Author Topic: trackIR problems  (Read 444 times)

Offline Urchin

  • Platinum Member
  • ******
  • Posts: 5517
trackIR problems
« on: February 01, 2008, 09:11:56 PM »
Got a couple issues with trackIR still.  Had it about a month, I'm starting to get used to it finally.  

One huge issue is my POV seems to move "away" from where ever I am looking.  Backwards.  So if I look straight up (which with a hat is a clear view in almost every plane, or can be made to be), it is almost like my guy lays down in his seat and then looks up.  If I look out the left side, it is like my guy backs up against the right side of the cockpit, then looks left.  Is there any way to stop this from happening?  I've tried my own profile and Mace's (I did look around on the forums before I asked this) - both profiles do the moving backwards thing.. it is really irritating.  

Another problem I have is that the thing seems to lose tracking when I look to the left sometimes.  Is there anything I can do to increase the sensitivity of it so this does not happen?  

Any help is appreciated, tons.  Thanks in advance.

Offline moot

  • Plutonium Member
  • *******
  • Posts: 16333
      • http://www.dasmuppets.com
trackIR problems
« Reply #1 on: February 02, 2008, 12:04:12 AM »
Did you check if it's not the head position for each cardinal snap-mode view that your TrackIR is going to?

You'd probably get more responses in the Hardware and Software forum.
Hello ant
running very fast
I squish you

Offline Octavius

  • Skinner Team
  • Platinum Member
  • ******
  • Posts: 6651
trackIR problems
« Reply #2 on: February 02, 2008, 02:04:08 AM »
Ah crap, Urchin you have the same problem as me.  To be honest I never pinpointed a source, but I following may variables may be the culprit:

1) orientation of the tracking camera - mine used to slide around due to lack of USB cord length... until I taped it down.

2) inconsistent location of head - my chair is on wheels and can't seem to 'find the sweet spot' in relation to the camera every time.

3) position of hat and clip - the clip would fly off at times and it would not be centered on the brim of my hat.  Also the centering of my hat would also change, so the clip position would again become incorrect and off center.

So basically the setup would have very minor alterations each time I got settled in to fly.  These alterations would cause huge annoyances.  I discovered the comfortable position of a hat on my head never really had the center of the brim in line between my eyes and on the center of forehead, so that was a starting point.  It drove me nuts for a while until I realized there was also a big difference in the camera position as well as my seat position in addition to the hat/clip orientation.  

I had to get used to the software update a few years ago (2004?), and after that it never felt like my yaw and pitch were moving with respect to pivot point, rather it felt as if this point turned into a point along a radius in front of the point, causing annoyance.  And more recently it became asymmetrical and inconsistent -- I would look right and it felt very good, very natural -- then I would look left and it would do the 'back up into cockpit glass' deal.  If any of the aforementioned variables were not right, I discovered, the 'pivot around a radius' would be exaggerated along one half of an axis.

I never did find a method to finding my sweet spot.  Every so often I'd sit down in a particular spot, adjust my hat, adjust my clip, and if need be, adjust the camera position on top of my massive CRT (I now have an LCD and it no longer moves), and everything would work out.  I can only suggest messing around with every variable until you find something that works good for you.  Once you find it - take a sharpie or tape and mark every position -- your hat, clip position on hat, camera position, chair position, position relevant to moon, position relevant to sun, position relevant to pluto, etc.  Then simply duplicate the process each time you fly. :D


Sorry for the wall-o-text, but it really seems like it's hit or miss with this stuff.

[edit:  Sit down for big adjustment sessions with AH closed.  Use the 'view head' to compare all six axes, you'll be amazed at how off it is between physical adjustments in front of the camera.]
« Last Edit: February 02, 2008, 02:12:56 AM by Octavius »
octavius
Fat Drunk BasTards (forum)

"bastard coated bastards with bastard filling?  delicious!"
Guest of the ++Blue Knights++[/size]

Offline Old Sport

  • Nickel Member
  • ***
  • Posts: 530
trackIR problems
« Reply #3 on: February 02, 2008, 11:06:17 AM »
Try Clipboard ->Options->Preferences->View Options-> TrackIR Object Relative Move

When I first installed TIR it did what you describe, i.e. put you in the pan of the seat when you look up, pushing your POV opposite from the direction you are looking, etc. This setting changed the display to more like actually being in a cockpit.

In addition to what Octavius says about getting aligned with the sensor, you might also recenter TIR (F12 key I think) on a regular basis. I've got recenter mapped to my stick and I recenter a lot.

Also, TIR seems to work best for me at a distance of 3+ feet from sensor to my clip (i.e. not too close), and I've got my sensor elevated above my monitor a few inches so that it is level or above the level of my clip. Hope that helps.

Mace's profile should be aggressive enough. I think the more aggressive, the better, i.e. the more TIR turns the virtual head per degree of real head turn the better.

All the best.

Offline Urchin

  • Platinum Member
  • ******
  • Posts: 5517
trackIR problems
« Reply #4 on: February 02, 2008, 11:52:17 AM »
Putting in a large dead zone on the Z axis helped somewhat.

Offline FireDrgn

  • Silver Member
  • ****
  • Posts: 1115
trackIR problems
« Reply #5 on: February 02, 2008, 02:31:22 PM »
Other light sources can affect trackir if your useing pro clip....
"When the student is ready the teacher will appear."   I am not a teacher.

Offline FireDrgn

  • Silver Member
  • ****
  • Posts: 1115
trackIR problems
« Reply #6 on: February 02, 2008, 02:32:10 PM »
also i find that not using the roll axis helped the most
"When the student is ready the teacher will appear."   I am not a teacher.