full head translation

gaming247
Posts: 35
Joined: Fri Jan 10, 2003 5:00 am

Re: full head translation

Post by gaming247 » Thu Feb 20, 2003 6:48 am

What about FS2002. Cockpits in Fs2002 are fully 3D. (most of them) You can also lean left and right, move forward and backward, up and down.

@ Jim. You should be able to track a 6DOF movement with the trackIR by using 3 dots. The dots would have to be placed like this:

...........O
........O....O

(will make a more detailed screenshot if u like )

mathematically it would be the projection of a accompanying tripod (don�t know what it is in english, in german it�s called Begleitendes Dreibein). It is used for describing the position on Space curves. ( e.g to control robot movement)
imagine an axis going through each dot.
it should be possible to track every movement by using three dots. u can also use four. one dot is the origin.

........O
........O <--- Origin
.....O...O

for an example look here: http://www.math.tu-dresden.de/~klix/Dif ... ibein.html

Jim if you are really interested i would like to hear your thoughts on this. I am student of mechanical engineering and would like to see what can be done with your device.

Jim
NaturalPoint Employee
NaturalPoint Employee
Posts: 1395
Joined: Mon Oct 14, 2002 5:00 am
Location: Corvallis, Oregon
Contact:

Re: full head translation

Post by Jim » Thu Feb 20, 2003 8:40 am

Hello:

Gaming247, I am very interested in this method of control. We have the ability to track many different points at once, so deriving the full 6DOF should not be a problem. I would be interested in working with you on the math, email me at jim@naturalpoint.com. It might take me a couple of days to get back to you on this as I am very busy right now.

I think that we will use MSFS2K for our technology demo platform for 6DOF. I am also planning on making a system for use with HMD's, that is all optical.

gaming247
Posts: 35
Joined: Fri Jan 10, 2003 5:00 am

Re: full head translation

Post by gaming247 » Thu Feb 20, 2003 9:02 am

Afaik optical tracking devices for hmd already exsit, but in a higher price segment.

there is only one major problem i see on getting the 6dof to work.
This comes from the limitation of the trackIR hardware, as it online scans the dot in 2D.
So tracking the distance between the unit and the dots will be a problem.

There are only a few ways to overcome this:

u need a second unit.

u track the size of the dots (but u already said that the resolution for this would be too low)

any other way ??

Maybe u can track the amplitude of the IR wave that is reflected by the dot. When the dot moves the amplitude should change.
Don�t know if that is possible with the unit but i guess not.

If u want to discuss this things and your ideas , feel free to mail me.
Also i would like to join your beta tests if u like.
I will send you some sketches or pics when i have time, as i am busy right now myself. Just remind me if i forget.

RSJeff
Posts: 27
Joined: Wed Jan 22, 2003 5:00 am

Re: full head translation

Post by RSJeff » Thu Feb 20, 2003 11:57 am

When I set up my TrackIR for FS2002, I tried to map the "change vantage point" controls to my joystick hat- in default, these commands are "shift+enter", "shift+backspace" etc, and they do allow you do move all around the virtual cabin with appropriate changes in the perspectives of the cockpit model. In other words I wanted to use the hat to sit up in my seat, or move closer to the window, and then utilize the TrackIR to look around - to add a 4th dimension as someone mentioned. I tried to change the default key assignment settings so that instead of changing my view (which I now do with TrackIR) my hat would now change my vantage point, but this didn't work -I'm not sure why. I'm still working on it, and it seems like a relatively easy way to allow even more realistic freedom of motion while flying. After all, until recently you were using your joystick thumb to "look around", so for now at least it wouldn't be that difficult to get used to using your thumb to "move around". BTW, I'musing a MS FF 2 joystick.

Siggi
Posts: 42
Joined: Fri Feb 07, 2003 5:00 am
Location: UK

Re: full head translation

Post by Siggi » Thu Feb 20, 2003 12:31 pm

Two TIR's would be the way to go, so you could kind of triangulate the head's position. I think radio would even be more precise than IR radiation? Certainly it would overcome the limitations of both range and FOV.

honda346
Posts: 40
Joined: Sun Feb 09, 2003 5:00 am

Re: full head translation

Post by honda346 » Thu Feb 20, 2003 5:39 pm

So Jim how much evangalizing are you all doing with the game companies for the full 6 DOF in upcoming games? Do you just have phone numbers with cold calls or are you staffed well enough to give them technology demos to prove the point, et?

While the other guy was thinking Jim Morrison I had the spoiled girl in Willy Wonka come to mind singing "I want it now!"...

gaming247
Posts: 35
Joined: Fri Jan 10, 2003 5:00 am

Re: full head translation

Post by gaming247 » Thu Feb 20, 2003 10:16 pm

with radio u always have the problems that it has to be very silent in order to work correctly otherwise u won�t get a clear signal.
IR Waves are much better. The FOV shouldn�t a problem when using active leds.

Charlie251
Posts: 955
Joined: Fri Jan 03, 2003 5:00 am
Location: Melbourne, Florida

Re: full head translation

Post by Charlie251 » Fri Feb 21, 2003 5:38 pm

I thought of the two TIR idea but it is to tedious and you would have to have an extra spot to put one of them. Then you have to worry about centering two dots perfectly instead of only one. To much hassle. A toggle switch to change from translation to trasaction would be better.

Active LED's have their own quirks as well. Right now I move forward and backwards to give myself more throw or less when I need it (closer the dot, more the throw) and I do not beleive an ALED will give this feature as it will paint the same size target on the unit regardless of how close you are. Mainly because were talking about a beam now instread of a reflection.

Using RF instead of IR is a VERY bad idea. Any kind of stray RF near a computer is bad. Especially for those of us with amatuer radio equipment on the desk next to us. Then you have baby monitor, wireless lans, bluetooth, ect. Your just asking for alot of interference. You could argue that there is enough bandwidth open to find a slot, but really there isnt. Especially in a frequency block where you going to want this to run. Besides, you still have to sovle the idea of how to track head movement using RF instead of IR. Very tough there. Finally, you have the fact that a cell phone next to my head is about as risky I am going to get with pumping RF into my brain cavity.

Jim
NaturalPoint Employee
NaturalPoint Employee
Posts: 1395
Joined: Mon Oct 14, 2002 5:00 am
Location: Corvallis, Oregon
Contact:

Re: full head translation

Post by Jim » Sat Feb 22, 2003 1:04 pm

Hello:

I mentioned some more 6DOF ideas in another thread just now, forget which one it was.

The methods that we are working on involve the use of 2 cameras in a new case, or more dots in a fixed patern or some combination there of. We have a new method for doing this that allows for enough resolution and without bulking up the system. It should work well. As mentioned, I will give it a try in FS2002 first. If I really think it improves the experience then we can move forward. I think that it might be very realistic, but it could distract the user instead of enhancing the experience. There is a fine line there.

Post Reply