Angus Dorbie (dorbie++at++sgi.com)
Mon, 13 Sep 1999 11:59:05 -0700
Normally you do not adjust the ipd.
The operation effectively scales the perceived size of objects in the
world. In otherwords, with a single eyepoint you really have no world
scale. When you have stereo the virtual ipd determines the world scale.
Normaly the virtual ipd should match the viewer and the frusta should in
turn match the relationship of the real world eyes with the display.
When you zoom in on a sand table or similar device there is a notion of
the scale you draw things at. You should recognize that there is an
correlating change in the view matrix operation and the sand table has a
dimension in the real world. It is straightforward to determine the
scale of the projected model on the sand table and adjust the ipd
accordingly. This is not a conventional style of navigation so it may
take some work to really understand what's going on. You could do this
with fixed IPD and different midelview but the terrain would appear far
below the table surface and there would be insufficient parallax to give
information on terrain relief, it would be more like being in real sized
world.
The more conventional style of stereo is where the ipd is fixed and
determined by the scale of objects in the world, this is by far the more
common method where you should attempt to match the real world ipd in
the virtual environment. If you get the frusta and view matrix correct
then you may still get problems if you bring virtual objects too close
to the eye, but in general it should be the most robust approach.
Cheers,Angus.
Guenter Schreiber wrote:
>
> Hello eveerybody,
>
> as I have seen in the slideset of
> http://www.sgi.com/software/performer/presentations/geometry/sld012.htm
> there is some interpupil distance correction proposed, depending on the
> scene size you are currently
> observing.
> So my question is: I have a big scene, and need while overflying a
> stereo impression.
> But as I come close, that interpupil distance is too big for proper
> stereo impression.
> Till now, I need to adjust the interpupil distance manually.
>
> So is there some algorithm, which adjusts this distance automatically.
> For a fast implementation, I thought, I could use the "ray-picking" with
> a diameter of 0.5 or bigger.
> So the closest object will determine the eye distance via some
> heuristics.
>
> Anybody experiences?
> How is the performance on the useage of the closest distance obtained by
> raypicking?
> Or should I use some results of the cull process ?
> Which interpolation function should I use for coming close - linear -
> quadratic ?
> Anybody sample code ?
>
> Thanks in advance
>
> Günter
>
> --
> Enjoy life -- enjoy juggling - God gave you that creativity.
> ...
> o o . .
> O o o O . O .
> |/ \| \|/
> | | |
> / \ / \ / \
> ______________________________________________________________
> | Dipl.-Ing. Günter Schreiber | phone: *49/*8153/28/1183 |
> | German Aerospace Center | fax: *49/*8153/28/1134 |
> | Institute of Robotics and | email: | /|
> | System Dynamics | Guenter.Schreiber++at++dlr.de
> --------
> | P.O. Box 1116 | priv: / | |
> /
> | 82230 Wessling | Katja_Guenter.Schreiber++at++gmx.de ---+-+--
> |________ http://www.robotic.dlr.de/Guenter.Schreiber _________| |/
>
> It is unlawful to use this email address for unsolicited ads
> (USC Title 47 Sec.227). I will assess a US$500 charge for
> reviewing and deleting each unsolicited ad.
> -----------------------------------------------------------------------
> List Archives, FAQ, FTP: http://www.sgi.com/software/performer/
> Submissions: info-performer++at++sgi.com
> Admin. requests: info-performer-request++at++sgi.com
--
"One of the best-known folk theorems of software engineering is that
60% to 75% of conventional software projects are either never
completed or rejected by their intended users. If that range is
anywhere near true (and I've never met a manager of any experience
who disputes it) then more projects than not are being aimed at goals
which are either (a) not realistically attainable, or (b) just plain
wrong."
Eric S. Raymond - The Cathedral and The Bazaar
For advanced 3D graphics Performer + OpenGL based examples and tutors:
http://www.dorbie.com/
This archive was generated by hypermail 2.0b2 on Mon Sep 13 1999 - 11:59:12 PDT