Hard Light Productions Forums
Modding, Mission Design, and Coding => FS2 Open Coding - The Source Code Project (SCP) => Topic started by: Dragon on July 29, 2011, 04:30:49 pm
-
Recently, I've managed to get FaceTrack to run on my computer, and, of course I'd like to use it with FS2Open.
The problem is, it only worked when set on "fake TrackIR" mode and even then, my view was always panned fully to the right and somewhat down, my head movement didn't seem to change it. Does anybody know what I could be doing wrong? My camera is placed below the monitor, a little to the left (the image of my head is fully centered in FaceTrack preview).
-
Perhaps, there is some light that confuse your camera.
-
Nope, in ArmA II it works well, though I need to adjust it.
On the other hand, ArmA II uses FreeTrack protocol, and I didn't change the setting between games, just the protocol.
-
Can we get a merge of sorts here? There's been many threads on FaceTrack Dragon, another one of which was your own!
http://www.hard-light.net/forums/index.php?topic=75274.msg1490214#msg1490214
http://www.hard-light.net/forums/index.php?topic=76615.msg1521303#msg1521303
-
I didn't notice your thread and didn't want to necro my own.
Maybe it should indeed go into some other board though, like GenFS, as it doesn't seem to be getting much attention here.
-
I got it working... Did you update from v1.30 to v1.50?
-
Yes.
Could you attach an .ini file with your settings?
Also, where is your camera positioned?
Mine is below the monitor, I've seen that TrackIR camera is mounted on top.
-
My ini is the same as the default, but when I get back from work in a few hrs I will post it just in case it isn't. I had the same problem before updating to v1.50 and then updating the filters (EWSA?). My camera is above my monitor. I would try it above just to see if that changes anything.
-
might be an interface issue. natural point has gone to great lengths to make sure that its proprietary, closed source, interface not be used by other devices/software. they even prevent you from using trackir hardware with free software like freetrack (though i do happen to have a copy of the version that supported it if anyone wants it). so long as head tracking and motion controllers have closed source interfaces were gonna have problems like this. thing i like about freetrack is it supports more interfaces, like mouse axes, joystick axes via ppjoy, its own more open interface, and trackir emulation mode. im not sure what kind of interface(s) your software supports, but id hope it has some kind of joystick emulation. we will also at some point want to consider other devices like kinect. then its just a matter of supporting freelook by axis, which should support everything out of the box including trackir (through tir2joy).
you dont really need to remove existing support, or change it in any way. you do however need multiple joystick support and available axes to bind them to. this will likely be covered on the move to pure sdl for input on all platforms. after that it should be trivial to add the neccisary 6 freelook axes. this also allows players to use extra joystick (maybe mouse) axes to make use of freelook features in the absence of head tracking hardware or software.
-
Of course, it'd be great of FreeTrack protocol was supported by FSO, I think that this should be doable.
Multiple controllers would also be an amazing addition, but I'm afraid it's not gonna happen anytime soon.
-
perhaps theres the possibility of swapping out the scptrackir.dll with an alternate interface dll. so long as all the calls and returns are identical should work.
-
I just got everything set up to 1.50 and the view sticks facing up and left, even when positioning myself to get the X and Y / rotX and rotY values between 0.5 and -0.5.
The camera is ontop of the monitor.
And oddly enough it suddenly starts working, though the lagtime between turning and the sheer sluggishness of the camera turning to match makes it not worth using.
-
I just got everything set up to 1.50 and the view sticks facing up and left, even when positioning myself to get the X and Y / rotX and rotY values between 0.5 and -0.5.
The camera is ontop of the monitor.
And oddly enough it suddenly starts working, though the lagtime between turning and the sheer sluggishness of the camera turning to match makes it not worth using.
That was basically my experience. Eventually it just started working (more or less), but the imperfect tracking (and especially the lag) made it not worth using.