

Joystick Control Feedback - 1.3.306
Started by Kyle Polulak, Jul 15 2014 09:37 AM
71 replies to this topic
#61
Posted 23 July 2014 - 12:04 PM
I think I follow. Yes I was oversimplifying and made an error. And it has been a long time since Z80 machine code, 2 s complement and bcd notation and all those arcane things.
#62
Posted 23 July 2014 - 02:25 PM
Ah, those were the days.
If you wanted to play a game, you had to type it in first
If you wanted to play a game, you had to type it in first

#63
Posted 25 July 2014 - 01:35 PM
Here's what I have in my config file, It works but the dead zone is there.
Here's the JS I'm currently using.
http://www.conrad.co...FB.EPS_1000.jpg
cl_joystick_gain = 8
cl_joystick_sensitivity = 5
cl_joystick_throttle_range = 0
cl_joystick_invert_throttle = 1
cl_joystick_invert_pitch = 1
cl_joystick_invert_yaw = 0
cl_joystick_invert_turn = 0
cl_joystick_deadzone = 0.040
;i_joystick_buffered = 1
;i_joystick_deadzone = 0.040
;i_joystick_deadzone = 0.0005 //The deadzone is Set at the lowest so that you get the maximum range of your joystick
;cl_joystick_deadzone = 0.0005 //You need both of these lines set the same to make the deadzone work properly
Here's the JS I'm currently using.
http://www.conrad.co...FB.EPS_1000.jpg
cl_joystick_gain = 8
cl_joystick_sensitivity = 5
cl_joystick_throttle_range = 0
cl_joystick_invert_throttle = 1
cl_joystick_invert_pitch = 1
cl_joystick_invert_yaw = 0
cl_joystick_invert_turn = 0
cl_joystick_deadzone = 0.040
;i_joystick_buffered = 1
;i_joystick_deadzone = 0.040
;i_joystick_deadzone = 0.0005 //The deadzone is Set at the lowest so that you get the maximum range of your joystick
;cl_joystick_deadzone = 0.0005 //You need both of these lines set the same to make the deadzone work properly
Edited by Celtic Warrior, 25 July 2014 - 01:37 PM.
#64
Posted 26 July 2014 - 08:06 AM
Yeah, I have a similar stick but I doubt it is gonna be any good for aiming in MWO due to it's lack of resolution.
#65
Posted 26 July 2014 - 10:13 AM
It's not bad but the dead spot is quite noticeable, If there were a better stick on the market I would consider purchasing it.
#66
Posted 26 July 2014 - 02:35 PM
I'm still waiting on Extreme 3D Pro support. Tried it, and it still sucks. C'mon, guys, do better than this.
#67
Posted 26 July 2014 - 02:35 PM
So one bug with the Joystick which is more pronounced now is that when you adjust it for Testing Grounds it is very sluggish in an actual online match. Previously the difference was very minor so you could just set the sensitivity up slightly higher than you wanted on Testing Grounds and then in-game it would be fine. Now the difference is so great you have to reset it while in a match as best as you can.
Absolute Inputs didn't give me the control method I wanted. Maybe if I used rudder pedals it would be more beneficial.
Absolute Inputs didn't give me the control method I wanted. Maybe if I used rudder pedals it would be more beneficial.
#68
Posted 26 July 2014 - 02:45 PM
Hawk819, on 26 July 2014 - 02:35 PM, said:
I'm still waiting on Extreme 3D Pro support. Tried it, and it still sucks. C'mon, guys, do better than this.
I am using a Logitech 3D Force Pro, which I believe uses the same Profiler as the Extreme 3D (named Logitech Profiler with blue lettering). It's working great for me. Here is my User.cfg sets, maybe try it and then adjust the Controller sensitivity slider in MWO/Options.
cl_joystick_gain = 3.277 cl_joystick_sensitivity = 4.777 cl_joystick_throttle_range = 0 cl_joystick_invert_throttle = 0 cl_joystick_invert_pitch = 0 cl_joystick_invert_yaw = 0 cl_joystick_invert_turn = 0 I_joystick_deadzone = 0.007 I_joystick_buffered = 0
In the User.cfg, Higher Gain makes large joystick movements faster and higher Sensitivity makes smaller movements more accurate. Fiddle with those until you like the results. I set my Sensitivity in the Logitech Profiler to about 17.
#69
Posted 27 July 2014 - 09:35 AM
Do you find the logitech to be a better JS then the thrustmaster? Years ago I think I tried them both out and preferred the thrustmaster but I don't think it's been changed. Has the logitech been updated?
I have my thrustmaster working pretty good but the dead spot is noticeable.
I have my thrustmaster working pretty good but the dead spot is noticeable.
#70
Posted 27 July 2014 - 09:38 AM
What I would like to see is the "select next weapons group" option in the JS setup. Not sure why this is missing it should be standard on any joystick sim game where multiple weapons groups are present.
#71
Posted 27 July 2014 - 05:45 PM
Test Note 3:
Having built a custom joystick, I decided to do a bit more testing. It seems the problems I noted with the stick not aligning to the mechs position in cockpit / first person view, has to do with arm lock.
When arm is toggled on, there is a very good alignment between the position of the mech's torso and the position of the stick, outside of the deadzone, in the cockpit view / first person. Micro adjustments for accuracy are possible,though probably not practical for most sticks.
When arm lock is off, the stick is controlling the absolute position of the arms. The torso tries to auto align to the position of the arm reticule. So while you have absolute control over the arms, you do not have absolute control over the torso of the mech. Hence it seems like the reporting discrepancy i described earlier exists, when in fact it does not. The behavior noted is a function of the mech's torso auto aligning to the arms. An illustration of the behavior is in the spoiler below
When using arm lock in cockpit / first person, the behavior of the system seems to be working well. Though it still needs some tweaks to make useable. In addition to the non linearity and saturation suggestion I made previously, a few other suggestion may make the system much more useable.
First is that deadzone needs be reduced, to around .002 for a 16 bit reporting system. That would ignore the first 136 positions. A noise filter or smoothing for joystick inputs would also be handy. Something that could filter out variances of +-100 of outputs from the stick. This would reduce the flickering and micro adjustments that occur, because of the small changes in reporting of the stick due signal noise, dusty sensors, ect.
Getting a unlocked torso to behave more intuitively would be greatly enhanced by such measures. Also, changing the behavior of the torso auto alignment would yield a more intuitive control scheme. If the arms and torso stayed aligned until the torso achieved it's maximum rotation, then the arms unlocked and were able to rotate beyond maximum torso rotation. You would get the benefits an unlocked torso, with the useability of a locked torso.
Honestly if you get the deadzone reduced, non linearity and saturation implemented and a noise filter. The system would work very well for torso locked mech's in first person. The cursor behavior doesn't seem like that big of change, it probably could come later.
image of my custom stick, made from a surplus M1A1 tank turret controller.
Having built a custom joystick, I decided to do a bit more testing. It seems the problems I noted with the stick not aligning to the mechs position in cockpit / first person view, has to do with arm lock.
When arm is toggled on, there is a very good alignment between the position of the mech's torso and the position of the stick, outside of the deadzone, in the cockpit view / first person. Micro adjustments for accuracy are possible,though probably not practical for most sticks.
When arm lock is off, the stick is controlling the absolute position of the arms. The torso tries to auto align to the position of the arm reticule. So while you have absolute control over the arms, you do not have absolute control over the torso of the mech. Hence it seems like the reporting discrepancy i described earlier exists, when in fact it does not. The behavior noted is a function of the mech's torso auto aligning to the arms. An illustration of the behavior is in the spoiler below
Spoiler
When using arm lock in cockpit / first person, the behavior of the system seems to be working well. Though it still needs some tweaks to make useable. In addition to the non linearity and saturation suggestion I made previously, a few other suggestion may make the system much more useable.
First is that deadzone needs be reduced, to around .002 for a 16 bit reporting system. That would ignore the first 136 positions. A noise filter or smoothing for joystick inputs would also be handy. Something that could filter out variances of +-100 of outputs from the stick. This would reduce the flickering and micro adjustments that occur, because of the small changes in reporting of the stick due signal noise, dusty sensors, ect.
Getting a unlocked torso to behave more intuitively would be greatly enhanced by such measures. Also, changing the behavior of the torso auto alignment would yield a more intuitive control scheme. If the arms and torso stayed aligned until the torso achieved it's maximum rotation, then the arms unlocked and were able to rotate beyond maximum torso rotation. You would get the benefits an unlocked torso, with the useability of a locked torso.
Honestly if you get the deadzone reduced, non linearity and saturation implemented and a noise filter. The system would work very well for torso locked mech's in first person. The cursor behavior doesn't seem like that big of change, it probably could come later.
image of my custom stick, made from a surplus M1A1 tank turret controller.
Spoiler
Edited by Grits N Gravy, 27 July 2014 - 05:54 PM.
#72
Posted 28 July 2014 - 01:41 PM
Well, step away from the game for a couple of weeks and wow, this made a mess of things. I've always used my Saitek X52 pro to play. Please tell me this will either be fixed next patch or at least rolled back, because the games is pretty much unplayable at the moment.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users