1. Reminder: Please use our affiliate links for holiday shopping!

Lens comparison (test) AIS 2.8/180, 2.5/105, 1.2/50 vs. 1.7/20 Panasonic

Discussion in 'Adapted Lenses' started by HoB, Jun 3, 2012.

  1. HoB

    HoB Mu-43 Rookie

    15
    May 31, 2012
    When I got my new camera, I was eager to see how well my old AIS Nikkors would do on the OM-D and how they would compare to a native lens.

    Secondly, I was curious as on the site www.lensetip.com :43: lenses score a higher resolution. The site points out that you can compare lenses only if they were tested on the same camera so I was eager to see for myself. In addition, I remembered from the time of film-photography, that lenses with a resolution of 40+ lines/mm was considered already very good, now the :43: score resolutions of 80+ lines/mm and I wanted to know wether it was something funky going on with the sensor or whether it was really the lenses.

    Please keep in mind, that while I have tried to control image parameters, but this is by no means a professional test. It is rather a comparison, with my personal commentary.

    Personally, I had my money on my Nikkors, so when I did the first test on JPEGs, I was somewhat shocked to see the Panasonic 1.7/20 blow the Nikkors out of the water. So I suspected the OM-D to recognize the Pana and to apply in camera corrections to the images. I've repeated the test with RAW images and developed them in Olympus Viewer 2 without any corrections to the images (all "as shot"). I am not very familiar with RAW development, so Olympus Viewer might have still snug some "unfair" processing into the images, but Sharpening (as well as all other parameters where this applies) was set to "as shot" and "use lens information" was unchecked.

    For each lens I did sets from wide open to f8.0 (except for the 50mm, for which this is pointless, I went to 5.6). I picked the best performing aperture from each set. If someone is interested in wide open, I can upload some of them as well.

    Obviously, I tried to maintain the same viewing angle for all lens by moving closer to the object for the shorter wavelengths. It worked quite well, as I was able to get the same crop with 973x729 pixels (shown below). I also tried to keep the optic axis as perpendicular to the object as possible, I picked intentionally a printed pamphlet. I wanted a flat object to avoid depth-of-field problems (these are really fast lenses and focussing is tricky enough as it is). I also wanted a text section and especially white on red background (not shown here). My experience has been that this is particularly difficult to image properly (might of course be just my imagination :) ).

    The images are 100% crops about 1/4 down and right of center, with the exception of the first image, which is a close-up with the Pana as REFERENCE to show, how the test object looks like (in particular the details of the printed image) to show the details that the lenses COULD have picked out in the test shots. It is not a crop but rather a resize.

    So, without further ado:

    REFERENCE:
    P6030042.JPG

    AIS 2.8/180 ED @ 4.0
    P6030020_copy.

    AIS 2.5/105 @ 2.8
    This might not be the best example, as I have dropped this lens pretty badly a few years ago and it may have gotten damaged. On the other hand (and this applies to all of the tested Nikkors) the image quality in real-world shots is still pretty good, as the picture in the "Adapted Lens" thread of our cat shows.
    P6030024_copy.

    AIS 1.2/50 @ 2.8
    I picked 2.8 though there is no appreciable difference between 2.0 and 2.8. Ken Rockwell's statement that this lens reaches max. performance at 2.0 seems correct, as far as I can tell
    P6030031_copy.

    Panasonic 1.7/20 Pancake
    And finally the Panasonic. I may be partial to my old Nikkors, but this is just no contest. The exception being distortion (not shown here), where it doesn't stand a chance due to its focal length.
    P6030037_copy.

    Commentary:
    Resolution is not my primary measurement of IQ. In fact I am a bit disappointed that sites like lenstip (which I really like) use mainly resolution as primary IQ measurement (apart from distortion, vignetting etc.).

    About 30 years ago, Zeiss pointed out, that they do not optimize their lenses for resolution but rather for max. contrast up to 20 lines/mm, as structures of this size have a much bigger visual impact as the really fine details. So in addition to the measurement of max. lines/mm at 50% contrast, I would wish that sites like lenstip or photozone (another excellent page in my opinion) would add a graph of max contrast at 20 lines/mm or (since in :43: the magnification to displayed image is twice that of a full frame, or the old 35mm-film) for :43: rather max contrast at 40 lines/mm.

    What strikes me in this comparison is that the Panasonic does not only show higher resolution as the Nikkors, but that it beats them contrast wise on the larger structures as well (and by a significant margin) - just look at the curly hair of the boy in front, or compare the printed text.

    That's it. Sorry for the long post. I am new here. I hope this was ok.
     
    • Like Like x 2
  2. heli-mech

    heli-mech Mu-43 Top Veteran

    959
    Mar 9, 2012
    Vancouver Island, Canada
    Andrew
    Interesting but a couple questions:
    Tripod used? IS turned off if on Tripod? Self timer used? Why the different exposures 180mm is 1/50 F/4 where as the 20mm is 1/10 f/4?

    One big variable is manual vs auto focus. Did you try "Focus bracketing" to ensure you are getting the manual lenses optimally focused?

    I haven't tested my 180 against my native lenses but I did with my Tokina 90 and the Tokina was sharper with equal contrast.
     
  3. HoB

    HoB Mu-43 Rookie

    15
    May 31, 2012
    Tripod: Yes, of course

    IS: Set to IS1 - on purpose. In my first trial I had IS off, same result. But in a real life shot, I use IS1 as well, so I decided to use IS1. I am really impressed with the IS of the OM-D. I did match the focal lengths for the IS system.

    Self timer used: Yes

    Different exposure: This was taken with the newspaper tacked against a wall underneath a skylight (I wanted to use natural light). It was an overcast day. The 180 was the first one up. During the comparison, the light actually darkened - not really all that noticeable to me but I noticed the exposure time change after the 180 (I record the exposure settings the old fashioned way with pencil and paper). It was constant for the following 3 lenses though. But this could indeed mean, that the 180 image might not be completely comparable contrast wise to the other three (by showing more contrast). Does nothing for the exceptionally good results of the 1.7/20 though.

    Manually vs. Autofocus: Who says I used autofocus :)? Seriously though, focussing is a big issue. The depth-of-field is so shallow on the 180 and the 50 wide open that focussing is extremely difficult. On the other hand DoF increases noticeably stopping down, so for the apertures shown, focussing is much less of an issue. All lenses were manually focussed (including the Pana) using the focussing magnifier at max. magnification and I took a loooong time ensuring that I got it as good as could be (again, in any real life shot you will not take as much time as I did for this comparison). This was actually the main reason why I redid the comparison. The 1.2/50 poses a particular problem. You always want to focus wide open, to utilize the narrow DoF to maximize focussing precision, which can then only get better when stopping down, but the image quality of the 1.2/50 wide open is actually too soft to get maximum focussing precision. So this was the only lens which I focussed trying different apertures.

    One remark about both the viewfinder as well as the rear screen of the OM-D. They are amazingly good, especially when focussing MF lenses with the magnifier. There was in fact, very little in the final images, which I didn't see already in the viewfinder or the rear screen.
     
  4. heli-mech

    heli-mech Mu-43 Top Veteran

    959
    Mar 9, 2012
    Vancouver Island, Canada
    Andrew
    I was intrigued so I did a "quicky" test with my ED 180mm and 25mm 1.4. Virtually identical, of course mine was much less scientific that yours. At first the 25mm had a slight sharpness advantage but a little less contrast. However I could add sharpening in PP to the 180mm shot and bring out the same detail, however there was no perceived advantage to sharpening the 25mm.


    I swear the camera bumps up sharpness and contrast with native lenses. I seem to have noticeably more PP room with my legacy lenses, whereas with native lenses they are nearly as good as they will get. (JPEG)


    I do wonder the merits of comparing lenses of such different focal lengths. I really don't understand the physic's of light enough to say but you would think there would have to be some advantage to being 3 ft away vs over 20ft. It is interesting to try though.