Re: [Leica] Forget about DOF scales - Mike Quinn's DOF calculations

[Mike Quinn <mlquinn@san.rr.com>]

Dear Walter,

I suppose you caught my little aperture slip.
OK. If you stop down to f2001 instead of f2, and use the suggested formula
(with an aperture larger than the cow), the DOF is not 2000 feet in front
and 4000 feet behind. As you cogently point out, it's only 13 feet in front
and 26 behind.

Of course that's still a range that puts the DOF 10' BEHIND the photographer
who is standing 3 feet in front of the cow. (This interesting effect should
be particularly useful with fisheye lenses.)

Mitch Alland has been kind enough to post a reference to the paper on which
his formula was based.
> Yes, the 1/3-2/3 DOF is what the scale of the lens shows. However, Merklinger
> shows convincingly that this is not the real DOF. See:

> http://fox.nstn.ca/~hmmerk/HMbook14.html

I've been studying that reference for the last hour and, although Merklinger
is clearly wrong, (among other things he deals with muskrats rather than
cows), I think he does make a few good points.

Given: We are all sometimes unhappy with the results of using the hyperfocal
distance. (That's why some people suggest stopping down two stops from the
values on the depth of field scale.)

Merklinger suggests focusing on the most distant important object and then
stopping down the lens to reduce the circle of confusion in the foreground
to an acceptable value. That's fine, but it requires you to calculate that
acceptable foreground value, and guarantees that the foreground will be
fuzzier than optimal. (DOF does exist, and that technique will put one edge
of it BEYOND the most distant important object.)

Merklinger also suggests that if we all focus at infinity and then set our
apertures at f16, we will be blessed with sharp enough focus everywhere.
(As long as we don't try to resolve a muskrat larger than the diameter of
the lens. Did I get that right?). Oh well, I'm sure that his muskrats are a
lot smaller than most cows...

What then to do? Either technique seems pretty good. One (Merklinger)
guarantees that the background will be sharp and the other (using the scale,
but stopping down further) gives you a pretty good shot at getting a wide
DOF around the thing you are focusing on. (Would there be any difference? I
can't tell from Merklinger's examples. When he leaves cannons and addresses
more animate objects (muskrats), he suggests focusing on the muskrat's eyes
and then calculating how fuzzy its tail will look. (I thought he was going
to focus on the tail and then calculate how fuzzy he wanted the nose to
look.)

From this I conclude that the closer you get to a cow, the more likely you
are to focus on the eyes and forget the tail (which is swishing all over the
place anyway).  

I think in the future I'll follow Merklinger's suggestions and only try to
use the formula with cows that are smaller than the aperture of my lens.
That should keep them in sharp focus most anywhere.  Of course, they'll be
very small cows. I'm glad I use a Leica. Can you imagine how hard it would
be to focus a SLR on their beady little eyes at f2001?

Thanks for your correction.
(I'm mooved by your attempt to help in this difficult matter.)

Mike Quinn
 
Walter Levy wrote:

> Mike, let me be the first (I hope) to say that your calculations are udderly
> wrong.  You should be more cowful with your cowculator!