 are directly mapped to the trigger through button 8 on the joystick.  Rather cute, however, is the pseudo button 11, which is "pressed" when the infrared sensor detects the hand leaving the joystick.  I could see some fun with this, like making it a jump, dock, or pause button. :)
 
As for buttons, things are a little more normal: Buttons 0 through 7 are directly mapped to the trigger through button 8 on the joystick.  Rather cute, however, is the pseudo button 11, which is "pressed" when the infrared sensor detects the hand leaving the joystick.  I could see some fun with this, like making it a jump, dock, or pause button. :)
  
What you need to map stuff properly is to get the program 'jstest', available (at least in Debian) as part of the 'joystick' package.  Locate your joystick device (/dev/input/js* or /dev/js*) and run jstest on it, then frob some of the controls around.  Things should become apparent quite quickly, and you may find some cool stuff (like my hand sensor).
+
What you need to map stuff properly is to get the program 'jstest', available (at least in Debian) as part of the 'joystick' package.  Locate your joystick device (/dev/input/js* or /dev/js*) and run jstest on it, then frob some of the controls around.  Things should become apparent quite quickly, and you may find some cool stuff (like my hand sensor). — [[User:Wisq|Wisq]] 13:29, 30 Jul 2005 (PDT)

Revision as of 21:29, 30 July 2005

SideWinder (and other) joysticks under Linux

Regarding the comment about the bindings not working in Linux: One thing to keep in mind is that the Linux joystick drivers (particularly my USB-based one) can receive inputs in the weirdest ways.

For example, on my joystick, I've got standard movement, twist action, a throttle, a digital hatswitch, and eight buttons: one trigger, three thumb buttons, four base buttons. But Linux sees my joystick as having 11(!) axes and 20(!!) buttons. So things are mapped a little differently.

On mine, for example, axes 0 and 1 are the left-right and up-down respectively (normal). Axis 4 is the left-right twist action, not 2 or 3 as you would expect. Axis 5 is the throttle. And even though the hatswitch is digital, axes 6 and 7 are the hat left-right and up-down, respectively — but they're always either -1, 0, or +1. The rest seem to be unused.

As for buttons, things are a little more normal: Buttons 0 through 7 are directly mapped to the trigger through button 8 on the joystick. Rather cute, however, is the pseudo button 11, which is "pressed" when the infrared sensor detects the hand leaving the joystick. I could see some fun with this, like making it a jump, dock, or pause button. :)

What you need to map stuff properly is to get the program 'jstest', available (at least in Debian) as part of the 'joystick' package. Locate your joystick device (/dev/input/js* or /dev/js*) and run jstest on it, then frob some of the controls around. Things should become apparent quite quickly, and you may find some cool stuff (like my hand sensor). — Wisq 13:29, 30 Jul 2005 (PDT)