Re: [photo-3d] viewer optics

[Paul Talbot <list_post@xxxxxxxxx>]

Re: Ferwerda's comment that 'the eccentric viewing may give
an unfavorable effect when the lenses are of a very simple
quality,' Peter Abrahams (2001-03-08) suggests:

  | the effect would very likely extend to other types of
  | lenses than 'very simple', especially when they are f2.

I don't suppose the 'official' stereo glossary defines either
'very simple quality' or 'unfavorable effect," so it is difficult
at best to reach a consensus on the point.  With his binocular
and telescope background, Peter no doubt is aware that lenses
vary widely in performance; a descriptive term such as 'achromat'
is not sufficient to assure a quality viewer lens.  So I relate
a tale of some experiences I've had with medium format viewer
lenses:

  - I received a pair of 50mm diameter PCX lenses to evaluate for
    a less expensive version of the SaturnSlide viwer.  While
    surprisingly good for a single element lens, they clearly
    had 'unfavorable effects' when used in a fixed interocular
    viewer in front of my narrowly spaced eyes.  These lenses
    would be well suited to a viewer such as Alan described as
    a good use of interocular: one with variable spacing of both
    the L/R images and the lenses.

  - I have used a MF 3D viewer prototype that uses achromatic
    loupes and found it had poor optical performance.  I also
    noted that moving my eyes away from a very small 'sweet spot'
    in the center of the lenses resulted in rapid deterioration
    of the image quality.

  - I have tested several achromatic lenses in the range of
    40mm (+/-) diameter and 80mm (+/-) FL and found a wide
    range of performance.  I remember one in particular that
    had extreme pincushion distortion.

  - I have used the 46x78 lenses recommended by Alan for his
    viewer kit, and find that they perform superbly in a fixed
    interocular viewer, even with my very narrow eye-spacing.
    (These lenses, BTW, have a 40mm diameter clear aperture.)
    Would they be even better if my eye looked dead straight
    through the center at the image?  Probably.  But, as Alan
    noted, shifting the lenses without also shifting the images
    won't result in that 'ideal' straight-through situation!

I do hope Peter has the chance to see some images in one of
Alan's viewers at his club meeting.  I know from experience that
Alan takes optical performance very seriously.  As he states on
his 3D viewer products page:
  "I not only assemble a viewer to look good, but it must perform
   correctly or I won't sell it."
http://members.home.net/3-d/products/products1.htm 

I know Alan has made substantial investments of time and expense
in evaluating lenses for suitability in various stereo viewers.
And I firmly believe the end results reflect that!

Based on my experiences, Ferwerda's caution of an unfavorable
effect can apply to many lenses, some of which are arguably
not just 'very simple quality' lenses.  But my experiences have
also shown that with adequately large lenses of sufficient
quality, interocular adjustment is generally not needed in a
stereo viewer.  (I acknowledge the possibility of a small
number of 'outliers' on the bell curve of IPD who may be
special cases.)  Most home-made viewers in the MF 3D viewer
gallery do not have interocular adjustment:
http://www.skep.com/3D/gallery.htm

BTW, my initial post against interocular intentionally
over-stated the case, and was directed at the particular
lens/interocular tradeoff I have seen in 1950s viewers.
The Red Button allows adjustment to a ridiculously narrow
setting (relative to the design of the stereo mounts).
If I were designing a viewer with interocular adjustment,
it would allow setting the lenses wider than normal infinity
spacing for the format, but very little adjustment to *less*
than normal infinity spacing.

Paul Talbot

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/