The principle of diffraction limiting is well known in the world of lasers.
With a laser you can easily calculate divergence for any given wavelength.
The optical trick to overcome divergence is to place a lens in the optical
path to make the beam collimated.
By collimating the light as it exits the pupil, you are able to redirect the
diffraction-induced divergence. Collimation is standard operating procedure
in lens designs for modern zooms as well as "made for digital" lenses. This
is one trick in the many books of tricks of optical engineers to overcome
the curvature of an optical wavefront. In the early days before aspheric
lenses, this effort to collimate the light would result in spherical
aberrations. This is very evident in some lenses where a point-source light
wouldn't have a visible Airy disc, but would be surrounded by distinct rings
or phantom images instead.
The calculation for determining the diameter of the Airy disk assumes a
"simple lens" with an uncollimated exit.
One way to illustrate this is how two lenses of the same focal length render
out of focus highlights. This is easy to see with an LED on any electronic
device. Take your camera, focus the lens at infinity and move the lens to
about 10mm from the light. Look through the viewfinder (or take a picture)
with the lens wide-open (or at a given aperture). Now repeat this test with
a lens of a different design, but at the same focal length and aperture.
You will see that this out-of-focus light will spread to a much greater
diameter on an old prime lens than a modern zoom lens where the actual
diameter of the LED is limited to about 200% of the actual diameter of the
light source. Depending on the design of the older lens, the LED's light may
spread to the maximum extent of the imaging circle of the lens.
I know this is a seperate issue than diffraction, but the point is that this
divergence control that affects out-of-focus highlights also affects
diffraction. If a modern lens is capable of limiting the spread of an OOF
highlight by 50%, it is also going to limit the spread of that Airy Disk by
50%. In almost all cases, this is done post aperture, although some
zooms are actually two or more lenses in one with one pre-aperture and the
other post-aperture. As you adjust focal length, the front-group moves, or
the rear group moves or even the entire mid-group containing the aperture
will move. The front group may move to adjust for wide-angle focal lengths
(or at least wider than the centering focal-length of the lens), the rear
group will move to adjust for telephoto focal lengths. If you watch some
lenses when you zoom you will see some things move at some point and other
things move at other points.
In summary (yeah right), what I am saying is that the lens engineers are
able to game the optical playbook in such a way that the mathematical
calculations written around optical wave principles apply, but can be fooled
into not being relevent when the output focus point is well known and fixed.
AG
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|