Re: f/32 testing

[T3D john bercovitz <bercov@xxxxxxxxxxx>]

Well, here's a progress report, anyway.  I made f/32, f/45 and
f/64 apertures for my 50 mm f/1.8 Canon lens and tested at a
magnification of 0.1 with subject distances for each test of 490,
534, 550, 568, and 628 mm.  The test objects gave lines per mm on
film of 60, 42, 30, 21, and 15.  Focus was on the 550 target, of
course.  (Magnification = 550/55 = 0.1; 1/550 + 1/55 = 1/50)

I have not tried projecting them on the screen yet, but it doesn't
look like it's going to be necessary.  I can see now that they're
pretty darned fuzzy due to physical optics (diffraction) problems.

At f/32, I can make out the 15 and 21 lpmm at all five distances.
At the 550 distance, I can additionally make out 30 lpmm.  At f/45
I can make out the 15 and 21 at all distances but the 30 at none.
At f/64 I can make out _none_ of the test targets' lines!  I _can_
still read the f/64 setup information tag I put under the targets.
8-)  Conclusion: conventional wisdom is vindicated.

I don't plan to pursue this at this time since I am trying to
write up my center of perspective paper, which I posted here a
while back, in a way that more people can understand it.

John B

PS: I did not use high contrast targets.  The black to white was
only two stops.


------------------------------