Jump to content

ALL JOYSTICK USERS

joystick flight system HOTAS configuration config controller

373 replies to this topic

#81 Loc Nar

    Member

  • PipPipPipPipPipPipPipPip
  • 1,132 posts

Posted 10 April 2013 - 11:35 AM

Quote

Any help is appreciated.


I can't help with your specific problem, but as a user of emulator based controls I have had a fully functioning setup since CB for my mechpit that has been immune to all the changes that have periodically ruined other users interfaces since MWO thinks it's a kbm. EvilC has an emulator program that should be able to get you back up and running, and also protect you from future sillyness and bugs (which there are several known ones that are being worked on, but still...)
http://mwomercs.com/...tual-joysticks/

#82 DBrando

    Rookie

  • FP Veteran - Beta 2
  • 8 posts

Posted 22 April 2013 - 12:54 AM

Anyone succesfully mapped minimap ("B") and center torso to legs ("F")?
My joystick seems can't "read" this two buttons.

Edited by DBrando, 22 April 2013 - 12:55 AM.


#83 MentalPatient

    Member

  • PipPipPipPipPip
  • 145 posts

Posted 23 April 2013 - 03:15 AM

For anyone interested in using the joystick in absolute tracking rather than relative which is currently the only supported way unless you have a thrustmaster stick, I am using a tool called joystick mouse tool, available here:

http://www.deinmeister.de/jmt_e.htm

I much prefer this way of moving the joystick rather than the relative movement way. This tool basically emulates the mouse using your joystick input values, and you can select either relative or absolute. It's one of the most accurate and bugless program's of its type I've come across. You have to adjust sensitivity and dead zone within your own profiler as the software doesn't allow adjusting these in absolute mode, but that isn't too much of a problem. Also I had to add in the user.cfg lines for mouse sensitivity, and had to set it way down as the default sent the view going crazy.

I feel now that this method of controlling is superior to the default joystick input, which when you move the stick, it moves the cursor at a certain rate, relative to where you place the stick. Absolute mode is closer to that of a mouse, although I am not superior to using mouse as yet, maybe I might approach that with time. The only problem is due to the high sensitivity of this mode, you almost need to have a stick extender for finer control. I am using a logitech g940 flight system with throttle, stick and pedals, and to me this is the most immersive way of controlling yet. I was using mouse for aim, throttles for speed and pedals for turning, which I feel is the most superior setup. But for those wanting to use their stick effectively I can reccommend using this software to enhance your aiming with the stick.

Edited by MentalPatient, 23 April 2013 - 03:17 AM.


#84 Loc Nar

    Member

  • PipPipPipPipPipPipPipPip
  • 1,132 posts

Posted 23 April 2013 - 08:06 AM

Quote

I feel now that this method of controlling is superior to the default joystick input, which when you move the stick, it moves the cursor at a certain rate, relative to where you place the stick. Absolute mode is closer to that of a mouse, although I am not superior to using mouse as yet, maybe I might approach that with time. The only problem is due to the high sensitivity of this mode, you almost need to have a stick extender for finer control.


You are probably only one of like 8 people that actually understands what it means that I built my own stick specifically for absolute inputs and mech piloting (mounted in my new mechpit). Rather than a spring centering pitch/roll gimbal it's an azimuth/zenith bearing and has greased plastic rubs for damping, since there is more to a mouse than it's use of absolute inputs that gives it the edge (see below). The bad news is that using a normal stick does quite get you there due to the pith/roll/spring centered gimbal being largely suited for relative inputs, so even if you use relative inputs it's still pretty goofy, just not as goofy as relative inputs, but you have identified one of the key pieces of the puzzle, which puts you in a much better position than most that have the desire to pilot with a stick.

Question: does your program allow independent control of x/y axis? I have found (corroborated by other stick users) that the y axis is noticeably more sensitive than the x and needs to be dumbed down a little more than the x does to get them proportionally correct with each other. From your post it sounds like there is no sensitivity control, but at least you can lower the mouse sensitivity in your user .cfg to at least get some level of controllability...
cl_sensitivity = 1.0  //mouse sensitivity, default = 1.0
Now that your on emulator controls, you can do nifty things like be able to toggle between different sensitivities on the fly and such. You are also immune to all the pitfalls of bugs/support everytime there is an update to the UI.

From spending far too much time on the forums, most folk assume there is some level of support or tuning that is supposed to happen on the development side of MWO that will then magically make joysticks equally competitive with a mouse. It's a myth, and I have identified what I feel are the underlying reasons why. In short what it breaks down to is not that a joystick is poorly suited for MWO, it's that an airplane joystick is poorly suited for MWO, which is unfortunately the only kind of joystick one can buy at the moment. The problem has everything to do with the opposing design criteria of a stick that controls something that moves like a mech/tank, which was designed around/benefits from unrestricted absolute inputs vs flying an aircraft, which is in turn designed around/benefits from restricted relative inputs. Comparing a mouse first to a normal airplane joystick, then to your joystick, then to my own stick:


-attributes (effects)

Mouse:
-moves in x/y Cartesian Plane (natural range of movement of the reticule)
-has no spring centering or detents (nothing fighting inputs, unrestricted movement)
-no deadzones (always in control, precision across centers no different than any other point in x/y)
-uses absolute inputs (obvious and easy to control, very precise)

Airplane stick:

-moves in pitch/roll (unnatural range of motion, not reflexively intuitive)
-spring centering, most likely with detents (fights inputs, negative interactions across the axes center's)
-requires deadzones (distracting, imprecise, disconnected feeling between inputs and in-game reactions, wasted range of motion)
-uses relative inputs (mech movements wind up either too slow or uncontrollable, difficult to not to overshoot past target, combined with the first 3 problems adds up to a non-viable control option)

Metal Patient's stick:

-moves in pitch/roll (unnatural range of motion, not reflexively intuitive)
-spring centering, most likely with detents (fights inputs, negative interactions across the axes center's)
-requires deadzones (distracting, imprecise, disconnected feeling between inputs and in-game reactions, wasted range of motion)
-uses relative absolute inputs (mech movements wind up either too slow or uncontrollable, difficult to not to overshoot past target, combined with the first 3 problems adds up to a non-viable control option obvious and easy to control, very precise)


Loc Nar's MWO stick
-moves in zenith/azimuth (pitch/twist -natural range of movement of the mech)
-no spring centering/detents, but has tensioned/greased rubs (nothing fighting inputs, unrestricted smooth damped movement, holds position when not being moved therefore maintains physical orientation related to on-screen state)
-no deadzones (always in control, precision across center no different than any other point in x/y)
-uses absolute inputs (obvious and easy to control, very precise)

#85 MentalPatient

    Member

  • PipPipPipPipPip
  • 145 posts

Posted 23 April 2013 - 12:32 PM

Thanks for your post loc nar, very informative. Currently with my stick setup, my stick has no actual spring as it is a ffb stick, and being ffb, I have the option to tune my settings to some extent as the g940 ffb is customisable. I hoping I can find a setting which might allow something remotely similar to what you describe. If not, I might use have to just remove ffb altogether and use it with no spring, which is fairly unstable when trying to keep the input still. I don't really want to pull my stick apart and alter the internal mechanisms as I also use it for flight simming.

#86 MentalPatient

    Member

  • PipPipPipPipPip
  • 145 posts

Posted 24 April 2013 - 05:14 PM

ok, so I have been playing around some more with settings for my joystick force feedback, joystick sensitivity, and game settings and sensitivities and I have stumbled upon somewhat of an increase in performance for my joystick aiming setup. The limitation I had before was, that as I am using absolute movement for my joystick now, it was somewhat limited in x axis torso movement, to the point where it would not travel its full range. I thought this might be due to the Joystick Mouse tool program I was using to emulate the absolute values for the stick, as its input only goes to the edge of the screen for instance in windows, I figure that it wasnt sending movement values when it reached that point. So it was a toss up between sensitivity of joystick movement - the higher the sensitivity, the further I could move the stick and torso, and torso travel. Unfortunately higher sensitivity means lower precision.

As this was causing a problem, I had pitch and torso movement of the regular relative movement bound to a ministick on my joystick, so that when I reached the edge of travel of my stick movement, I would then move the ministick to give more torso travel, however trying to couple this with aiming with the stick is difficult. Just for laughs I decided to bind the joystick x and y axis to pitch and torso twist, so that I would have both a relative movement from the typical joystick input, and an absolute movement from the joystick mouse tool.

WOW! I must say this setup really improves accuracy of the joystick as well as ability to place the cursor on the target, for the FULL range of torso travel. ;) Its a notable improvement over just absolute aiming, due to its limitations, at least for my stick. Now, when there is a target to the periphery of my torso, moving the joystick towards that direction I can still have the advantage of absolute precision of cursor movement, but can still reach the target for full torso deflection thanks to the added relative input. I had to adjust the relative input sensitivity to be fairly low, so it didnt interfere too much with the precision of the absolute joystick input. Here is my user.cfg file i am using:

Quote

## mouse sensitivity
cl_sensitivity = 0.005
## mouse acceleration off
i_mouse_smooth = 30
##i_mouse_accel = 0
i_mouse_accel_max = 0

gp_mech_view_look_sensitivity = 0.004 //Normal view
gp_mech_view_zoom_level1_sensitivity_modifier = 0.002 //Zoom 1.5x
gp_mech_view_zoom_level2_sensitivity_modifier = 0.005 //Zoom 3x
gp_mech_view_zoom_level3_sensitivity_modifier = 0.009 //Zoom 4x

i_joystick_buffered = 0


Sensitivity in my joystick profiler (mine is logitech) is set to 100%, so it does not apply any curves, aiding the absolute control. No deadzone either.

If anyone has a logitech g940 and wants to take advantage of the FFB (can remove spring centering while keeping a dampening force to resist movement from where your stick is currently positioned) here they are:

In specific game settings:
Overall effect strength: 150%
Spring effect strength: 0%
Damper Effect strength: 150%

Enable centering spring: disabled
Centering spring strength: 0%

Its still not perfect, i still need to play around with the sensitivities and possibly adding a deadzone for the relative input, so as to reduce interference with the precision of the absolute input. Feedback is appreciated.

Edited by MentalPatient, 24 April 2013 - 05:31 PM.


#87 F lan Ker

    Member

  • PipPipPipPipPipPipPip
  • 827 posts
  • LocationArctic Circle

Posted 26 April 2013 - 02:06 PM

S!

Interesting and good thread. I have been using Thrustmaster Warthog + CH ProPedals setup since day one when I got Founders status. I tried with mouse, just was not my stuff. The support for peripherals has gotten better but still a long way to go as we can see, needs file editing to get desired results and stuff.

Going to try out some tips and stuff seen in the thread to see how it will work. Thanks again for the tips!

#88 Ronin13m

    Member

  • PipPipPipPipPip
  • The Raider
  • The Raider
  • 136 posts
  • LocationNC

Posted 02 May 2013 - 11:06 AM

View PostMentalPatient, on 24 April 2013 - 05:14 PM, said:

ok, so I have been playing around some more with settings for my joystick force feedback, joystick sensitivity, and game settings and sensitivities and I have stumbled upon somewhat of an increase in performance for my joystick aiming setup. The limitation I had before was, that as I am using absolute movement for my joystick now, it was somewhat limited in x axis torso movement, to the point where it would not travel its full range. I thought this might be due to the Joystick Mouse tool program I was using to emulate the absolute values for the stick, as its input only goes to the edge of the screen for instance in windows, I figure that it wasnt sending movement values when it reached that point. So it was a toss up between sensitivity of joystick movement - the higher the sensitivity, the further I could move the stick and torso, and torso travel. Unfortunately higher sensitivity means lower precision.

As this was causing a problem, I had pitch and torso movement of the regular relative movement bound to a ministick on my joystick, so that when I reached the edge of travel of my stick movement, I would then move the ministick to give more torso travel, however trying to couple this with aiming with the stick is difficult. Just for laughs I decided to bind the joystick x and y axis to pitch and torso twist, so that I would have both a relative movement from the typical joystick input, and an absolute movement from the joystick mouse tool.

WOW! I must say this setup really improves accuracy of the joystick as well as ability to place the cursor on the target, for the FULL range of torso travel. :D Its a notable improvement over just absolute aiming, due to its limitations, at least for my stick. Now, when there is a target to the periphery of my torso, moving the joystick towards that direction I can still have the advantage of absolute precision of cursor movement, but can still reach the target for full torso deflection thanks to the added relative input. I had to adjust the relative input sensitivity to be fairly low, so it didnt interfere too much with the precision of the absolute joystick input. Here is my user.cfg file i am using:



Sensitivity in my joystick profiler (mine is logitech) is set to 100%, so it does not apply any curves, aiding the absolute control. No deadzone either.

If anyone has a logitech g940 and wants to take advantage of the FFB (can remove spring centering while keeping a dampening force to resist movement from where your stick is currently positioned) here they are:

In specific game settings:
Overall effect strength: 150%
Spring effect strength: 0%
Damper Effect strength: 150%

Enable centering spring: disabled
Centering spring strength: 0%

Its still not perfect, i still need to play around with the sensitivities and possibly adding a deadzone for the relative input, so as to reduce interference with the precision of the absolute input. Feedback is appreciated.

Mental Just want to say thank you. I also have the G940 and had shelved it since I could not figure out why I hated it for aiming. Just today started thinking about absolute aiming. I will have to try this set up and see if I can finaly play Mechwarrior as I always wanted to with a joystick in hand.

Thanks again Will let you know what I settle on for my calibrations incase you want to give it a shot.

#89 Viper69

    Member

  • PipPipPipPipPipPipPipPipPip
  • 4,204 posts

Posted 02 May 2013 - 11:32 AM

I shelved my stick,rudder and throttle as well. There is no simulator mode where you can enjoy not getting slaughtered by people using mouse and keyboard so I had to join the horde of mouse and keyboard users. Mechwarrior 4 had wonderful joystick support but for some reason its still lost on these guys or so far back on the burners as to not get any play.

#90 F lan Ker

    Member

  • PipPipPipPipPipPipPip
  • 827 posts
  • LocationArctic Circle

Posted 03 May 2013 - 02:08 AM

S!

It seems MWO's joystick "support" has an insane dead zone as default and I could not get rid of it even by editing user.cfg ;) How hard can it be to implement proper peripheral support with adjustment options?! I currently play with both mouse and joystick, brawl with stick and sniping with mouse :)

#91 Loc Nar

    Member

  • PipPipPipPipPipPipPipPip
  • 1,132 posts

Posted 03 May 2013 - 08:42 AM

Quote

It seems MWO's joystick "support" has an insane dead zone as default and I could not get rid of it even by editing user.cfg


Emulators are pretty much the only way to achieve a viable (functional, reliable, etc) MWO joystick scheme, and if your observations are accurate this is yet another reason why this is objectively true. It's nice to be unaffected the ever present bugs that seem to continually plague joystick users, seeing as every patch sets a different part of the analog controls on fire...

#92 Gremlich Johns

    Member

  • PipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 3,855 posts
  • LocationMaryland, USA

Posted 03 May 2013 - 02:02 PM

I do nothing special for my Logi extreme 3D Pro. My scores are decent and the JS makes the mech drive like a mech, not a battle armor suit.

#93 F lan Ker

    Member

  • PipPipPipPipPipPipPip
  • 827 posts
  • LocationArctic Circle

Posted 04 May 2013 - 12:25 PM

S!

The deadzone is there. I tried both online and in training grounds. The movement required to make any input even register is quite big. Mouse is instant. The movement required on throttle is also there. To add to the insult movement speed is also abysmal unless you adjust sensitivity ^_^ I have not fiddled with the TARGET software yet as ALL flight sims I have played with it have supported point & click config + plenty of response and other settings.

The Ask the Devs shed some hope that improvements are incoming in following patches, but until they REALLY put in a proper control panel to adjust response curves, dead zones etc. It is pretty useless as people with their varying taste of control response want to tweak and without the tools and only one "default" mode = meh.

Edited by F lan Ker, 04 May 2013 - 12:26 PM.


#94 Mechteric

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 7,308 posts
  • LocationRTP, NC

Posted 10 May 2013 - 05:12 AM

One thing I would like to be added is for the analog chassis turning action to also have separate analog turn left and analog turn right actions. This is because the pedals on my CH setup are actually different axes (one is X the other Y). So currently the only way I can use my pedals for turning is by using the slider as opposed to the pedals themselves (which can feel a bit awkward).

Posted Image

#95 MentalPatient

    Member

  • PipPipPipPipPip
  • 145 posts

Posted 10 May 2013 - 02:16 PM

View PostCapperDeluxe, on 10 May 2013 - 05:12 AM, said:

One thing I would like to be added is for the analog chassis turning action to also have separate analog turn left and analog turn right actions. This is because the pedals on my CH setup are actually different axes (one is X the other Y). So currently the only way I can use my pedals for turning is by using the slider as opposed to the pedals themselves (which can feel a bit awkward).

Posted Image

I currently use pedals to turn to, but I prefer the standard rudder turn setup as opposed to the actual toe brakes as pedals. I would imagine using separate axis to turn would create conflicts, as imagine if you put in a bit of left and right toe brake, it would either have to cancel each other out or would conflict. I think a single axis for turning is simpler and easier to use.

#96 Loc Nar

    Member

  • PipPipPipPipPipPipPipPip
  • 1,132 posts

Posted 10 May 2013 - 02:48 PM

Quote

I think a single axis for turning is simpler and easier to use.


It's actually not easier, at least not easier to use for controlling a mech in MWO (definitely easier to code for...). It is easier for an aircraft (pilot here...), which has quite different underlying control principals. In flight, rudder inputs are subtle, gradual, generally few/far between, and balancing the feet against each other for tactile reasons is a benefit for sustained maneuvers. For MWO however you wind up either tapdancing or undercontrolling if you need to use both feet each time you want perform a single action such as turning either left or right, which more than doubles the workload.

As for how the separate analog turn inputs would work, they would cancel each other out, but proportionately, not binary. Left pedal returns 0 to 255, right pedal returns 0 to -255 (*or left pot uses positive voltage, right pot uses negative, and MWO only sees the sum). Pressing both fully returns 0. pressing the right fully and the left halfway would be the same as holding half right pedal etc. It's actually not terribly complicated, although I might not be explaining it well...

For the time being, in my cockpit my own pedals are fixed to the floor, and I use the toebrake pots to generate 'a' and 'd' commands. I find that being able to discretely use my feet even for full right/left turn inputs is still easier than using a single central analog axis, since that requires both feet and MWO requires too many inputs for this to be ideal.

*this will be my hardware workaround for if this is not addressed soon

#97 Mechazoid

    Rookie

  • Knight Errant
  • 2 posts

Posted 10 May 2013 - 07:03 PM

This is sad... I can't find user.cfg file. Im in my mechwarrior online folder. I see system.cfg. systemoverride.cfg, etc. Please help...

#98 AztecD

    Member

  • PipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 656 posts
  • LocationTijuana. MX

Posted 10 May 2013 - 08:48 PM

View PostMechazoid, on 10 May 2013 - 07:03 PM, said:

This is sad... I can't find user.cfg file. Im in my mechwarrior online folder. I see system.cfg. systemoverride.cfg, etc. Please help...


C:\Games\Piranha Games\MechWarrior Online

#99 MentalPatient

    Member

  • PipPipPipPipPip
  • 145 posts

Posted 11 May 2013 - 01:50 PM

View PostLoc Nar, on 10 May 2013 - 02:48 PM, said:


It's actually not easier, at least not easier to use for controlling a mech in MWO (definitely easier to code for...). It is easier for an aircraft (pilot here...), which has quite different underlying control principals. In flight, rudder inputs are subtle, gradual, generally few/far between, and balancing the feet against each other for tactile reasons is a benefit for sustained maneuvers. For MWO however you wind up either tapdancing or undercontrolling if you need to use both feet each time you want perform a single action such as turning either left or right, which more than doubles the workload.

As for how the separate analog turn inputs would work, they would cancel each other out, but proportionately, not binary. Left pedal returns 0 to 255, right pedal returns 0 to -255 (*or left pot uses positive voltage, right pot uses negative, and MWO only sees the sum). Pressing both fully returns 0. pressing the right fully and the left halfway would be the same as holding half right pedal etc. It's actually not terribly complicated, although I might not be explaining it well...

For the time being, in my cockpit my own pedals are fixed to the floor, and I use the toebrake pots to generate 'a' and 'd' commands. I find that being able to discretely use my feet even for full right/left turn inputs is still easier than using a single central analog axis, since that requires both feet and MWO requires too many inputs for this to be ideal.

*this will be my hardware workaround for if this is not addressed soon


That's interesting loc nar. I personally find using the single axis of the pedals perfectly adequate. But you could say I'm not the normal kind of pedal user as I have been virtual formation flying for years, so they've had some practise.

#100 Mechazoid

    Rookie

  • Knight Errant
  • 2 posts

Posted 11 May 2013 - 05:12 PM

View PostAztecD, on 10 May 2013 - 08:48 PM, said:


C:\Games\Piranha Games\MechWarrior Online


Okay, I have C:\Program Files (x86)\Piranha Games\MechWarrior Online

No USER.CFG is located inside this folder, or any of the sub folders inside?





6 user(s) are reading this topic

0 members, 6 guests, 0 anonymous users