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

Can of Worms

Discussion in 'Open Discussion' started by chalkdust, Jul 23, 2016.

  1. chalkdust

    chalkdust TalkEmount Veteran

    282
    Sep 25, 2015
    Bert Cheney
    In a different thread about the upcoming Sony-Zeiss 50mm Planar 1.4, Tom02013 tossed in two links. I took the bait and looked at them. Though I was a physics major, that was many decades ago. I find that there is more "hand waving" than explanation in these links. Nevertheless, considering optics from something other than ray tracing (such as field) is not silliness, in my opinion. I am among those who see the variations in rendering of depth in the images given in those links.

    I have noticed this depth rendering for years and have sought it in images and in my choice of lenses. I have also noticed that it happens best with some classic lenses of few elements. The 5 element 3 group classic Heliar 50mm f/3.5 is my go to example for this. It often provides images in which the contours of surfaces look to me like they are unbound from the flat computer screen or flat paper. It seems to happen best when photographing things in the 1 to 3 meter range, I think.

    I am considering setting up a fairly well controlled comparison between my 50mm f/3.5 Classic Heliar and my modern 55mm f/1.8 Sonnar. I am still thinking about the best way to control variables in that (any advice from other lens comparison people as welcome).

    What do the rest of you think about these things? (looking at you tom02013)
     
  2. tomO2013

    tomO2013 TalkEmount Veteran

    375
    Dec 11, 2014
    Hi Chalkdust, sorry for the delay responding.

    I think the question that I was asking was whether adding more glass (while good for correcting things like spherical aberration) impacts the actual color and depth. I agree too, I would have loved to see a little less hand waving and a little more scientific evidence that is both measurable and repeatable.

    Truthfully I hadn't intended my post to shit-stir! That being said I was helping a buddy with a shoot yesterday in bright outdoor sun. I notice on DSLR bodies that often I prefer the rendering of slightly older designs that were designed with a particular aesthetic in mind over the more technically clinical. That being said, for a paid job I'd take the more technical design, with more accurate focus motor any day of the week! Older glass (e.g. FA Limited on K3ii tends to play havoc with the AF in bright outdoor sun if you are trying to shoot wide open. More modern glass has no such problemes) so I guess there is a practical element to this too.

    I'd love to see a controlled comparison between your 50mm heliar and 55mm Sonnar. I suspect that you will see a lot more vignetting and spherical aberrations on the heliar but that the bokeh will look smoother. :)

    Tom.
     
  3. cvt01

    cvt01 TalkEmount Regular

    164
    Jan 3, 2015
    I think I read every article on that site and I agree that it brings up some interesting aspects but the evidence is not the best. For example every time he bashes sigma he picks the world's most uninteresting subject lit by the flattest light possible at f/8 and pits it against his favorite lens and a photo of something interesting bathed in amazing light for comparison :D
    I still agree with the theory nevertheless, the more interesting question how much of the difference can be made up in post processing... like microcontrast for example.
     
  4. tomO2013

    tomO2013 TalkEmount Veteran

    375
    Dec 11, 2014
    Yup - on the flip side as the saying goes " it's easier to tone something down than big it up!". In that sense , you cannot add sharpness in post if it isn't there to begin with.
    I've found that many of the older glass that I use I have to be very selective of where I use it as spherical aberration while not as troubling for film is a pain in the butt for digital PDAF sensors whether on the focal plane itself or off plane.
    Basically if I am wanting to shoot wide open in bright outdoor sun, I'd favour the modern Sony Zeiss/Zeiss any day of the week. Studio, indoors, cloudy day, indirect sunlight then it would be a wash dependent on the look that I wanted to achieve.

    Tom.
     
    • Informative Informative x 1
  5. chalkdust

    chalkdust TalkEmount Veteran

    282
    Sep 25, 2015
    Bert Cheney
    I am working on this comparison. I set up a little still life about 1 meter from the camera. It is shot indoors with incandescent light from above. The camera was set up on a tripod and the tripod setup was not moved between exposures or lens change. Both were shot at 1/10 second and f/5.6 as indicated on each lens. Each lens was focused manually using magnification on the crossed arms of the figure. A ten second delay timer was used on the shutter. Both images are extra fine jpg from camera. No adjustments have been made other than clipping the cropped sections shown here from the original images.

    I am not using a person face because I believe we are all overly influenced by very subtle changes in facial expression. I am not at this time naming which lens was used for which shot because I think we are all overly influenced by brands and reputations.

    I see a very slight difference in the way each lens shows the contours of the surfaces. In particular the face of the figure has different flatness (has less or more contour detail) in the two images.

    Does anyone else see this? Which lens produced better contour information to your eyes?

    I do not think either lens produced a poor image.
    compare-02.
     
  6. roundball

    roundball TalkEmount Hall of Famer

    Oct 8, 2013
    USA
    To me...the left photo is better in all respects...brighter, sharper, more life-like, etc
     
    • Agree Agree x 1
  7. chalkdust

    chalkdust TalkEmount Veteran

    282
    Sep 25, 2015
    Bert Cheney
    The only thing in focus in either image is the little Cycladic figure of the person. Everything else is not in focus. The "one lens" image on the left is brighter. I think that may be why it shows less contour information on the figure's face than the right image. Both images show the same sharpness in the grain of the Cycladic figure.

    I need to come up with a better way to make the exposures more consistent with different lenses in order to do good lens comparisons. In this comparison, brightness trumped everything else. If I post process the right image to darken the face of the figure so that it matches the face on the right image, then the left image becomes dark and dull.

    Both Roundball and Ad chose the 55mm f/1.8 Sony-Zeiss Sonnar over the 50mm f.3.5 Heliar. (No surprise)
     
  8. addieleman

    addieleman Passionate amateur

    Nov 13, 2012
    Netherlands
    Ad Dieleman
    To be honest, I don't see too much of the subtle differences in (un)sharpness rendition. My preference is based on the general tone of the images, the right one looks murky in the shadows, an almost sure sign of a lens with lower general contrast.
     
  9. tomO2013

    tomO2013 TalkEmount Veteran

    375
    Dec 11, 2014
    Have you disabled in lens corrections for the jpeg test? A7 corrects vignetting for fe55?

    Sent from my ONE A2005 using Tapatalk
     
  10. cvt01

    cvt01 TalkEmount Regular

    164
    Jan 3, 2015
    The left is definitely brighter. I think before we can compare the two their brightness levels should be brought up to equal levels.
     
  11. chalkdust

    chalkdust TalkEmount Veteran

    282
    Sep 25, 2015
    Bert Cheney
    I continued my comparisons. Several times comparing 50mm f/3.5 Heliar 2 f/8 with 55mm f/1.8 Sonnar @ f/8 at object distance of about 1 meter.

    I found that though I think I see slightly better 3D from the Heliar, it is very, very slight. I could find it sometimes when I zoomed in close to 100% when viewing the image, but then it was highly dependent on precise focus. In other words, it is not something that can be depended upon - so effectively, for everyday photography, the difference does not exist.

    Heliar classic is 5 elements in 3 groups ; 55mm Sonnar is 7 elements in 5 groups.

    So I decided to increase the difference in elements and groups because that was the original hypothesis (more elements and groups tends to flatten the image). I compared an old Nippon Kogaku Nikkor-PC S mount 10.5cm f/2.5 with a Sony 70-200 f/4. Both were shot at f/8 for an object about 3 meters away. The 70-200 was shot at focal length of 113mm (set "by eye" close to the image given by the 10.5 cm).

    The old Nikkor 10.5cm is 5 elements in 3 groups. The new Sony 70-200 is 21 elements in 15 groups.

    Again, though there might be a slight improvement in 3D contours with the old lens, it is overwhelmed by all the other possible factors.

    I chose f/8 for all of these comparisons because I am not looking for subject separation due to DOF, but for 3D contour clarity within the focused regions.

    My conclusions:

    1. Improved lighting at time of exposure + careful post processing is better than counting elements for 3D effect in an image.

    2. The Heliar classic 50mm and the Nikkor-PC S mount 10.5cm are delightful manual focus lenses.

    3. There MAY BE a slight improvement in 3D effect with the older, smaller element count lenses - but it is not a particularly good factor for lens choice.

    So, for me, at this time, I am closing this can of worms by saying that the new, high element count lenses may sometimes flatten the image slightly - I have never found them to be more 3D than the older lenses - but often they are not less 3D either. Wise choices in lighting and post processing render the slight differences (in in-focus 3D effect as described here) insignificant.