As you are no doubt aware, there are many ways an interested party can consider photographic optical performance.
For going on thirty years I've been looking at camera system resolution. Now I would like to begin to consider how lenses transition from sharp to out of focus regions behind the point of focus.
There is a useful approach outlined by Mariane Oelund in a discussion thread on DPReview. For my look at the question I displayed a 2 pixel wide "dot" on a computer screen and photographed it with lens apertures set to wide open from sharp to increasingly out of focus.
Here are the results for a few lenses shot wide open, with and without a Lens Turbo II focal reducer.
Comments and Observations -
The immediate question is how to interpret what we see? Zeiss provides an good starting point. On a practical level, Jakub Travnik provides a useful illustration.
If an out of focus circle is bright around the outside edges, it is said to be over-corrected for spherical aberration. Some people refer to this as "soap bubble bokeh."
If the out of focus circle is evenly lit across the disk, it is said to be neutrally corrected. Zeiss currently champions this approach to lens design.
If the out of focus circle is bright in the center and less bright toward the edges then the lens is said to be under-corrected for spherical aberration.
With these things in mind, we can see that my Nikon Nikkor 50mm and 55mm lenses used with a Lens Turbo II focal reducer increases under-corrected spherical aberration. The Nikon Nikkor 105mm and 135mm lenses seem to be under-corrected for spherical aberration to very similar degrees when used with or without the Lens Turbo II focal reducer.
An important question is how does this work in the real world?
We already have a hint from the way the over-corrected spherical aberrations are interpreted photographically. That is, such lenses create "soap bubble bokeh" in the out of focus regions.
For neutrally corrected lenses we have two very good examples. One is Zeiss' own 135mm f/2 and, perhaps surprisingly, the Rokinon/Samyang 135mm f/2. These lenses are said to provide neutrally corrected out of focus renditions.
Lastly, the thing that started me on this quest (which I am only now beginning and am by no means understanding as fully as I would like) were comments made about the history of Nikon lens designs. Read carefully I found out that Nikkor designers from the outset deliberately under-corrected for spherical aberrations in certain designs (1950's Sonnar formula RF lenses - 5cm, 8.5cm, and 105cm as well as many standard and longer than stardard focal length lenses for the F-series SLRs). Nikon appears to have continued this approach up until rather recently.
Nikon designers felt that under-corrected spherical aberrations in the out of focus areas behind the point of focus lead to a "delicate rendition" of a scene. This property of a lens would change as the aperture was stopped down. One or two stops down from wide open would change a lenses character to being sharp across the field and quickly reduced/eliminated spherical aberration in out of focus areas.
At the same time I read the Nikkor lens history site I stumbled upon information provided by the Metabones adapter manufacturer. They suggest that their focal reducer adds a little under-corrected spherical aberration. While the Zhongyi Lens Turbo II I use was likely not designed by Metabones, I'm wondering if it too doesn't add a bit of spherical aberration to some lenses? Nearly all of my 50mm-ish lenses seem to "improve" their out of focus renditions when mounted on the Lens Turbo II.
I really appreciate Nikon's lens designer comments and feel they have given us an idea of why their lenses behave the way they do. If anyone knows of similar sources of lens design choices for Zeiss, Leica, Canon, and other photographic application lens manufacturers, please let me know.
For going on thirty years I've been looking at camera system resolution. Now I would like to begin to consider how lenses transition from sharp to out of focus regions behind the point of focus.
There is a useful approach outlined by Mariane Oelund in a discussion thread on DPReview. For my look at the question I displayed a 2 pixel wide "dot" on a computer screen and photographed it with lens apertures set to wide open from sharp to increasingly out of focus.
Here are the results for a few lenses shot wide open, with and without a Lens Turbo II focal reducer.
Comments and Observations -
The immediate question is how to interpret what we see? Zeiss provides an good starting point. On a practical level, Jakub Travnik provides a useful illustration.
If an out of focus circle is bright around the outside edges, it is said to be over-corrected for spherical aberration. Some people refer to this as "soap bubble bokeh."
If the out of focus circle is evenly lit across the disk, it is said to be neutrally corrected. Zeiss currently champions this approach to lens design.
If the out of focus circle is bright in the center and less bright toward the edges then the lens is said to be under-corrected for spherical aberration.
With these things in mind, we can see that my Nikon Nikkor 50mm and 55mm lenses used with a Lens Turbo II focal reducer increases under-corrected spherical aberration. The Nikon Nikkor 105mm and 135mm lenses seem to be under-corrected for spherical aberration to very similar degrees when used with or without the Lens Turbo II focal reducer.
An important question is how does this work in the real world?
We already have a hint from the way the over-corrected spherical aberrations are interpreted photographically. That is, such lenses create "soap bubble bokeh" in the out of focus regions.
For neutrally corrected lenses we have two very good examples. One is Zeiss' own 135mm f/2 and, perhaps surprisingly, the Rokinon/Samyang 135mm f/2. These lenses are said to provide neutrally corrected out of focus renditions.
Lastly, the thing that started me on this quest (which I am only now beginning and am by no means understanding as fully as I would like) were comments made about the history of Nikon lens designs. Read carefully I found out that Nikkor designers from the outset deliberately under-corrected for spherical aberrations in certain designs (1950's Sonnar formula RF lenses - 5cm, 8.5cm, and 105cm as well as many standard and longer than stardard focal length lenses for the F-series SLRs). Nikon appears to have continued this approach up until rather recently.
Nikon designers felt that under-corrected spherical aberrations in the out of focus areas behind the point of focus lead to a "delicate rendition" of a scene. This property of a lens would change as the aperture was stopped down. One or two stops down from wide open would change a lenses character to being sharp across the field and quickly reduced/eliminated spherical aberration in out of focus areas.
At the same time I read the Nikkor lens history site I stumbled upon information provided by the Metabones adapter manufacturer. They suggest that their focal reducer adds a little under-corrected spherical aberration. While the Zhongyi Lens Turbo II I use was likely not designed by Metabones, I'm wondering if it too doesn't add a bit of spherical aberration to some lenses? Nearly all of my 50mm-ish lenses seem to "improve" their out of focus renditions when mounted on the Lens Turbo II.
I really appreciate Nikon's lens designer comments and feel they have given us an idea of why their lenses behave the way they do. If anyone knows of similar sources of lens design choices for Zeiss, Leica, Canon, and other photographic application lens manufacturers, please let me know.