Hi Mike:

I have always considered that the optimum resolution of a stereo slide
is entirely subjective.

I do get bothered when I see exaggerated measurement numbers being
posted, which I consider to be an attempt at inferring increased
accuracy: Example:

"....I use a CoC of 0.03115mm to support a resolution of 7.625 lp/mm,
after 4.21 magnification in my adjustable focus 78mm Saturn Slide
viewer". your ....viewer ....is capable of even higher
magnification...."

Why was 0.03115mm used instead of .03mm. Who would be able to see the
difference ?

Also why was 7.625 lp/mm used, rather than 7 lp/mm. Who would be able
to see the difference.


I am at a loss as to being able to understand how anyone can suggest,
with any accuracy, (3.39X and 4.1X), as being the magnification provided
by a MF viewer ?


Incidentally, my viewer contains 68mm fl lenses, which when focused at
infinity produce a, "theoretical", magnification of 3.7X, only 9 percent
larger than
3.4X.

Also, I do have my doubts as to your suggestion that using f/32 should
not be a usefull diaphram setting, when shooting MF stereo slides.


Best regards,

DON

Don,

I apologize for using so many insignificant digits. With that said, can we
get back to the topic we were discussing?

Unanswered question (paraphrasing): Do you consider CoC's that can only
support a resolution of 4 lp/mm delivered to the final image, after
magnification, to be "sharp" enough for your tastes?

Mike Davis

I for one ABSOLUTELY do not want to get involved with the COC/lmm
issue, but I am curious about the quote below. Has anyone seen Gary
Cullen's f:45 shot in the current folio? I have some f:128 shots from
older cameras. I would be interested in other physical (NOT
mathematical) examples of the effects of MF3D images stopped down to
the extreme. These would be quite useful for those who do MF macro.

Sam

--- In MF3D-group@yahoogroups.com, Don Lopp wrote:

>
> Also, I do have my doubts as to your suggestion that using f/32 should
> not be a usefull diaphram setting, when shooting MF stereo slides.

Sam Smith wrote:

> Has anyone seen Gary
> Cullen's f:45 shot in the current folio?

A'yup. It looked good to me :)

I've taken (I think) only one shot on my
Rolleidoscop at f/32 (and I doubt I could
find it again on the mess that passes for
my desk) and except for the great depth of
field, I didn't notice anything wrong with it.

My 'scop will go even smaller but I haven't
actually shot film at its smallest stop.
--
John Thurston
Juneau Alaska
http://stereo.thurstons.us

John and Sam,

I promise to use no math in this thread. :-)

When considering whether or not diffraction is "visible" at a given
combination of aperture and enlargement factor (viewer magnification),
please remember that diffraction affects the entire image uniformly. It's
far easier to detect "out-of-focus" areas in the foreground and/or
background of an image captured with insufficient Depth of Field (too wide
an aperture), than it is to detect the degradation caused by diffraction
(too small an aperture.)

It's easier to detect insufficient DoF because the areas of inadequate
focus are immediately adjacent to areas that appear to be sharp. When
diffraction goes to work on an image, it's hard to detect the degradation
it causes (a loss of fine detail) until the damage is well along. Again, I
say this because diffraction does its thing to the whole image
simultaneously. You have to compare entire images - one taken at f/22 to
one taken at f/32, for example. You can't be sure you're discerning things
correctly looking at only the f/32 shot. You've also got to remove DoF
from the comparison.

So, to determine at what aperture diffraction becomes visible for a given
enlargement factor, with no concern for variations in DoF, try shooting a
relatively flat, finely textured subject - like a brick wall. Subjects
like this, taken with the lens axis perpendicular to the plane of the
subject, from a distance of say 5 feet, will allow every f/stop to
compete without DoF being an issue. Aberrations will still disadvantage
the widest apertures, but such factors will be greatly reduced when stopped
down for comparing a shot taken at f/22, to f/32, or f/45, etc. I also
recommend using a tripod and cable release to remove camera motion as a
factor, too.

Focus carefully just once, then shoot the wall at each f/stop, adjusting
shutter speed to maintain a consistent exposure. Mount the stereo pairs
and have fun looking for the stop at which degradation kicks in (if indeed
your camera lens stops down far enough for diffraction to be an issue at
your viewer magnification).

For the record, I believe the lower magnification (3.4x) imposed by the
fixed-focus 75mm lenses in the Uni-Colour/3D World STL viewers, will allow
the use of f/32, without inducing "visible" diffraction, vs. a limit of
f/22 for the higher magnification (4.2x) imposed by a 78mm SaturnSlide
viewer focused at an image distance of only 10 inches. I don't think f/45
and smaller stops can be used in either viewer without seeing at least
some loss of detail.

Here are some example photos that illustrate what diffraction looks like:

http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml

Mike Davis

My original point was twofold Mike: Did you actually suggest that
"f/32 should not be a usefull diaphram setting" as quoted? If so, I
think what needs to be defined is what you would lose versus what
could be gained. The TL-120 has showed me two things, one that
incredibly razor sharp images are obtainable with a handheld camera,
and also the weaknesses of being limited to f:22. The gained sharpness
on the focused object contrasts significantly with the softer
foreground and background, making the gain also a loss. I will modify
my own camera to f:45, as I would rather have "OK sharp" throughout at
times than "outstanding sharp" in only a portion of the image. Other
times it would be the opposite, but I want to have the choice.

> Here are some example photos that illustrate what diffraction looks

like:

>
>

http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml

This link appears very deceiving to me. There is no variance in DOF
field throughout the examples, making me think it is the same image
softened electronically. Surely they could come up with a better
example to prove their point.

In the folios, most attempts at small base closeups are compromised by
soft foregrounds and backgrounds. A sharp stamen and a soft petal just
doesn't cut if for me anymore. I want to see the whole flower sharp,
or it's just another flower picture (example only- not meant for
anyone's particular work). It would be interesting to see a few
samples of the goods and evils of extremely small apertures,
especially when 35mm macro cameras get away for f45 and f80.

Sam

Hi Sam!

At 09:26 PM 11/14/2006, you wrote:

My original point was twofold Mike: Did you actually suggest that
"f/32 should not be a usefull diaphram setting" as quoted?

This morning I wrote:

For the record, I believe the lower magnification (3.4x) imposed by the
fixed-focus 75mm lenses in the Uni-Colour/3D World STL viewers, will allow
the use of f/32, without inducing "visible" diffraction, vs. a limit of
f/22 for the higher magnification (4.2x) imposed by a 78mm SaturnSlide
viewer focused at an image distance of only 10 inches. I don't think f/45
and smaller stops can be used in either viewer without seeing at least
some loss of detail.

Testing shots made on a tripod, at various apertures, with a finely textured, flat subject is the best way to tell what you will find acceptable.

If so, I
think what needs to be defined is what you would lose versus what
could be gained.

It's a subjective choice ultimately.  We can allow the entire image to go a little bit soft, by stopping down so far that diffraction kicks in, rather than backing away from the nearest subject to use the next widest aperture for sufficient DoF.   The good (?) news is that many people aren't likely to notice a uniformly degraded image, even when this same cross-section of your audience might immediately notice that the limits of the subject space are softer than the middle-ground.

The TL-120 has showed me two things, one that
incredibly razor sharp images are obtainable with a handheld camera,
and also the weaknesses of being limited to f:22. The gained sharpness
on the focused object contrasts significantly with the softer
foreground and background, making the gain also a loss.

Well said.  Lenses of high resolution and contrast have a habit of doing that.  :-)

I will modify
my own camera to f:45, as I would rather have "OK sharp" throughout at
times than "outstanding sharp" in only a portion of the image. Other
times it would be the opposite, but I want to have the choice.

That's certainly your prerogative - to forfeit overall sharpness if doing so will permit a composition that couldn't be had otherwise.

> Here are some example photos that illustrate what diffraction looks
like:
>
>
http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml

This link appears very deceiving to me. There is no variance in DOF
field throughout the examples, making me think it is the same image
softened electronically. Surely they could come up with a better
example to prove their point.

He used a 180mm lens on a Phase One P45 back (which has a 49x37mm sensor (about the size of two 35mm frames stacked one atop the other.)  Given that the diagonal of the sensor is only 61mm, a 180mm lens would be similar to using a 233mm lens on a 6x6cm camera.  With so long a lens, the working distance could have been great enough to explain the apparent lack of change in DoF across that range of stops.

Here's another link that discusses diffraction in more detail, but have a look at the interactive graphic under the section header "What it looks like":

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

Just sweep your mouse across the f/stops shown beneath the photograph of the corduroy fabric and watch what happens to the detail as you stop down.

Mike Davis

> > Here are some example photos that illustrate what diffraction looks
>like:
> >
> >
>http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml
>
>This link appears very deceiving to me. There is no variance in DOF
>field throughout the examples, making me think it is the same image
>softened electronically. Surely they could come up with a better
>example to prove their point.

Actually, the top pair is a cross-eyed stereo pair. The stereo is very
slight, but it's there.

But yeah, there's something about this that doesn't look right.

_________________________________________________________________
All-in-one security and maintenance for your PC. Get a free 90-day trial!
http://clk.atdmt.com/MSN/go/msnnkwlo0050000002msn/direct/01/?href=http://www.windowsonecare.com/?sc_cid=msn_hotmail