C.H.Ling wrote:
> ----- Original Message -----
> From: "Moose" <olymoose@xxxxxxxxx>
>
>
>> I believe sample variation is unavoidable. In factory new units, I believe
>> it was quite small, and probably insignificant for all but the occasional
>> outlier. With used lenses, I think the possibility of a below average sample
>> is increased, as the effects of handling show up in some samples.
>>
> I have disassembly and clean more than 15 different OM Zuikos, I have
> confidence on the OM lens construction. Unless the lens has been dropped,
> wear out or dry out without proper grease in between the threads (serious
> one may cause left/right resolution unbalance) otherwise sample variation
> shall be rare.
I think we really agree. Significant sample variation was rare new, and
may be less rare with misused used lenses.
> ....
>
> This does not apply to the latest DZ lenses, I had a 14-45 which is one side
> sharper than the other and a 11-22 which had some play in front element group
> due to wear out by zoom action and unable to focus to infinity at 18mm when
> it is slight pull out.
>
Unfortunate.
> ......
>
>>> My experience was based on the same sensor (E-1/E-3) so APS vs FF does not
>>> apply
>> I believe it does, as you are using results with an even smaller sensor to
>> compare to results with FF film. If you magnify each the same amount, so
>> that the 4/3 image is half the height of the film image, there's no problem.
>> If you magnify the 4/3 image twice as much, the the image with the same lens
>> may look quite different from the two systems.
>>
> I'm comparing different lenses on the same 4/3 body, the best one will still
> be the best in FF
I think this is not always true in the digital realm, with digital
resolution and anti-alias filters. I believe it is possible for a lens
to "out resolve" the sensor system on a large sensor, but be "out
resolved" by a small senor system. If another lens out resolves both
systems, it will equal the first lens on FF, but be superior on 4/3.
> (at least the center, that matter more in this comparison).
>
Again, I disagree. The center may matter more for some subjects and
uses, and not for others. According to Gary, the 18/3.5 "... lens design
emphasizes outer zones at expense of center image zone (which gives a
sharper overall impression of an image)"
Dpreview says of the C & N 70-200/2.8 lenses:
"The Nikon lens clearly outperforms the Canon for sharpness on the
smaller DX/APS-C format, however this comes at the cost of rather
compromised performance on full frame, with significantly higher
distortion, vignetting and chromatic aberration, plus extremely soft
corners. This leads us to conclude that the two lenses were optimized
differently, the Canon for full frame and the Nikon for DX, and
illustrates how the different demands of the two formats appear
difficult to reconcile in a single lens design."
Do "... significantly higher distortion, vignetting and chromatic
aberration, plus extremely soft corners." not matter? I'm not claiming
that is the case with the lenses at hand, only that the generalization
may not be valid.
>
>> If the test was indeed performed using an OM-1, whereas the test I referred
>> to was performed with an OM-4, mirror and aperture prefire and the Bogen
>> lens support, that alone would account for the differences.
>>
> I doubt they are using sample printed resolution chart and a camera body for
> the test, I think pop photo using special equipment.
That's correct, but I suspect Modern was still using film shots of test
targets in 1973.
> BTW, high speed flash illumination will eliminate all vibration problem.
>
Of course. And if one is only going to shoot with flash, there's no
problem. I almost never use flash and the images that started this all
didn't either. If Gary had used flash, his tests would indeed only have
tested the pure optical quality of the lenses. How much more useful to
have practical tests that show what real world problems the lens/camera
combinations have and solutions to minimize the problems.
>> ......
>>
> Thanks for the example, but it has nothing to do with 180/2.8 and 200/4
> comparison or if 200/4's performance is on par with other OM teles. Different
> people has different demand on lenses quality, some are easliy
> satisified.
>
Good luck with claiming that AG isn't fussy about lens IQ. :-)
> My 200/4 is MC, cosmetic 75-80%, perfect glasses with smooth and tight
> focusing, my 180/2.8 is 70% cosmetic, very little scratches on front
> glasses, I have used light grease to make the focusing easier. Both should
> have no problem in optical quality.
Maybe we will have a chance to see how my 200/4, which is about 98%
cosmetic, performs. If I can overcome my inertia.
> My 200/4 is not bad but just not as good as 180/2.8.
>
But by how much? And for what uses would the difference be significant.
See below.
> ....
> It is now totally different, using live view and flash, error is absolutely
> minimized. Here is a test I just made, tripod mounted 40D, 430EX bounce off
> ceiling, live view, 2 second shutter delay. At least three shots for each
> lens with re-focus and selected the best one (actually very little difference
> for the same lens). Same development parameters, WB 5200K, sharpness=2. Both
> lenses set at F4, the results speak for themself.
>
Yes, they do indeed. However, I'm not sure they tell the story you think
they do. As you pointed out before, perceived sharpness isn't actual
resolution.
I saw the same thing you did when I first looked at the tests images.
Then I did a little experimenting. I used PS to increase the contrast,
so only actual resolution, not a combination of resolution and contrast,
would be considered in looking at the sharpness of the samples. My
conclusions are:
- The 180 is sharper at the bottom of the subject, but not by a great deal.
- Over most of the central area, it's a toss-up. If I mixed up sections
of prints, I don't think anyone could tell which is which.
- On the upper end, the 180 is slightly sharper.
It's actually sometimes easier to see which is sharper by the size of
the type. Where the 200/4 is less sharp in the bottom corners, the
slight spread of the black areas ends up making them bigger when
contrast is increased and black is clipped.
<http://www.moosemystic.net/Gallery/tech/Lenses/Zuiko200-180/200-180.htm>
It's hard for me to imagine an application in which I would use either
lens where such small differences in resolution would be significant. I
understand that apparent sharpness differs by more than the actual
resolution difference. The 200 obviously has lower contrast. However,
that's not the issue, at least to me, that it was in the film era.
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|