On 2/19/2022 12:07 PM, Mike Gordon via olympus wrote:
<<I'm not sure I "get" this technique. Not only visually disorienting, but
highly
<<subjective and non-repeatable/comparable.
https://phillipreeve.net/blog/how-to-check-how-decentered-your-lens-is/
It is fast, easy and works well enough for the intended purpose.
When I say 'I'm not sure I "get" this technique.', I understand how it works. What I don't "get" is why one would use
it, if aware of a better alternative.
Fred on FM devised a slightly modified version of this technique, but they all
work about the same. Decentering is a most common malady in wide fast lenses.
Sigi seems to have the best QA but about 1/3 of pixel peepers on FM have sent
back a 14-24/2.8. Fred couldn't even find a centered copy of teh Sonnie 14/1.8
and gave up. The Siggie at f2.8 is as good or better than the 14 prime.
<<<You tell me; can't you clearly see the differences, wide open, between these
highly rated and very similar lenses, including
centering?<http://www.moosemystic.net/Gallery/tech/Lenses/PanvsOly25mm/PanvsOly25mm.htm>
Nothing wrong with that methodology.
My argument is that there's something better about it.
The Oly is decentered, BTW, with the left side being weak especially the left
lower.
1. See how easy that is? A few moments, and you can see it all.
2. See how easy a comparison is? A few moments of looking and I know that the Oly is slightly sharper in the center, but
poorly enough centered that two corners are really bad. The Panny is well centered and slightly better in the corners.
(As I recall, Ctein saw no difference in center sharpness with his samples.)
If I were looking for an ideal copy of a Blogogon 8/1.0, all it takes is a coupe of minutes and a few shots, to document
each one, as I rent or buy and return. No adjustment/accommodation for different foliage, light, etc. as with this
outside technique.
Oly tends to use designs less susceptible to small perturbations, so they don't
have too many lemons.
Is this from Roger? I'm wondering how someone outside of Oly production
engineers can know this, especially statistics.
Sigi appears to use more checks during manufacturing to reduce sample variation
an allow more tweaking. Roger from lensrentals has said it is cheaper to
replace a decentered lens than dissemble and attempt to fix it. Whether they
toss the returns or repackage them to less fussy customers, I have no clue.
The economics of this stuff are tricky, and unknowable from outside. Way back at the early stages of digital, local
Canon repair centers were replacing modules. The bad modules were then all sent to one world wide, low cost place, where
locals were trained as specialists in repairs of only specific modules. I have no idea if they still do that. The point
is that repairs by generalists in high labor and operating costs countries would be much more expensive than the above
scenario.
My first father in law was always looking for ways to make money and rather good at it. He was working on a scheme to
gather watches brought in for repair to local jewelers, ship them to the Philippines for repair, then back. Probably
would have made good money, had not digital watches arrived on the scene.
Roger has also stated MFT lenses have more sample variability than others.
Might this be related to scale? Missing spec by 0.01 mm makes a larger
difference for 4/3 than FF?
I reprocessed the 7mm shots in DXO and cancelled the linear distortion
correction that cannot be avoided with ACR.
OK, this raises a question. Why does linear distortion matter - for this specific purpose? If the stars are nice
pinpoints, does it matter if their locations differ a bit from a true star map?
I included a tad of deconvolution sharpening. The lens looks centered, but
soft in the corners. I had manually focused with max mag with the chimney in
the center of the frame. I bet it is the field curvature doing the damage.
(back to) 3. Again easy peasy to check with the target. If using PanOly, focus too close and take a focus bracket
series. A run through in LR, FastStone, etc. will quickly reveal field curvature. For Sony, etc., move the AF point from
center to middle distance, then corner. All will be revealed. No need to speculate. (The OM 21/2 was notorious for field
curvature at close focusing distances.)
I had hoped avoiding the distortion correction would sharpen it up a bit as
it avoid pixel torture. Moose had previously posted the definitive treatise
on the algorithms used for distortion correction. I have never ever seen any
more informative evaluation than this.
https://www.mlarchive.de/Olympus-OM/2016-09/msg00059.html
Thanks!! Neither have I. 😅
Here are the new images.
https://www.olyendomike.com/Oly-7-1428/i-98WZKTR/A
https://www.olyendomike.com/Oly-7-1428/i-hBvK8Lr/A
https://www.olyendomike.com/Oly-7-1428/i-m3xPPR8/A
https://www.olyendomike.com/Oly-7-1428/i-KWwQqB2/A
Too soft for me, Mike
Nope, still mystified at what these tell me, without some comparison or
standard.
Standardized Moose
--
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/
|