I noticed that on 11-9-06, at 1:12 PST, we were told that, "a CoC
diameter of .0222mm equates to a CoC diameter of .0167mm. (just take the
reciprocal to convert between lp/mm and diameter."

Actually, lp/mm is LINE PAIRS per mm, and should not be confused with
dpi, (dots per inch). A, "pair of lines", consists of a white line
paired with a black line, both lines of the same width.

Interesting, as the measurements of the black line in a resolution test
chart image of 1, (one), lp/mm actually measures 0.5mm. The above
equation would have you believe that the black line of a 1, (one), lp/mm
would measure 1.0mm in width.

In other words, I consider the CoC numbers being used in DoF tables are
twice the diameter as they should be. I proved my theory in an
Optical Physics class, some 50 years ago, using a Leica camera. I
positioned a series of resolution test charts, spaced 3 inches apart.
The camera was focused on the middle row of test charts which was
positioned ten feet from the camera. I had to use f/8.0 to get the
results predicted when using f/5.6, by the depth of field tables
provided by Leica, which were, supposedly, based on a resolution of 30
lp/mm.

I am not impressed by the so-called experts that believe in the
reciprocal conversion theory, as I do not understand how a CoC diameter
can equal the width of a paired set of equally spaced black and white
lines. It took years to determine that the Piltdown Man was a hoax.


Best regards,

DON

Hi Don,

I am not alone in my practice of converting a desired resolution in lp/mm
to a maximum permissible CoC diameter in mm by simply taking the reciprocal.

1 / lp/mm = CoC diameter in mm

And although you are among a minority in this regard, you are not alone in
doing the conversion as I do, then dividing by two, to make your CoC's half
again as small as mine to support a desired resolution in lp/mm.

1 / lp/mm / 2 = CoC diameter in mm

Who is right? Who is wrong?

You get the results you desire by doing things your way. I get the results
I desire by doing things my way.

From your perspective, anyone who listens to Mike Davis will end up with
photos that are only half as sharp as they should be.

From my perspective, anyone who does what Mike Davis does will get the
same results I'm getting.

Given that my stereo views are routinely praised for their "sharpness," I
can find no incentive to shrink my CoC's (and Airy disks) by a factor of
two with a penalty of slower shutter speeds and/or increased working
distance to the nearest subjects.

I'm nevertheless willing to concede the possibility that the minority
school of thought is correct regarding the conversion. When I submit to
that possibility, I run smack into this: Accepting your conversion
formula as "gospel" means that when I think my 0.03115 mm maximum
permissible diameter for CoC's (and Airy disks) is supporting nearly 8
lp/mm in the viewer after 4.21x magnification, it's really supporting only
4 lp/mm. I don't actually know what resolution I'm achieving after
magnification in the viewer, so perhaps 4 lp/mm is all it takes to make
people jump and down over how sharp my views are. Unfortunately, I
suspect you are not willing to concede that a post-magnification resolution
of only 4 lp/mm would be all that satisfying.

Do you see the paradox, here? If I concede that your formula is correct,
you must concede that 4 lp/mm is sufficient to garner the accolades my work
receives. What say ye? I'm OK with that, if you are.

I personally don't care what the actual resolution is. For me,
specification of a desired lp/mm after magnification, in the various
formulas I use, is nothing more than a "throttle" that I can adjust until I
get the results I require.

All I know for sure is that I had to get my on-film maximum permissible CoC
diameter down to 0.03115 before I was satisfied with the results had by
following the DoF calculations performed with that CoC. And this required
plugging a value of 7.625 lp/mm into the formula that I use for converting
lp/mm to CoC diameter.

My advice to anyone following this discussion: Use the formula of your
choice and then adjust the "desired resolution" until your DoF calculations
produce the empirical results you desire.

Using a desired resolution of 8 lp/mm in this formula:

1 / lp/mm = CoC diameter in mm

will produce the same CoC diameter as a desired resolution of 4 lp/mm will
in this formula:

1 / lp/mm / 2 = CoC diameter in mm

1 / 8 = 1 / 4 / 2 = 0.0125mm

Now divide this 0.0125mm by the magnification of your viewer to get the
on-film CoC diameter you should use for DoF calculations. Testing may
reveal a need to go to larger or smaller CoC's. The final choice is
entirely subjective, no matter how we get there.

Mike Davis