[Leica] Test versus trial?, part 2

[imx <imxputs@knoware.nl>]

Bokeh is generally defined in a very loose way, which in itself hinders a
good discussion of the phenomenon.
Often bokeh is presented as a recent discovery which is not true. The intra-
and extra focal out-of-focus blur is a wellknown topic in every optical
handbook. New is only the idea that these blur effects have a subjective
dimension and are supposed to be instrumental for the qualitative
appreciation of a picture.
Blur has a real and a subjective aspect.  The analysis of bo-keh is in
itself simple: we know from optical theory how  the focused rays will be
distributed in the unsharpness areas in front of an behind the plane of best
focus. We can also calculate the energy distibution within the blur circles,
located on the several unsharpness planes at any required distnce form the
plane of focus. This is Canon's approach and by studying the changes in
energy distribution patterns, they can predict what effect will result. The
optical cause of these different energy distributions within the blur circle
is the level of correction of spherical aberration. So if I know the level
of SA in a lens, I can safely predict the way the image will be reproduced
in the unsharpnes plane.
An optical manual from 1933 (no typo!) already gives a lucid account of this
phenomenon.
If we would do a series of experiments with our lenses to see how they
reproduce the out of focus blur circles and how these have impact on the way
the out of focus images are shaping (deforming) the image structure and
details.
As it seems the case that the characteristics of the image (shape, structure
of details, contrast) in the out-of-focus plane are decisive for the type of
blurring in the  rendition of out-of-focus foreground and background  and
the  perception of the eye to this type of blurring, we need to establish
some clear guidelines what structures we have to look at.
Some experts note that bokeh is confined to out-of-focus highlights, which
are influenced by optical design and number of diaphragm blades. But not one
experiment (and I checked more than a hundred sites on internet to become
familiar with what is the general opinion) has conclusively decided which,
if any, of these supposed causes is the real one.
Others hold that bokeh is only relevant for macro- and portrait photography.
Bokeh is variously described as "hard edged blurs", "abrupt transitions from
light to dark", "poor definition in out of focus objects", " doubling of out
of focus lines","soft edged blurs", "smooth even transitions in areas of
contrast". Then we have all kinds of descriptions of the blur spot itself:
donut shaped, butterfly shaped etc. Not one description refers to focal
length, aperture, distance and relative position of fore and background
location. Now "(fore)background"  is not an isolated plane, but a portion of
reality with depth. So where exactly are you loking at when discussing
bokeh?
As the origin of bokeh is the blur circle and its physical shape, dimension
and energy distribution, any study of bokeh would have to control the usual
blur circle effects: focal length, distance to plane of focus, distances
from PoF and print magnification. Enlarge a negative 5 times or twenty times
and the size of the blur circle changes and so the relative diatance from
unsharpness blyr to PoF also changes. Diffeernt portions of the out of focus
image are affected.
To get ahead in this barrage of definitions and options, we need to
establish a common situation. You can not compare a picture taken with a
50mm at 5.6 of a girl at 5 meters in a sunny environment and enlarged 8
times with the background at 15 meters, with a picture of a 90mm lens at 4
of a flower at 1 meters in contre jour to get nice specular highlights in
the background, where the background is at 3 meters and the enlargement is
15 times. And so on.
Without a description of an experimental situation, that can be accurately
reproduced in several testsessions, without the definition of the camera
position and distance, aperture and the definition of the specific o-o-f
planes where the bokeh effect should be studied, without specified
enlargement factors, without a specification of the blur effects to look for
and so on, we are simply getting nowhere in the direction of studying the
phenomenon.
I do assume, wading through hundreds of pages of lens descriptions by people
who discuss bokeh, that  concepts like "openness", "surrealness", "airbrush
creamy smooth" and so on are all attempts to describe the impression that
the observer gets from looking at a picture.
Such an approach makes for a fine discussion that may be satisfactory for
the participants, but is so open-ended and vague that I wonder what the goal
is here.
If you pick up a art book like " how to look at photographs" or 'The
photograph", the content and meaning of a photograph is discussed in terms
of visual language and the whole weight of visual art critique.
If you pick up an optical book, you are treated with physics, and the
fundamentals of visual perception.
Both approaches cover the whole spectrum of how to analyse and discuss a
photograph. And it is very enlightening to study both fields.
The bokeh discussion on the contrary is quite confusing and has no added
value (at least not in the present state of vagueness and lack of any
experimental base). In the Japanese culture,  the bokeh discussion may be of
great value as it fits in in their tradition of visual awareness and
sensitivity. 
If we wish to improve the quality of the bokeh discussion, we should move
beyond the trial stage and do some substantial fact finding based on
repeatible experiments.
Erwin