Wayne Harridge <Wayne.Harridge@xxxxxxxxxxx> moved upon the face of the 'Net and
spake thusly:
> Wrote a program to do the very same thing in FORTRAN 20 years ago !
Reimplementing the wheel in ones Language Of Choice is a favourite
pastime of programmers everywhere. :-)
My Forth driven optical calculator allows you to have an interactive
conversation, remembering default parameters once you set them. I
hope to port it to Palm Pilot or similar device when I actually get
one of the beasties.
You can choose your film format and other parameters. You can enter
and display values in whatever units you prefer (internally everything
is in metres).
Can calculate magnification from distance, or distance to gain
required magnificaiton. Angle-of-view vs. focal length can be worked
both ways also. There are also routines not listed in the banner to
grind out depth-of-field tables for lenses which lack DoF scales.
Here's a trivial session. I will precede my typing with '$'.
Everything else is response from the calculator:
$ gforth optic.f
-------------------------------------------------------------------------------
Optic calculator environment by Christopher Biggs
Make value: micron mm cm inch metre degree
Show value: in-mm in-cm in-metre in-inch in-degree
Set Film : 35mm 6x6cm 4x5in 8x10in
Set var : focal-length aperture-ratio subject-distance
circle-of-confusion film-format (WxH)
Attributes: fov mag[nification] red[uction] dof nlf flf hfd
Details : show-field show-parms show-subject show-hfd
Conversion: mag2dist mag2dof fov2f
-------------------------------------------------------------------------------
Current parameters:
Film format is 36. mm x 24. mm.
Circle of confusion is defined to be 0.0333 mm
Lens has f=50. mm and is stopped down to f/8.
Subject is at 1. m.
ok
$ 25 micron circle-of-confusion
ok
$ 50 mm focal-length
ok
$ 450 mm subject-distance
ok
$ show-parms
Film format is 36. mm x 24. mm.
Circle of confusion is defined to be 0.025 mm
Lens has f=50. mm and is stopped down to f/8.
Subject is at .45 m. ok
$ show-field
Subject at 45. cm is in focus from 43.4 cm to 46.7 cm (3.24 cm ) at f/8.
Blur factor is 55.56 ok
$ dof in-inch f.
1.277 ok
$ f/1.4 aperture-ratio
ok
$ show-field
Subject at 45. cm is in focus from 44.7 cm to 45.3 cm (.573 cm ) at f/1.4
Blur factor is 314.3 ok
$ show-subject
With f=50. mm, subject at .45 m is reduced 4.5 times.
Subject area is 16.2 cmx 10.8 cm ok
> My concern is the "Circle of Confusion", what is acceptable in some
> circumstances may not be for others, e.g. the acceptable (what degree of
> unsharpness will I tolerate) size may depend on the degree of enlargement of
> the final image and the viewing distance, etc., etc.
>
See the (formerly Ilford) _Manual_of_Photography_ for one discussion of
these issues and a survey of common approaches.
> I guess there is a "Standard" size for the CofC, what is it, and for
> what situations is it "optimized" for ?
Depending on the physical basis for your choice and whether you're
doing wave or particle optics, various figures are used.
f/1000 is one limit that arises from certain physical models.
33um (1/30 mm), 30um and 25um are common figures that are plucked from
the air and used by lens manufacturers.
I seem to recall from somewhere that nikon uses 33um, while Leica and
Zeiss use 30 or 25. I've never thought to do the numbers and see
what CoC choices the DoF scale on my Zuikos imply.
A fixed value for CoC assumes that the bigger the print the bigger
the viewing distance. The f/1000 value corrects for enlargement by
assuming that a print will be viewed from its "proper" distance,
i.e. where the print subtends an angle of view equal to the angle of
view of the original taking lens.
When you really start to get critical and ask the hard questions, you
could argue that the whole concept of depth-of-field is bunk
anyway. :-)
There was a big argument about this on Usenet recently (regarding
whether one should focus at infinity or hyperfocal-distance for
landscapes).
> When I was doing calculations of
> DOF I allowed the CofC size to be specified by the user, depending on how
> critical the sharpness requirement was.
Me too. I choose 33 micron as a default because it's widely used.
> Sorry, just my ramblings, nothing is ever "absolute".
>
Yes, that's why I qualified the figures with what CoC value I used.
cjb.
--
------------------ Linux hackers do it in protected mode -------------------
| Christopher Biggs - Software Engineer, Stallion Technologies, Australia |
| chris@xxxxxxxxxxxxxx - Ph+61-7-3270-4266 - Fx+61-7-3270-4245 - PGP&MIME |
------- Mathematics and alcohol don't mix --- Never drink and derive -------
< 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 >
|