Re: Bokeh in 3d

[Greg Erker <erker@xxxxxxxxxxxxxx>]

>I very recently (last week) bought one of the new Voigtlander Bessa-L
>cameras with a 15mm/F4.5 Heliar lens and tried some slide-bar stereo
>shots.  The lens is incredible - 110 degree field of view
>(corner-to-corner), no detectable distortion, crisp (even in the corners
>of 11x14" prints, though the extreme corners of the full film frame are
>a little soft), and amazing depth of field.  The lens will focus as
>close as 0.3m.

  Very cool camera. Actually the lens moreso.

>The good bokeh in the center of the frame is due (I feel) mainly to an
>aperture with 10 blades.  Out-of-focus highlights tend to take on the
>shape of the aperture and in this case it is nearly circular (the reason
>many Leica lenses are highly praised for their bokeh).

  Aperture blades do have an effect but they
aren't the entire reason. Some lenses have
good bokeh wide open and some have bad bokeh
wide open. In both cases the lens opening is
perfectly round.

  From what I have read the bokeh is different
for near versus far out of focus images. IE a
lens with good bokeh on infinity objects may
be unpleasant for objects in front of the
plane of focus.

  Nikon make a couple of lenses with Defocus
Control (DC). In the neutral setting out of
focus highlights (OOFH) will be evenly lit across
them. ASCII cross section drawing below:
   ___
  |   |
__|   |__

  When the defocus control is set to rear
then far away OOFH will have brighter middles
than edges (this is good because it helps
to blur thing together more):
    _
   / \
__|   |__

  But the OOFH closer than the plane of
focus will look like this:


  |\_/|
__|   |__

  Brighter at the edges and dimmer in the
middle. This tends to give you double edges
on tree branches and other blurred lines.

  When the defocus control is set to front
then the near OOFH have soft edges and the
far ones are sharp edged.

  Even in non-adjustable lenses with good
bokeh, I hear, that it tends to be good
at near or far and less good on the other
side of the plane of focus.

  Gary's Rolleidoscop seemed to be pretty
neutral with evenly blurred tree branches
in front of and behind the plane of focus.

Greg