[Leica] Hyperfocal distance

[henningw at archiphoto.com (Henning Wulff)]

Optical design of retrofocus or telephoto design has no bearing on 
dof and consequently hyperfocal distance. Only focal length and 
aperture are of importance in 'normal' lenses. Things like true 
telecentric lenses are excluded from the definition of 'normal' 
because they don't play by the rules with respect to focal length and 
such.

The arguments and calculations that you lay out are valid; the only 
thing is that the coc of .03mm is, as noted, too loose. For casual 
photography that will not be enlarged beyond 4x6 prints it's OK, but 
for more demanding purposes a factor of 1.4 (1 stop) or 2 is, as you 
noted, more appropriate. With digital and the pixel peeping that goes 
hand in hand with that, even a 2.8 (3 stop) tightening is noticeable 
and sometimes desirable. At that point the hyperfocal distance loses 
its appeal for standard and longer lenses, so you might as well just 
focus on your main point of interest :-). Wideangle lenses of course 
are helped most by the use of hyperfocal distance, since it's often a 
short enough distance.




At 6:27 PM -0400 4/7/09, Lawrence Zeitlin wrote:
>Here is a question for LUG technical experts.
>
>I was checking through my 1948 edition of the Graphic/Graflex manual 
>and I found a rule of thumb for computing hyperfocal distance. The 
>rule states that the hyperfocal distance of any lens is equal to the 
>effective diameter of the lens opening x 1000. (H=d x 1000)
>
>Example: A 50 mm fl lens at f4 has an effective aperture diameter of 
>12.5 mm. The hyperfocal distance is thus 12500 mm (12.5 meters or 41 
>feet). At f8 the lens has an effective aperture diameter of 6.25 mm 
>so the hyperfocal distance is 6250 mm (6.25 meters or 20.5 feet. I 
>checked the formula calculation against Kodaks chart of hyperfocal 
>distance for various focal length lenses and apertures and the 
>figures seem to correspond closely.
>
>The lenses referred to the the Graphic/Graflex manual and the Kodak 
>tables were all normal prime lenses. None were retrofocus designs. 
>Further the Kodak tables, and I assume the Graphic/Graflex formula 
>used a circle of confusion equivalent to 2 minutes of arc, roughly 
>1/1720 of the focal length. That works out to about .03 mm for a 50 
>mm lens, about a resolution of 34 lines per mm. I recall that this 
>was the resolution that the late great Modern Photography deemed 
>acceptable (but not great). Doubling the calculated resolution to 64 
>lines per mm by increasing the multiplier in the equation to 2000 
>should bring the formula more in line with the demands of LUG 
>listees.
>
>Now my specific questions. Is the formula still valid for modern 
>retrofocus lenses as used in most DSLRs and the M8? Does anyone 
>still use hyperfocal distances nowadays? I confess that I make a 
>stab at it in shooting my Rollei 35. An estimate of hyperfocal 
>distance is also useful in street photography when you can't take 
>time to focus. But do the estimates work for modern optic designs?
>
>If you can get your hands on it, the old Graphic/Graflex manual is 
>quite a read. The chapter on lenses was written by Rudolph 
>Kingslake, the chapter on printing by Ansel Adams, that on 
>composition by Berenice Abbott, and so on. War photos were of WW2. 
>News photos by WeeGee. The section on PR photography was illustrated 
>with photos of young actors and actresses who have become household 
>names. Cameras weighed five pounds and up. Heiland flash guns 
>mounted four flashbulbs in case a lot of light was needed for slow 
>color films at ASA 12.
>
>Thank God for relatively tiny Leicas.
>
>Larry Z
>


-- 

    *            Henning J. Wulff
   /|\      Wulff Photography & Design
  /###\   mailto:henningw at archiphoto.com
  |[ ]|     http://www.archiphoto.com