[Leica] Hyperfocal Focusing

[msmall at aya.yale.edu (Marc James Small)]

At 07:25 PM 8/30/05 +0200, Philippe Orlent wrote:
>How do you come to a CoC of 1/1000, thus 0,001?
>
>My very handy hyperfocal chart takes 0,03, thus 1/33, as CoC, making a
>difference per type of camera
>(see http://www.vividlight.com/articles/3513b.htm and
>http://www.vividlight.com/pdf/hyperfocal.PDF)
>
>What determines these numbers?

Let's get back to basics here.  The Gevaert Manual of Photography (5th Rev
Ed, 1962) carefully reminds us on p. 31 that "the conception of depth of
field is purely relative and depends on the criterion of sharpness
considered necessary for any particular application".  That means, DOF
varies substantially depending on the actual sharpness necessary for the
job at hand.

This depends in large measure also on the quality of the film used.  The
"circle of confusion" is a mathematical statement of the resolution
required.  A rather grainy film will permit a low circle of confusion;
that is, if the film grain comes to, say, 1mm, in diameter, than the circle
of confusion required will be 1mm (roughly 0.04" or, in franctions, 4/100")
 That would be a really coarse-grain film.  More common are grain sizes in
the area of 0.1mm, or 1/250", which would be that of, say, a relatively
high-speed older film such as Royal-X or the like.  In today's world,
finer-grained films with higher speeds are readily available, so that it is
probably the case (I have NOT looked any of this up, but I am relatively
certain of the data presented), a rough grid could be developed as follows:

slow films (say, 100 Delta)             1/1000" or 0.025mm
medium-speed films (say, PanF)  1/750"or 0.034mm
fast films (400 Delta &c)               1/500"or 0.05mm

This is supposed to be the size of the average grain actually produced by
these emulsions and, agian, I've not checked this out against Ilford's data
but I am fairly certain that I am in the ball-park.

The basic formula is:  the hyperfocal distance in Imperial measures equals
the focal length of the lens in inches squared divided by 12 times the
aperture, this quantity then being multipled by the reciprocal of the
circle of confusion in fractions of an inch.

So, again, let us consider this 35mm f/3.5 lens when shot at f/3.5 under
varying conditions of film resolution.

With a slow film, we get the following:

The lens focal length of 35mm in inches is 1.378".  This quantity squared
is 1.9.  Divide this by 12 x 3.5, or 420, to achieve a factor of 0.0045.
Multiply this by the reciprocal of the circle of confusion your film
requires, and you get the following:

0.0045 times 1/1000 (low-speed film) = 4.5 feet
0.0045 times 1/750 (medium-speed film) = 3.4 feet
0.0045 times 1/500 (high-speed film) = 2.26 feet

I believe that my maths are right but, then, maybe not.  In any event, the
results I get are as follows:

100 ASA film:  set the lens to 4.5 feet and everything between 2 feet 3
inches and infinity will be focus to the limit of the film.
200 ASA film:  set the lens to 3.4 feet and everything between 1 foot 8.4
inches and infinity will be in focus to the limit of the film.
400 ASA film:  set the lens to 2.26 feet and everything between 1 foot 1?"
will be in focus to the limit of the film.

,Note that the range of the focus increases as the circle of confusion is
reduced.

Trust me on this one:  the press photographers who dominated American
photo-journalism from the 1920's to the 1960's could run this formula in
their heads while they were humping their Speed Graphics from one location
to another.  These guys really lived up to the Ben Hecht iconography and
most smoked cigars and most carried hip flasks and all of that -- but they
could crank out a hyperforcal distance in a heartbeat.  It was the stuff of
their lives, as they only got ONE chance to get a GOOD shot of, say, the
local Mayor being hauled to the jailhouse or of a local pastor fleeing a
whorehouse.  So, yes, hyperfocal distances were a key part of their lives.
(I had an encounter with one of the Speed Graphic guys from the dead and
much-lamented Baltimore News-American around 1975 at nine in the AM:  he
was already drunk but explained hyperfocal distance to me so clearly that
the concept has lived with me to these days.)

Bear in mind that the circle of confusion varies from with the type of film
used.  That is why those Depth-of-Field scales printed on the lenses are
most suspect, unless you know just what circle of confusion was used in the
calculations to produce these ranges.  In some cases, such as the 4.5/21
Carl Zeiss Biogon, it really doesn't matter due to the extreme shortness of
the lens coupled with the slow basic speed of the lens, but it does make a
lot of difference with, say, a modern 1.4/35 Summilux-M whether you are
using your reserve stock of PanF or 3200 Delta.  

Marc



msmall@aya.yale.edu 
Cha robh b?s fir gun ghr?s fir!

NEW FAX NUMBER:  +540-343-8505