At 23:16 1/13/02, Moose wrote:
Thanks for the info, John. I'm not sure we're quite on the same wave
length here. I was talking about DOF in the final image, let's say an 8x10
print. You say under b. "if maximum allowable circle of confusion diameter
remains unchanged". Since I am positing the same allowable CoC diameter
for identical size prints from different size negatives (this is the same
for slides, just change the names), I do allow a larger COC on the larger
neg, as it will be enlarged less to produce the final image.
Yes, very true.
I stated the condition (CoC remains unchanged) only so that the statement
about DoF changing with the square of the focal length would be technically
correct. It presumes all other variables are held constant. Think of a
film format change from 35mm to medium format as a two step process.
(a) Change the focal length; DoF decreases at square of focal length. This
is the "two steps backward."
(b) Next change the CoC to match the increase film size; DoF increases
linearly with increase in film size. This is the "one step forward."
Another interesting thing about DOF that hasn't been raised in recent
discussions is the underlying assumptions.
[middle part snipped out]
The allowable CoC diameter is then derived from this kind of subjective
data and all kinds of wonderful calculations are made and tables published.
All very well and wonderfully useful to the practical photographer. For
the thoughtful photographer involved in all aspects of the dance with
light and technology that creates the final image, it is useful to
remember this is just a guideline that doesn't necessarily work for the
particular purpose at hand.
[remainder snipped out]
I agree completely . . .
The subjective aspect is too often forgotten. Most "on line" calculators
for 35mm film format use a max CoC diameter of 0.030mm (CoC size
_on_film_). The value *is* subjective. I use a smaller value of 0.025mm
for 35mm format and 0.050mm for 645 medium format in my
spreadsheets. Why? Large exhibition prints tend to be examined from a
closer distance than a "casual viewer" would use. It is inherently
presumed by most DoF calculators that an 8x12 print will be viewed at
*twice* the distance as a 4x6 print. If so, apparent DoF between the two
remains the same. However, if both are viewed at the *same* distance,
apparent DoF in the 8x12 will be shallower!
For exhibition prints, I presume a 10 inch viewing distance regardless of
print size (8x10, 11x14, etc.). This is the distance I use for evaluating
them after they're framed and mounted on a wall with image center at 60
inches above the floor (approx. "gallery height"). It is shorter than the
casual viewer would use. "Critical" viewers, including exhibition jurors,
will move closer to examine details!
OTOH, Kodak uses approx. 0.033mm max CoC, an "average viewing distance" of
14 inches for a 4x6 print, and changes viewing distance at the same rate as
changing print size to puff up their "sharpness" claims. This pushes to
the ragged edge of perceived sharpness; it's based on 900f viewers
claiming it's UNsharp at their viewing distance before it becomes
UNsharp. Yet another reason to keep in mind the subjectivity surrounding
underlying assumptions when reading claims a particular digital camera can
easily create "sharp looking 8x10 prints" (based on same principle; max CoC
diameter becomes pixel size)! Kodak *also* assumes the image will be
printed on something like a dye sublimation printer, and that zero overkill
with pixel resolution is required to enhance overall sharpness.
Final advice:
Use hyperfocal focusing techniques for subject material such as landscapes
*very* carefully! It's based on the subjective max. CoC diameter. On the
infrequent occasions I use it, the lens is stopped down one additional stop
to ensure the very distant background will *not* look soft in a large
print. I have seen exhibition prints in which I *know* the photographer
used hyperfocal focusing. The extreme DoF combined with slight softness
with detail in the most distant background are the "telltales." This
wouldn't show in a 4x6, but at 11x14 it can.
-- John
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|