> The first is the performance of the lenses at f/4... an aperture much
> more easily handled by the 4/3 lens which should be designed to shoot
> high quality images at a couple of stops wider than the Zuiko 24/2.8.
I could have shown the F5.6 or F8 comparisons, but those are more
nuanced and not so blatant because of the increased DoF at those
apertures does include the distance objects and then at F8 you just
start seeing the contrast loss from diffraction. I used F4 in this
case because it was easier to see what I was pointing out AND it is
consistent with other results when shot closer to an object at higher
F-stops. Inotherwords, this is a good representation of real life
results. Besides, I did this during lunch hour while eating tacos.
> The second is the focusing accuracy achievable at 5 meters. By your
> calculations, DoF extends to 10.3 meters which implies using a CoC of
> 0.015mm. But that's at f/4. Presumably you were focusing at f/2.8 with
> the Zuiko and probably about f/3.2 with the zoom (which has an aperture
> range of 2.8-3.5). Since you're using a magnified image for focusing
> I'll allow that the effective CoC is cut in half and tightens up to
> 0.0075mm. That would allow your actual focus at f/3.2 to be as far away
> as 8.5 meters instead of 5 meters and still not detectable to the eye.
> Then, when the CoC is relaxed back to 0.015mm and f/4 taking aperture
> the far DoF can extend all the way to 70 meters.
I'm not sure I'm following your explanation. Each camera was focused
to the same point--as close as live-view with 10x magnification on the
L1 is able to achieve. I always do the back and forth routine to fine
tune focusing to the halfway point between known defocuses. This is a
technique I've used for decades and works well for me. In the image
files, the images appear to achieve maximum sharpness at the same
point. Yes, I focused both wide-open. Neither of these two lenses
experience problematic focus-shift with aperture adjustment that I've
seen.
Again, this isn't my first comparative test and this one does reflect
the results from the others.
The explanation that the exiting light cone is narrower is also my
belief too. The Zeiss DoF/Bokeh article is a good reference in this in
that it talks about the difference between Depth of Focus and Depth of
Field. One is object side and the other is sensor/film side of the
lens.
Let's think about a fictitious lens for a moment. The effective
focal-length is 50mm on a 35mm camera/film format. But unlike our
usual 50mm lenses, which, if symmetrical and non-tele or non-retro
will place the convergence point 50mm from the film surface when
focused at infinity. In this example, the exiting light rays are at
the same angle as the incoming light rays. Now let's redesign the lens
so the convergence point is placed 500mm from the film surface. The
exiting light rays will be at a very narrow angle. Since the maximum
amount of light-ray spread is only a fraction of that of the
symmetrical non-tele/retro lens, the CoC of the sensor-side of the
lens is greatly reduced.
Now, let's really blow people's minds...
Most zoom lenses aren't variable focal-length lenses. They're a
wide-angle lens with a second lens group which zooms in or crops or
optically selects a portion of the resulting image. They are actually
two lenses in one. A 14-50 Leica when set at 24mm isn't a 24mm lens.
It's a 14mm lens with a variable teleconverter added behind.
(obviously it's a little more complicated than this as front elements
are moved around too, but this is to correct for distortion, not
magnification). There are a handful of lenses which are reverse--they
are the longer focal-length with variable <1x optics. A pure varifocal
zoom is a different animal, though, because whatever focal-length it
is set at is true, not magnified. Actually, the Leica is a hybrid lens
which possesses both varifocal optics and variable teleconverter. This
is why it requires refocusing after zooming.
Now, to bring this back around to my fictitious 50mm lens with 500mm
base... A "telecentric" lens is optically doing this very thing. It is
taking a very short focal-length (like the DZ 7-14mm) and moving it
away from the sensor plane. This is done for mechanical reasons, but
also it is for optical reasons. By moving the exit pupil far from the
sensor plane the exiting light rays leave at an extremely narrow angle
in comparison to the subject-side light-rays.
I haven't even gotten into the subject of lens elements which aren't
uniform in shape and what they do to out-of-focus light rays.
Just as an aside, the Zuiko 14-54 mk1 lens produced results closer to
the OM-Z 24mm lens than the Leica. The Leica is in a different league
when it comes to this kinda thing and it regularily gives an effective
DoF of an additional stop. I've said it before and I'll say it again,
this lens is definitely different than anything I've used before. In
fact, when putting together the test results for this article I had to
keep catching myself because I was trying to match up the wrong
images. The Zuiko at F5.6 was extremely close to the Leica F4.0 image,
and the F8 matched up to the F5.6 image. Others with this lens have
said the same thing, so it isn't just me.
AG
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|