Wide-angle: the Achilles' heel of SLRs?

[georg16nik]

Enjoyed the lens design story, thanks georg16nik.

As several have noted... In terms of ease of focus and composition... rangefinders are well suited to wide angle, while SLR's are better suited to telephoto.

Bill, You are welcome!

I don't see SLR lens designers fooling themselves that something could be improved (at a reasonable cost) in SLR wides. Probably most users are totally cool with distortion even in 50mm. After all, its not big deal if You shoot cats.

Nikon and the rest did their best and with current photo manipulation techniques, distortion is just "one of those things" to be taken care of.
The Zeiss Sonnar "look" is probably a menu item in some software by now.

IMHO, no RF lens improvements appeared in the recent decades.. the firm base was reached in the 50's.
If You print optically from film and You shoot RF LTM (or M39), then using the same camera lens on Your enlarger might be revealing.

 

[Steve Smith]

Very interesting how retro-focus can actually be an advantage.

I hink it's probably even more of an advantage with that alternative. electrical image recording technology.


Steve.

 

[Diapositivo]

I am referring specifically to geometric distortion, not wide angle perspective "stretching". The distortion I refer to is the failure of the lens to form a true rectilinear image due to asymmetry and the position of the diaphragm. These compromises cause either barrel distortion (straight lines "bow" outward, most visible along the edges of the frame), pincushion distortion (the opposite of barrel) or complex distortion (a combination of barrel and pincushion - typically barrel transitioning to pincushion toward the corners, hence the term "mustache" often used to describe it). Generally in retrofocus lenses barrel distortion is the common defect while pincushion distortion usually affects telephoto designs (to a lesser degree) although either type of distortion (or both) can be present in both long and short focal lengths.

Wide angle RF lenses typically suffer significantly less barrel distortion than SLR retrofocus wides.

I understand. I thought it was generally accepted to call barrel/pincushion/moustache distortion simply "distortion" (or lens distortion) while the "wide-angle perspective stretching" is I thought it were generally referred to as "geometric distortion" meaning it's the result of a law of geometry (nothing to do with the lens scheme or quality), but I see usage is not uniform and consequently there will always be a certain amount of ambiguity when talking about distortion of wide-angle lens.

 

[benjiboy]

I let lens designers worry about problems like this, I've never had any problem with unacceptable distortion on the images provided by my wide angle lenses on my SLR s even my widest Tamron S.P. 17 mm f3.5.

 

[Steve Smith]

I've never had any problem with unacceptable distortion on the images provided by my wide angle lenses on my SLR s

You're doing it all wrong. You're not supposed to look at actual pictures, just test charts and pictures of graph paper and parrallel lines!


Steve.

 

[markbarendt]

You're doing it all wrong. You're not supposed to look at actual pictures, just test charts and pictures of graph paper and parrallel lines!


Steve.

Steve, Steve, Steve (insert head shaking smiley here)

 

[E. von Hoegh]

Very interesting how retro-focus can actually be an advantage.

Perhaps there will be others with relevant comments on this topic. All I have ever heard is how 'compromised' the SLR wide-angles are but I see nothing but sensational results from such 'inferiority'.

Mark Crabtree: you are very correct to state that focusing them is a problem with SLRs but some are better than others. - David Lyga

Don't pay attention to everything you hear. My Nikon SLR WA lenses are superb - both of them.

 

[benjiboy]

You're doing it all wrong. You're not supposed to look at actual pictures, just test charts and pictures of graph paper and parrallel lines!


Steve.

Sorry Steve, I apologise for not being a gadget wanker, I often wonder if these test bench jockeys after all their tests ever actually ever take any worthwhile pictures

 

[mopar_guy]

Just use what makes you happy

I occasionally use wide angle lenses on my SLRs and I am not troubled by distortion. Then there are exceptions to any rule. For example there is the 18mm f3.5 Zuiko for the Olympus OM cameras. The following is taken from The OM System Lens Handbook from 1983.

A super wide angle, "orthogonal projection" lens
This Zuiko 18mm lens features both a super wide angle of view and, on the other hand, it is completely corrected for distortion aberrations so that it can be used for normal pictures just like other ordinary lenses. This means that straight lines in the scene are shown as straight lines no matter where they are located. This feature makes this an "orthogonal projection" lens that gives a correct image of the subject according to the lens perspective.

For more on this lens you can look here:
http://www.mir.com.my/rb/photography/hardwares/classics/olympusom1n2/shared/zuiko/htmls/18mm.htm

 

[georg16nik]

Being aware of and/or interested in how things work does not make someone any less a photographer, nor does it make someone a "wanker" of any kind. The wankers post to technical threads to remind everyone photography is about pictures. Brilliant.

1+
...and just talked about optical variety.

 

[markbarendt]

Being aware of and/or interested in how things work does not make someone any less a photographer, nor does it make someone a "wanker" of any kind. The wankers post to technical threads to remind everyone photography is about pictures. Brilliant.

The big problem with technical discussion is agreeing on the criteria to use as "the" standard.

A technical best can only be determined when the output requirements are defined.

If "pictures" means "artistic fun stuff for my walls" is the defined output there is one standard.

If "pictures" means essentially "technically accurate representations of a scene" the standard is quite different.

Us wankers can pick our own standards.

 

[barzune]

David,
Nikon came out with an excellent wide-angle lens, similar to the Zeiss Biogon in design (although with a slightly reversed formula), which you can see at http://www.mir.com.my/rb/photography/companies/nikon/nikkoresources/ultrawides/21mm.htm .

Unfortunately, this gem cannot be used on any but the more pro-line cameras, that offer MLU. That negates the "reflex" advantage, requiring a separate viewfinder and zone focusing.

While this neither confirms nor defeats the "Achilles' Heel" premise, it does allow for an adaptation that works.

Dan

 

[E. von Hoegh]

David,
Nikon came out with an excellent wide-angle lens, similar to the Zeiss Biogon in design (although with a slightly reversed formula), which you can see at http://www.mir.com.my/rb/photography/companies/nikon/nikkoresources/ultrawides/21mm.htm .

Unfortunately, this gem cannot be used on any but the more pro-line cameras, that offer MLU. That negates the "reflex" advantage, requiring a separate viewfinder and zone focusing.

While this neither confirms nor defeats the "Achilles' Heel" premise, it does allow for an adaptation that works.

Dan

That lens was developed for the Nikon rangefinder cameras, then adapted to the SLRs. It was replaced with the 20/3.5 Nikkor UD, a big, heavy, and superb lens which does not require MLU. The Nikkormat Ft, FtN, and some others offered MLU.

 

[benjiboy]

Being aware of and/or interested in how things work does not make someone any less a photographer, nor does it make someone a "wanker" of any kind. The wankers post to technical threads to remind everyone photography is about pictures. Brilliant.

I've been involved in photography for more than fifty years and in my experience very few of the many photographers I have ever known who were obsessive testers of their equipment and darkroom chemicals and processes ever took a worthwhile photograph, their raison d'être was the testing, more often than not producing technically perfect images with little or no meaning or soul in them.

 

[Diapositivo]

OP asked a technical question. I responded with a technical answer. Distortion is a practical difference between most RF and SLR wide angle lenses. Distortion is not typically a problem in landscapes but for those of us who often shoot subjects with straight lines parallel to the the edges of the frame distortion is visible and distracting. It has nothing to do with test charts and graph paper. Sorry.

I totally agree.

There are circumstances, taking pictures of architectural features, where straight lines are important and distortion, if present, can be very evident and quite disturbing.

https://www.zoonar.com/photo/detail-of-the-eni-building-in-the-eur-district-in-rome_830873.html

http://www.alamy.com/thumbs/6/{6ACD7C4A-9056-4073-8FF7-98E4EE6DA4DC}/BDE8MM.jpg

http://www.alamy.com/thumbs/6/{784A734F-242F-4B62-BF51-294ABB820C3E}/BWNKAA.jpg

 

[benjiboy]

I totally agree.

There are circumstances, taking pictures of architectural features, where straight lines are important and distortion, if present, can be very evident and quite disturbing.

https://www.zoonar.com/photo/detail-of-the-eni-building-in-the-eur-district-in-rome_830873.html

http://www.alamy.com/thumbs/6/{6ACD7C4A-9056-4073-8FF7-98E4EE6DA4DC}/BDE8MM.jpg

http://www.alamy.com/thumbs/6/{784A734F-242F-4B62-BF51-294ABB820C3E}/BWNKAA.jpg

These distortions are a red herring, because they not because of inherent lens aberrations but because the camera has been pointed upwards to cause perspective distortion because it wasn't capable of lens movements in relation to the film plane.

 

[markbarendt]

These distortions are a red herring, because they not because of inherent lens aberrations but because the camera has been pointed upwards to cause perspective distortion because it wasn't capable of lens movements in relation to the film plane.

Yep.

 

[Ian Grant]

I have to agree with Michael R

Being aware of and/or interested in how things work does not make someone any less a photographer, nor does it make someone a "wanker" of any kind. The wankers post to technical threads to remind everyone photography is about pictures. Brilliant.

The OP asked a valid question and there's no doubting that the best extreme WA lens sdesigns have been made fo rangefinder cameras and aren't made for SLR use (unless mirror lock is used).

That doesn't mean that there aren't good extreme wide angles for SLR use, it's just that they behave slightly differently. If they weren't there would be no need for Hasselblad to have made the SWC cameras, Nikon their special 2.1cm f/4-16 Nikkor-O etc.

The OP's asking about those differences and unless you've used both types of designs it is hard to appreciate the benefits of the purer designs of wide angle lenses for rangefinder cameras, or in the case of the SWC and Nikon lens where no mirror is used.

Ian

 

[darinwc]

Originally Posted by Diapositivo View Post
I totally agree.

There are circumstances, taking pictures of architectural features, where straight lines are important and distortion, if present, can be very evident and quite disturbing.

These distortions are a red herring, because they not because of inherent lens aberrations but because the camera has been pointed upwards to cause perspective distortion because it wasn't capable of lens movements in relation to the film plane.

Actually benjiboy is incorrect. Diapositivo was referring to barrel or pincushion distortion, where straight lines bend due to lens distortion.

Converging verticals, a result of the angle of the camera, is not 'distortion', it is just perspective. The lines of a building will still be perfectly straight.

 

[wiltw]

That doesn't mean that there aren't good extreme wide angles for SLR use, it's just that they behave slightly differently. If they weren't there would be no need for Hasselblad to have made the SWC cameras...

It should be kept in mind by all that the Biogon 38mm on the Hassy SWC was designed back in a day when retrofocus lenses indeed had design limitations in portrayal of rectilinear features without considerable barrel distortion, while the first version 40mm Distagon for the SLR exibitied distortion quite considerably. Later Hasselblad updated with a newer 40mm Distagon, which was a notably sharper lens than the older Biogon and which also greatly reduced lens abberations in spite of its retrofocus design.

Similarly, it used to be that zoom lenses were considerably inferior to fixed FL lenses, back when computers were not used for lens design. Today's zoom lenses often rival and even exceed the performance of even fixed FL lenses which have not been updated in the optical formula with newer computerized designs.

In short, what used to be true is not always still true today!

 

[AgX]

Actually benjiboy is incorrect. Diapositivo was referring to barrel or pincushion distortion, where straight lines bend due to lens distortion.

Converging verticals, a result of the angle of the camera, is not 'distortion', it is just perspective. The lines of a building will still be perfectly straight.

Perspective distorsion refers to extreme difference in reproduction scale within a subject. This will be due to the subject-camera distance being of the magnitude as the distances within the subject itself. This is easily achieved by employing wide-angle lenses.

Though those converging lines are a matter of perspective or rather the inclination between subject- and film plane, the angle of convergence can be influenced by perspective distorsion. Thus both phenoma can add to each other.

 

[LJSLATER]

The problem I have with 35mm rangefinders is the imprecise framing. I shoot a lot of buildings and I often create formal (symmetrical) compositions. I sometimes notice barrel distortion in photos made with my Nikon wides; the actual amount seems to vary based on focus distance, and the noticeability is highly dependent on the subject. Despite that, I can’t imagine doing what I do without a 100% accurate viewfinder with a grid.

Is there such a thing as an “architecture” 35mm rangefinder? Ideally, it’d have both the hot shoe and the tripod socket centered directly above and below the lens, so all one would need to worry about was parallax on the x-axis. Why do so many rangefinders have offset hot shoes and tripod sockets anyway? Are there any highly-corrected accessory viewfinders with grids?

The obvious solution would be to simply use a digital camera or move up to a larger film format, but it’s still fun to think about.

 

[Ian Grant]

Interesting questions. I use LF cameras handhe;d with the wire frame finders and wide angle lenses and shoot with a 6x17 camera and a view finder with no issues, I think you quickly learn the camears. In the months I shot with a 21mm Leitz lens and its finder I never had any problems either.

Ian

 

[David Lyga]

Wiltw quote: "back when computers were not used for lens design"

My question: What was the year that things changed? Or please give a short continuum of years defining the (sudden or gradual?) 'improvement' in zooms due to computer determination of formula, as opposed to manual determination. My rudimentary guess is the 'late 70s'. - David Lyga

 

[Diapositivo]

These distortions are a red herring, because they not because of inherent lens aberrations but because the camera has been pointed upwards to cause perspective distortion because it wasn't capable of lens movements in relation to the film plane.

What I mean is that those pictures DON'T show barrel or pincushion distortion because either it was not there or it was corrected (guess how). The perspective effect, the convergence of lines toward a vanishing point, is not "distortion". Why people lie this word so much that they use always use it?

"Distortion" is a lens defect and in images like those, when uncorrected, would show very clearly and would disturb the subject a lot.

Distorted means deformed, wrong, weird, not-straight. Geometric and perspective effects should be rendered with some other word than "distortion" because they are perfectly normal real-world phenomena and show nothing "deformed".

I understand that people use the term geometric or perspective "distortion" when talking about the exaggerated nose of somebody photographed with a wide-angle lens from short distance.

But in general how do we define "barrel-pincushion/moustache distortion"? Should we define each time "of the barrel-pincushion kind" to be clear?