Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D

Notice
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
<-- Date Index --> <-- Thread Index --> [Author Index]

Re: Lens separation, window, Realist ads


  • From: bercov@xxxxxxxxxxx (John Bercovitz)
  • Subject: Re: Lens separation, window, Realist ads
  • Date: Wed, 13 Dec 95 13:44:31 PST

Stephen Kearney writes:

> If I take a stereo pair with twin cameras mounted 6" (lens to lens) apart,
> as opposed to the usual 2.5", what impact will it have on foreground
> subjects? The 1/30 rule says the closest object can be 15' away, but what
> if I violate the rule? Will the depth simply look exagerated or will it
> become an 'unfusable' stereo pair? I understand the 1/30 rule only applies
> to scenes that include infinity.

Well really the rule is that the maximum on-film deviation is 1/30th of 
your viewer's focal length.  So now we need your lenses' focal length.
Would that be about 50 mm as in a Star D?  Then your maximum on-film 
deviation would be 50/30 = 1.66 mm.  Note that for best results, you would
want your camera's lenses to also be 50 mm focal length.  (Sorry, sometimes 
Orthoman just can't resist.  _Some_times?)  I'll assume you did the right 
thing and therefore your taking and viewing lenses have the same focal 
length.  In this case, a recent post applies:

> In some situations, you may wish to go to the trouble of using the 
> exact formula for on-film deviation to decide the maximum separation 
> for the two perspective points of the stereo pair.  If you have a 
> really close-up shot, this formula is worth using.  
> 
>              af*an 
> b0  =  d* [ ------- ( 1/f  - 1/a) ]
>              af-an 

You can fudge the formula a little if you don't willy-nilly frame in such 
a way that near foreground objects partially occlude far background objects.

Assuming your focus is the best focus between the near and far objects in 
your scene (a = (2*af*an)/(af+an)), and that your taking and viewing lenses 
both have 50 mm focal length, we can make a chart like so:

d  = Allowable on-film deviation
an = distance from the camera lens to the nearest object in the scene
af = distance from the camera lens to the farthest object in the scene
f  = The focal length of the camera's lenses
a  = The distance at which the camera is focussed
b0 = maximum allowable camera lens separation
All values in mm

      d   an    af    f   a   b0
    1.66  500   750  50  600  46
    1.66  500  1000  50  667  31
    1.66  500  1250  50  714  26
    1.66  500  1500  50  750  23
    1.66  500  1750  50  778  22
    1.66  500  2000  50  800  21
    1.66  500 10000  50  952  17
    1.66  750  1000  50  857  94
    1.66  750  1250  50  938  59
    1.66  750  1500  50 1000  47
    1.66  750  1750  50 1050  42
    1.66  750  2000  50 1091  38
    1.66  750 10000  50 1395  26
    1.66 1000  1250  50 1111 159
    1.66 1000  1500  50 1200  95
    1.66 1000  1750  50 1273  74
    1.66 1000  2000  50 1333  64
    1.66 1000  2250  50 1385  58
    1.66 1000  2500  50 1429  53
    1.66 1000  2750  50 1467  50
    1.66 1000  3000  50 1500  48
    1.66 1000 10000  50 1818  36
    1.66 1500  2000  50 1714 193
    1.66 1500  2500  50 1875 121
    1.66 1500  3000  50 2000  97
    1.66 1500  3500  50 2100  85
    1.66 1500  4000  50 2182  78
    1.66 1500 10000  50 2609  57
    1.66 2000  2500  50 2222 325
    1.66 2000  3000  50 2400 195
    1.66 2000  3500  50 2545 152
    1.66 2000  4000  50 2667 130
    1.66 2000  4500  50 2769 117
    1.66 2000  5000  50 2857 109
    1.66 2000 10000  50 3333  82

Remembering that 150 mm is about like 6", it looks like you 
wouldn't want to shoot too close and too deep with that separation.  
(I bet you knew that, huh?)

John B


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