[Leica] A seeming conundrum

[Mike Johnston <michaeljohnston@ameritech.net>]

> DOF is perceived sharpness.  The key word is perceived.  There is only a
> single plane which is sharp in a picture and that is the plane which the
> lens was focussed on.  Everything else is more or less unsharp.


Martin,
Your comments are the best in this thread yet IMO, and without contradicting
you I have one amplification to contribute. "There is only a single plane
which is sharp in a picture" may not always be true simply because the plane
of best focus (p.o.b.f.) itself may not be rendered sharply. The "perceived
unsharpness" of objects farther from the plane of best focus is also
dependent on the perceived _comparative_ sharpness of those objects to the
perceived sharpness of the p.o.b.f.--i.e., how much differentiation the mind
of the viewer registers.

For instance, when the plane of best focus is not particularly sharp, say
because of camera shake or because there are no image objects on the plane,
perceived d.o.f. actually appears to be _greater_, because more of the
unsharp area comes closer to the standard of the plane of best focus. This
helps explain a seeming conundrum, which is that perceived d.o.f. becomes
_worse_ as the plane of focus gets _better_. When there is some visual cue
linking the p.o.b.f. to the o.o.f. areas--such as graininess--perceived
d.o.f. is also better. To demonstrate this, print a picture straight and
then print the same picture through some sort of texture screen. The
"sharpness" in the print made with with texture screen suffers, but the
perceived d.o.f. seems to improve. You really have to have this demonstrated
to believe it, but it's an impressive demonstration.

Just as d.o.f. is entirely a matter of perception, so is "sharpness," and
this has been proved repeatedly by empirical means. Various contributory
technical properties of "sharpness" can be described rigorously, but viewers
will still sometimes describe an image as "very sharp" that does not have
these technical properties, and as "unsharp" certain images that have the
given technical property in abundance. One specific example of this is that
in pictures with lots of front-to-back image objects, viewers decisively
label as "sharper" those prints with more d.o.f. even if the p.o.b.f. is
much worse, to prints where the p.o.b.f. is much better but the d.o.f. much
worse. Another specific example I can give is this: the Leica 35mm
Summicron-R has very high large-structure (i.e., 5 lp/mm) contrast but not
terribly good fine-detail resolution. The Pentax 43mm has the
opposite--something which can be made obvious by simply putting a loupe on
the fine detail. Yet in comparing prints made with the two lenses, people
consistently prefer the images of the Leica lens because perceived
"sharpness" is higher, even though actual recorded fine detail is worse.

Another example is that people will consistently prefer photographs in which
the objects are what is called "in drawing" (presented arranged so that
their structure is comprehensible from their surfaces) and well lit, to
pictures in which the optical imaging properties are superb but the
structure of the objects is confusing because of poor arrangement or
confusing lighting. This is what Ted's referring to when he says "Quite
often I do nothing more than walk around the room looking at  my hand to
see how the light appears when I hold it so it imitates the shape of the
head, as in I make a fist, then I determine where the subject should be
placed and away we go! :-)."

Pictorial photography begins and ends with perception. This is what Ctein
means when he says "if you can't see it in the print, it doesn't count," and
the same thing Ted is referring to when he says "I'd much rather folks were
looking at my pictures for the content and impact" and many other of his
comments along the same lines. KISS yes, but "it's the picture, stupid," is
also true. If it looks good, it is good. If it doesn't, it ain't. Perception
is the whole ball game, not only in the meaning of the subject, but in
pictorial photographic technique, too. I'll give a specific example in the
next post.

I've unpacked thousands of books since my move, but maddeningly there are
still far too many in boxes. I simply can't find the references on "image
gestalt" I need to tell anyone where to go to get more information about
d.o.f. and perception. However, the person I know of who's done the most
work in this area is Dick Zakia, professor emeritus of the School of
Photographic Arts and Sciences at Rochester Institute of Technology. My
memory is that he's written several good, understandable books for laymen on
the subject, although it's been years since I've read them.

- --Mike