On 12/30/2016 5:03 AM, bj@xxxxxxxxxxxxxxx wrote:
Recently I bought an Olympus AF 50-200mm ED f2.8-3.5 lens,
second-hand. Fortunately I was able to pay just 25% of the new price
for it.
I had the idea that I would take sharp, detailed photographs
of wild animals and birds that matched the wonderful quality of some
other brand equipment photos I've seen.
It seems to be in top condition. Despite reviews saying it is a fairly small
lens,
I found it to be a large and heavy lens. No problem with that - my assessment.
From the first shot I took with it I had the opinion that it did not
deliver the wonderful images that Olympus and reviews said it would.
So I have done comparison shots with a range of other lenses, manual
and autofocus, which I have. One shot taken with the M5, the rest with
the E-3. All shots taken at the same spot 6 metres away, on heavy
tripod, at f/8, IS off.
I'm uncertain what meaningful conclusions may be drawn about the lenses themselves from these images. They may well be
indicative of what results you, personally, may expect, as opposed to the inherent qualities of the lenses themselves,
or to the results others might get with different samples of the lenses.
1. At least one sample was very sharp, according to the IR tests:
"/Sharpness/
Olympus has done some extraordinary work with this lens. Meant to be versatile across a wide range of apertures and
focal lengths, the 50-200mm ƒ/2.8-3.5 is very sharp wide open at ƒ/2.8, though sharpness is at its best below 100mm. Set
to a focal length greater than 100mm at ƒ/2.8, the sharpness profile becomes a little uneven across the frame.
Curiously, according to our lab results, there is a ''sweet spot'' of sharpness at 70mm. When set to ƒ/2.8, even at its
''worst'', sharpness does not exceed 2 units on our scale.
Sharpness improves at ƒ/4, and the uneven sharpness seen above 100mm settles down. Below that, sharpness registers at 1
unit, which is about as sharp as we can measure. Optimal across-the-frame sharpness is obtained at ƒ/4, 70mm.
Diffraction limiting starts to set in at ƒ/16, but even here you're not exceeding 2 units on the blur scale. Even at its
worst, ƒ/22 at 200mm, you're at 3 units on the scale. Superior results."
<http://www.imaging-resource.com/lenses/olympus/50-200mm-f2.8-3.5-zuiko-digital/review/>
2. You used FF thinking, doing the test @ f8. Note in the above test that the sharpest aperture for this lens was f4.
Optimal aperture for 4/3 is generally f4-5.6. F8 is still quite useful, but generally not quite sharpest.
3. My own experience is that at least µ4/3 lenses out resolve the sensors. I expect this to be true of top line 4/3
lenses, as well. I have taken otherwise identical test shots with the E-M5 II at both the normal resolution and with the
~50 MP HR Mode. The considerably greater detail in the HR images is proof that the limit @ 16 MP is in the sensor, not
the lenses.
By using the 10 MP E-3, with roughly 25% less resolution than your E-M5, you
further limit the performance you experience.
4. I've said it before, and will probably do so again, shooting JPEGs further limits resolution compared to optimal Raw
shots properly converted. I've gone into the details before.
5. You present wildly different exposures. That might be fine for some sort of electronic eye, but human vision is very
highly sensitive to brightness, likely because brighter images allow more effective use of the cones in our retinas. In
any case, take the same image and present it light and dark, and most people will find the brighter one sharper.
6. By maintaining the same tripod spot for different focal lengths, you make the flowers different sizes in the samples.
Trust me, I know the difficulties of making the target the same size at each focal length - I've just been doing some
fairly extensive testing comparing various lenses myself (Using the HR mode of my E-M5 II). However, without that, the
vagaries of human vision make accurate comparisons hard.
Yes, I know that means the tests are carried out at different distances. (and logistics are a pain) However, those
minuscule possible differences in lens performance will be much smaller than the problem of different subject sizes.
7. The focal plane is not quite the same in the different images. Yes, it's hard to nail focus perfectly on exactly the
same part of a deep, 3D subject over and over again, with different lenses and focal lengths. But it makes a difference
in how we humans judge/compare sharpness.
Flat resolution test charts are really boring, and there's a lot about lens performance that they don't test. But for
simple questions of resolution, they are good for separating it from other factors and eliminating some of the problems
above. I am at the moment afloat in a sea of images of them.
========================================================
My personal conclusions:
You have made so many compromises in your test procedures as to make it essentially impossible to compare the resolution
of these lenses.
You may have a poor sample of the 50-200 ED, as that's always possible, but I'd not be comfortable making that call with
the samples presented.
You are comparing unprocessed JPEG images taken with an old, low resolution camera (with, as I may recall some other
problems?) to ". . . sharp, detailed photographs of wild animals and birds . . ." taken with up to date equipment, in
Raw, and expertly processed. I know you would like it to be so that simply taking the shot would result in the best
possible image, but it is not true. And it's never been true, post daguerreotype and tintype. Take a look at one of the
videos of St Ansel's darkroom work.
You would be amazed at the difference between original capture and the finished image in these photographs you admire -
and in mine. Yet, as with equipment and capture format, you eschew first rate process and editing software, and learning
how to use them.
The Oly 75-300 I used so extensively is a bit soft at the long end. As it happens, it's a sort of softness that responds
beautifully to deconvolution processing, resulting in nice, clear, sharp details. The data is there, but not in a form
our eyes can see, without it being reorganized. If you used it at 300 mm, you might well conclude it isn't that good,
yet the finished images are from good to great. I've posted here many sharp, clear, detailed photographs of wild animals
and birds taken with it.
"I had the idea that I would take sharp, detailed photographs of wild animals and birds that matched the wonderful
quality of some other brand equipment photos I've seen."
That's like buying a years old race car, filling it with regular gas, using street tires, and wondering why you lag
immediately far behind, and eventually are DNF.
Or perhaps like buying a used, out of tune, upright piano, relying on rusty skills from childhood lessons, and expecting
to make music as beautiful as a concert pianist playing a Steinway - right away, without the tools, blood, sweat and tears.
Moose D'Opinion
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|