P3D Re: "Bowl" distortion in viewers

[erker@xxxxxxxxxxxxxx (Greg Erker)]

>Just consider that the maximum separation between the near point
>and infinity is only 1.2 mm, which is just a small fraction of
>the 20-30 mm diameter of the lens.

  True, but consider that you can detect a tree at 100'
as separate from the infinitely far background. What
fraction of a mm shift is that? (John B will tell us
if we aren't careful :)

  The the lens distortion could bend the edges of the
"flat" test image in from infinity. Not as far as 6'
(1.2mm shift) but noticably to someone sensitive to it
(IMO).

>Also, such small 3-d distortions
>would be impossible to detect in an image with full range depth.
>Larry B. himself talked about seeing it (easier?) in 2D images.

  Yes it should be masked in a normal 3D image
(again IMO).

Greg E.



------------------------------