Comparing macro lenses using MTF - invitation, take 3

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Miljenko
Posts: 171
Joined: Sat Jun 01, 2013 11:53 pm
Location: Zagreb, Croatia
Contact:

Post by Miljenko »

Lou Jost wrote: So for the short-focused Sigma LSA, the predicted neutral resolution (obtained by simply shrinking the 5x image to 4.5x) would be about 1890. You observed 1987, so yes, this tube lens does improve when short-focused.
Thanks for chiming in, Lou. The basic math is clear to me but sometimes real-life examples deviate from this rule of thumb. I've found such cases in the past and one of them I presented not so long ago (I remember you've seen it): https://www.photomacrography.net/forum/ ... hp?t=40770.
Besides strange resolution ups and downs there is always non predictable CA amount where tube lens can either improve or deteriorate it.
The good news is that sometimes incidental tube lens can act like manufacturer's dedicated CA correction TL. This happened to me with Nikon LU Plan 5x 0.15 and reversed Kenko #5 macro attachment lens. This objective has terrible CA measuring from 0.30 to 0.88 depending on TL used. With Kenko #5 it measured negligible 0.08!
All things are number - Pythagoras

Miljenko
Posts: 171
Joined: Sat Jun 01, 2013 11:53 pm
Location: Zagreb, Croatia
Contact:

Three 10x objectives tested

Post by Miljenko »

Last two lenses/objectives Filip has sent me for MTF testing are Nikon Plan 10x 0.30 160/0.17 and Reichert aus Austria 10x 0.25 160/-. In order to get the absolute feeling of their performance I decided to test them side by side to currently my only 10x objective, infinite Nikon Plan 10x 0.30 oo/0.

Image

Two finite objectives were extended on plain tubes for exact 10x magnification. Shorter tubes were tried for lower magnification/higher resolution but the edge performance suffered considerably along with pronounced vignetting. My infinite Nikon objective was stacked to Nikkor 200mm f4 AiS stopped down to f/5.6. This tube lens is not the best I own overall but somehow it gets the best out of this objective.
For starters let me mention slim Reichert objective. Simply put, this vintage piece of optics performs at level not worth mentioning. I've seen cheap Chinese 10x objectives performing better than this so European quality & performance is not always a universal rule :D

Image

On the other hand, Nikon objectives performed much better, except for one well known drawback of excessive CA. Although I usually suggest CA is easy to treat contrary to lack of resolution, when over certain amount blue and red tinted object edges are simply uncurable. This was true for both Nikon objectives, more so for the newer infinity model. This one had equal edge to edge resolution with just slightly curved focusing plane. Filip's finite variety had worse edge resolution but higher center resolution which is more important. One peculiar thing about this lens (generally rarely seen) is reversely (outwards) curved focusing plane. Curvature is mild (those are both „plan“ objectives) and no big deal anyway since we are always stacking at this magnification level.
This test concluded this series of lens testing and now I announce short brake for my free lens testing service offer. This is because my own lens/objective newcommers have piled up in the last couple of months (18 of them and couple more on the way), waiting for MTF testing. I have a strong feeling some of those might provide pleasant surprise. Test results will follow....
All things are number - Pythagoras

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

The basic math is clear to me but sometimes real-life examples deviate from this rule of thumb.
Yes, that's my point. It is important to know the expected change in resolution due to magnification changes, so that we have a baseline on which to judge the real-life variation from this.

Scarodactyl
Posts: 1619
Joined: Sat Apr 14, 2018 10:26 am

Post by Scarodactyl »

Miljenko wrote: The good news is that sometimes incidental tube lens can act like manufacturer's dedicated CA correction TL. This happened to me with Nikon LU Plan 5x 0.15 and reversed Kenko #5 macro attachment lens. This objective has terrible CA measuring from 0.30 to 0.88 depending on TL used. With Kenko #5 it measured negligible 0.08!
This seems like potentially the biggest news of the lot. You got an LU plan to perform like an apo (better than the mitutoyos in the other test) with a 15 dollar tube lens?

Maybe I am misinterpreting the CA numbers, since the normal .3-.8 range on the low end would also match the mitutoyo 5x, but that doesn't seem to be the usual experience with these objectives.
Last edited by Scarodactyl on Tue Feb 18, 2020 11:46 am, edited 1 time in total.

Miljenko
Posts: 171
Joined: Sat Jun 01, 2013 11:53 pm
Location: Zagreb, Croatia
Contact:

Post by Miljenko »

Scarodactyl wrote: Maybe I am misinterpreting the CA numbers, since the normal .3-.8 range on the low end would also match the mitutoyo 5x, but that doesn't seem to be the usual experience with these objectives.
With my constant test setup (Fuji X-T2, Capture One and Imatest) my sample of Mitutoyo Plan Apo 5x 0.14 at nominal 5x magnification produced between 0.072 (Nikkor 4/200 AiS) and 0.33 (Sigma LSA). As shown in earlier post, Mitty with Raynox DCR-150 when short focused for 5x produced CA halfway between those two figures (0.15). So according to my tests Mitty has on average 4x less CA than Nikon LU Plan 5x 0.15. According to my experience, CA area of about 0.3 pixels roughly separates visible and invisible CA.
All things are number - Pythagoras

Scarodactyl
Posts: 1619
Joined: Sat Apr 14, 2018 10:26 am

Post by Scarodactyl »

That makes sense--so the lu plan on the best previous setup about matches the mitty on the worst setup--but with the reversed kenko that 0.08 on the LU plan still looks pretty amazing.

jurkovicovic
Posts: 81
Joined: Sun Jul 16, 2017 11:16 am

Post by jurkovicovic »

Looks like the next round of tests is over.
The results of some lenses are really nice and surprising.
Everything is a little clearer now.
I would like to thank Miljenko for his work and the information he has given me. Thank you Miljenko!
canon EOS *

Miljenko
Posts: 171
Joined: Sat Jun 01, 2013 11:53 pm
Location: Zagreb, Croatia
Contact:

Post by Miljenko »

I'm glad if I could help you to select good lenses for your future macro work, Filip. You'll get full data Excel sheet with your lenses for the future reference. There were some nice discoveries as well as some disappointments, I believe.
Now you can concentrate on real macro shooting and I will do the same. Shooting insects waiting for so long in dozen of boxes will be pleasant change from boring resolution targets. :D
Best,
Miljenko
All things are number - Pythagoras

MacroLab3D
Posts: 96
Joined: Tue Jan 31, 2017 11:40 am
Location: Ukraine

Post by MacroLab3D »

I want to thank you Miljenko for the tests. It was a nice read and information. Based on that i bought Kenko No5 and Raynox 150 and compared them both with my Canon 200mm 2.8L II. Raynox 150 reversed looks almost as good as Canon 200mm 2.8L II and i am very happy to have custom alternative to the Canon. Kenko in other hand is not as good, which matches to your conclusions too. Awesome!

Edit: i made another test - installed Kenko No5 in normal orientation and it performed much better. They really close now. But Raynox still better a tad bit.
Last edited by MacroLab3D on Sat Mar 21, 2020 6:28 am, edited 3 times in total.
-Oleksandr

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

MacroLab3D, that's an important observation that your 200mm Canon L did better than the Raynox. Which version is yours? I see there are at least three.

MacroLab3D
Posts: 96
Joined: Tue Jan 31, 2017 11:40 am
Location: Ukraine

Post by MacroLab3D »

Lou Jost wrote:MacroLab3D, that's an important observation that your 200mm Canon L did better than the Raynox. Which version is yours? I see there are at least three.
Mine is Mark II.
-Oleksandr

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Thanks. Looking at the reviews, on paper the 70-200 zoom is actually better than this prime. I would not have guessed that it would be a great tube lens.

https://www.lenstip.com/327.5-Lens_revi ... ation.html

MacroLab3D
Posts: 96
Joined: Tue Jan 31, 2017 11:40 am
Location: Ukraine

Post by MacroLab3D »

Lou Jost wrote: 70-200 zoom is actually better than this prime.
This probably because i am on Crop and according to this test - the center is excellent. This is why i am preferring crop. It is always better in overall.
-Oleksandr

Post Reply Previous topicNext topic