Author Topic: head tracking device for input  (Read 1509 times)

Become dust

  • Sr. Member
  • ****
  • Posts: 255
    • View Profile
Re: head tracking device for input
« Reply #30 on: November 01, 2017, 03:21:03 PM »
I will figure it out together with Dan from EdTracker who can do wonders

Not sure if he can do much if the problem is GPB being not capable of seeing two different edtracker devices at the same time.
Maybe you should use vJoy to remap the two EdTrackers to two "joysticks", as GPB normally handles well multiple input devices at the same time.

When I wanted to try to use the edtracker as input device (for rider lean or even for bike lean) I wanted to fine tune the response curves so for me it wasn't enough to use edtracker as is (seen by gpb as a joystick). I wanted to use Joystick Gremlin to alter the curves and this makes me use vJoy. Bit overcomplicated, but it works. So you may be able to use vjoy (you don't strictly need joystick gremlin, unless you want to fine tune the curves etc) to "remap" the two edtrackers to two new joysticks and assign them to functions in GPB.

Actually I think what you need is to to have 1 edtracker remapped to a joystick (for body movements) and one edtracker for head tracking (no need for vjoy here, just opentrack).

That's an interresting way of going about. 

poumpouny

  • Sr. Member
  • ****
  • Posts: 266
    • View Profile
Re: head tracking device for input
« Reply #31 on: November 02, 2017, 05:32:57 AM »
Become dust > can you please explain how you get it to work ?

I think the best solution is to have a 6Dof head tracking, where the gyro rotation controll the view and the translation controll the rider movement, i don't know if such a head tracker exist !

doubledragoncc

  • Hero Member
  • *****
  • Posts: 4596
  • If it dont fit HIT it!!!!!!!!!
    • View Profile
    • GPBOC Live Streams
Re: head tracking device for input
« Reply #32 on: November 02, 2017, 08:11:52 AM »
PP you cant have one device to give 2 seperate movement groups. You HAVE to have 2 seperately mounted devices giving 2 seperate sets of inputs or it gets messy lol.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls

Teamspeak: Server IP: 82.43.148.211  user password "yougetin"

i5 7600 5.0GHz z270 ASUS Maximus Code Mobo 16GB 3866MHz DDR4 RAM ASUS Strix GTX1050Ti OC 4GB DDR5 Kraken X62 AIO Cooler

HornetMaX

  • Hero Member
  • *****
  • Posts: 5864
    • View Profile
Re: head tracking device for input
« Reply #33 on: November 02, 2017, 08:26:21 AM »
PP you cant have one device to give 2 seperate movement groups. You HAVE to have 2 seperately mounted devices giving 2 seperate sets of inputs or it gets messy lol.
If the device can be "seen" by vJoy, you can. But I'm not sure 6dof devices (there are some, e.g. TrackIR) can be seen by vJoy: most of the time their have a proprietary interface.

doubledragoncc

  • Hero Member
  • *****
  • Posts: 4596
  • If it dont fit HIT it!!!!!!!!!
    • View Profile
    • GPBOC Live Streams
Re: head tracking device for input
« Reply #34 on: November 02, 2017, 09:22:34 AM »
No Max it does not matter about IF it can do the axis its about your head has to me tracked and so does the body but they move individually and therefore need a seperate input device otherwise you are getting unreal input. Hard for me to explain but you cant have ONE device for both input groups. One device will cause the input to GPB to be false due to how the head and body act seperately, yet together. Its the same as why nobody has built a realistic motion system for a bike as it also needs to groups of motion due to the why a bike works.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls

Teamspeak: Server IP: 82.43.148.211  user password "yougetin"

i5 7600 5.0GHz z270 ASUS Maximus Code Mobo 16GB 3866MHz DDR4 RAM ASUS Strix GTX1050Ti OC 4GB DDR5 Kraken X62 AIO Cooler

HornetMaX

  • Hero Member
  • *****
  • Posts: 5864
    • View Profile
Re: head tracking device for input
« Reply #35 on: November 02, 2017, 09:40:54 AM »
A 6dof device is in fact 2 3dof devices. The advantage is, theoretically, that it should handle better the interactions between the 3 translations and the 3 rotations.
Another difference is that 2 edtrackers will track 6 rotations while a 6dof device is 3 rotations and 3 translations.
 
But your head and your body in reality they don't move individually (independently): if you lean your body left, yor head leans too (unless you compensate, of course).

doubledragoncc

  • Hero Member
  • *****
  • Posts: 4596
  • If it dont fit HIT it!!!!!!!!!
    • View Profile
    • GPBOC Live Streams
Re: head tracking device for input
« Reply #36 on: November 02, 2017, 10:00:30 AM »
Half of what you say is true Max, the problem is that there will be conflict if you use one device as a body movement in a certain axis will be read as a head movement by the device and so will be inputed to gpb when in fact you have not moved your head in that direction so you get a fauls rinput in gpb. It really is simple but hard for me to write. I have been working on this for some time and beleive me ONE device will give conflicting inputs. A 6DOF sounds correct in thinking but the practical use is in fact wrong. You need a device connected to the head and a seperate one for the body. Try leaning right from the waste and then leaning you head left at the same time with ONE device, how are you going to get 2 inputs that use the same axis on ONE device? Think about it a biy.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls

Teamspeak: Server IP: 82.43.148.211  user password "yougetin"

i5 7600 5.0GHz z270 ASUS Maximus Code Mobo 16GB 3866MHz DDR4 RAM ASUS Strix GTX1050Ti OC 4GB DDR5 Kraken X62 AIO Cooler

HornetMaX

  • Hero Member
  • *****
  • Posts: 5864
    • View Profile
Re: head tracking device for input
« Reply #37 on: November 02, 2017, 11:00:02 AM »
Try leaning right from the waste and then leaning you head left at the same time with ONE device, how are you going to get 2 inputs that use the same axis on ONE device? Think about it a biy.
They don't use the same axis. A 6dof thingy tracks the position and orientation of your head.
Now one valid point that is against using that kind of device for our needs is that what you want to put in place is not the tracking of position and orientation of the head, but more something like "orientation of the head + position of the ass" (or body :) ). That because this is how GPB wants things to be.
So yeah, for that a 6dof device is not ideal. But the reason it's not that "it's one device" or "you have 2 inputs on the same axis".

speedfr

  • Sr. Member
  • ****
  • Posts: 327
  • Partir, c'est pourrir un pneu !
    • View Profile
Re: head tracking device for input
« Reply #38 on: November 22, 2017, 01:30:57 AM »
Hello Riders,

Very interesting posts in here  :)

I use a EDTracker Pro (USB) and i've made several videos about using it on my YT stuff (Assetto Corsa, Star Citizen, DCS -F15C and ArmaIII).
I just came back to this Gp Bikes project (i think i have alpha1 and 3 still on my disk).

Need to sort out the feeling with the game (can't play with a pad, i'm too old) but i tried EdTracker and i works like a charm, i just need to get to the lean back and fore stuff.
i used the same profiles as DCS and activated the smooth option settings in GPB, and as written above no need for Opentrack or FaceTrackNoIr.
Will see later if i'll make a video like i did for other simulators, if needed  :) but i've read that doubledragoncc was going to do one.
Need to check if we can keep the Lean and using the Freelook option at the same time or maybe its a stupid idea...  ::)


But yes, headtracking is a must have for this nice simulator.

Missing Gp500 (Microprose)

Testing EDTracker Pro on YT

i7-4790K/16Go/GTX1070/1440p/Full WC