[W.j.christy]

I am in the process of acquiring a zenitar 16 f2.8 lens. Does anyone have experience on how it will perform on a k3? I have searched for information on it but have only found speculation on how well it ot will do.

Any insight would be welcome

[stevebrot]

Mine does very well, though I don't have any on my Flickr stream for the combination. Let me see what I can find on my hard drive.


Steve

[Arjay Bee]

Worked fine for me but sold it when I moved to a Sigma 8-16. Centres are good but at edge sharpness and colour fringing deteriorates.

[stevebrot]

Mine does very well, though I don't have any on my Flickr stream for the combination. Let me see what I can find on my hard drive.


Steve

I was wrong...the photo below is on my Flickr stream, but is a crop:




I did a little look-see in my archives and found another that better shows what the lens is capable of on the K-3. I will PM you a link to a full resolution version of the crop.




Steve

[W.j.christy]

Thanks for the replies. I had readup on the lens on Ken Rockwell's website and he's quite negative about it. But hes pretty negative about most everything it seems. So I thought I would ask the folks who would best know how to use it.

[WPRESTO]

I looked into the Zenitar but never owned one (have owned three other fish-eyes). My impressions from reading numerous reviews: 1) obviously much less than 180 degree coverage on APS-C, coverage probably less than a Sigma 10~20 @ 10mm**; 2) center sharpness marginal @ f2.8 but improves significantly @ f4, and more at smaller apertures; 3) edge sharpness dismal @ f2.8 and improves very gradually; 4) best overall IQ @ f11 or 16 where it is quite good edge-to-edge; 5) at least some versions have a manual diaphragm = you must twist the aperture ring to stop down after focusing but before using the green button to set exposure; 6) very difficult to focus = take advantage of the enormous DOF at small apertures (f11~16) and guestimate the hyperfocal distance, or better: make a test run at different focus-distance settings on the focus ring and take notes on a card to carry along, unless your memory is better than mine (not difficult).

**On an APS-C body although you'll get far less than the 180 degree coverage, you will get the fish-eye barrel effect IF that is what you're after.

[W.j.christy]

I looked into the Zenitar but never owned one (have owned three other fish-eyes). My impressions from reading numerous reviews: 1) obviously much less than 180 degree coverage on APS-C, coverage probably less than a Sigma 10~20 @ 10mm**; 2) center sharpness marginal @ f2.8 but improves significantly @ f4, and more at smaller apertures; 3) edge sharpness dismal @ f2.8 and improves very gradually; 4) best overall IQ @ f11 or 16 where it is quite good edge-to-edge; 5) at least some versions have a manual diaphragm = you must twist the aperture ring to stop down after focusing but before using the green button to set exposure; 6) very difficult to focus = take advantage of the enormous DOF at small apertures (f11~16) and guestimate the hyperfocal distance, or better: make a test run at different focus-distance settings on the focus ring and take notes on a card to carry along, unless your memory is better than mine (not difficult).

**On an APS-C body although you'll get far less than the 180 degree coverage, you will get the fish-eye pincushion effect IF that is what you're after.

Not getting the full 180 degree field of view isn'tnecessarily a huge problem, though its something to consider. I am getting, what I consider, a deal almost to good to be true so at least I can try it out and see if I like it. I was hoping, and have had it answered that I could at least take decent full res pictures with it and make sure it truly would be worth my time if I picked it up.


I do appreciate you help though it does give me a good starting point to work from.

[Ontarian50]

Honestly, I think you're wasting your time putting the 16mm fisheye on an APS-C body.
I have a Nikon 16mm (same idea as Zenitar) but it's only a 180 degree fisheye on a full frame body. When I had an APS-C Nikon, it was worse than useless. There was no real fisheye effect, just a wide angle with bad distortion.
All the fun of a fisheye lens comes from the outer third of the image.
By all means get the 16mm Zenitar - but save it for a K-1.

[stevebrot]

Honestly, I think you're wasting your time putting the 16mm fisheye on an APS-C body.

Damn! All this time and I thought I was doing good work with my Zen on APS-C as well as 35mm film. Now I know that my tool is worse than useless. I am so discouraged


Steve

(..."only 180 degree" on FF?...I thought that was the traditional fisheye FOV, though I have been wrong before...)

[stevebrot]

There was no real fisheye effect, just a wide angle with bad distortion.

To be more serious, there is more to shooting with a fisheye than "fisheye effect". While many may buy one as a novelty lens, I use mine (both the Zenitar and Rokinon) as special purpose wide-angle lenses with specific optical characteristics that work well with many subjects.

The bending of lines is not barrel distortion. It is a design characteristic. "Fisheye" is the common name applied to a family of lens designs that use some form of circular projection. Most camera lenses use rectilinear projection (straight and parallel lines appear geometrically true). Those straight lines come with a price when applied to wide and ultra-wide angle lenses. The price is something called "volume anamorphosis" where objects assume unnatural, odd, or distorted shapes. While present through most of the field, this effect is most obvious towards the periphery.

Translation? A rectilinear projection is not consistently true for all aspects of geometry.

A good parallel might be the Mercator projection used for maps where places such as Greenland appear to be disproportionately huge. By using a circular projection, this distortion is largely avoided, though at the cost of non-straight lines in some orientations. What does this mean to the photographer? Consider a small set of talking points:

• The human vision is not rectilinear
• The field of human vision is not flat
• IMAX lenses are not rectilinear
• Much of what is termed "fisheye effect" is simply perspective resulting from the wide FOV. That is why a cropped fisheye image appear less "fishy".
• Straight/parallel lines are very rare in nature. Archeologists leverage this fact when doing aerial or satellite image surveys in search of ancient ruins/cities.
• As long as the horizon appears straight and level, landscape photos taken with a fisheye lens appear quite natural to most viewers
• Fisheye line bending depends heavily on the line's orientation to the lens axis
• In cases where the horizon and other lines appear bent, the scene may still appear quite natural to most viewers, if perhaps a bit disorienting
• For subjects where critical portions "wrap" the viewer, a circular (fisheye) projection will provide a more realistic representation to the viewer

It is that last feature that drives most of my fisheye shooting. Despite the obvious novelty aspect of shooting with the Rokinon 8mm on APS-C, I seldom use the lens in that way. I have used both it and the Zenitar for general landscape photography and for events (wedding receptions are an obvious) and other settings where tightness is a characteristic of the scene. While I have several excellent means to "de-fish" in PP, I seldom do so.

Back to whether a fisheye designed for FF is useful, on APS-C, I would give it a qualified yes. I can and have shot "fishy" with the Zenitar on my crop format cameras. I can and have shot "straight" with both the Zenitar and the Rokinon on APS-C. I have also shot "straight" with the Zenitar on FF 35mm film.

Zenitar 16mm on K10D...most viewers never notice the curve...If the lens axis is level to the earth, the horizon will always be straight, but I was not concentrating on the lens axis...




This Zenitar shot on K10D is very fishy, though you would be hard-put to notice if you had not done the shot.




Rokinon 8mm at my daughter's wedding reception...no alien heads in this scene...



When I show people that last one, they are always surprised when I mention it was shot with a "novelty fisheye" lens.

Steve

[WPRESTO]

V-A-V the comments above. It is true the human eye does not see things in rectilinear perspective, but our minds make images appear that way. Whether the rendering by a lens has "natural" perspective depends on where you place your eyes relative to the image. For a fisheye lens rendering of a scene, to make it look "natural" (= as your eyes would see it) you need to view the image from a matter of inches away, for example, you'd need to place your face almost on top of a computer screen turning slightly so your nose did not touch the screen in order to see an uncropped fisheye image in "natural" perspective. Fisheye lenses can/do serve two or three purposes: 1) to include far more of a scene than is possible to record with a single shutter click with any other lens; 2) to use fisheye distortion as a creative element; and sometimes, possibly 3) to obtain enormous DOF. A stitched panoramic covering 180 degrees horizontally looks very different from a fisheye rendering of the same scene because of, well, fisheye distortion.

[W.j.christy]

I love the photos and am grateful for the insight. I am excited to say I pulled the trigger and picked up the lens. Thanks for the help in deciding!

[stevebrot]

For a fisheye lens rendering of a scene, to make it look "natural" (= as your eyes would see it) you need to view the image from a matter of inches away, for example, you'd need to place your face almost on top of a computer screen turning slightly so your nose did not touch the screen in order to see an uncropped fisheye image in "natural" perspective.

This is quite correct, though a little exaggerated. Probably the best example would be sitting in the IMAX theater where the image is projected onto a hemisphere defining 180 degrees of arc in all directions.

Lacking an IMAX screen, sometimes it is enough to simply add a strong horizontal crop or tilt the frame so that the composition leads the eye to one of the corners where the projection is strongest. It may also depend on the person viewing the photo. I tend to see with a wide rather than long eye when just walking around. Other people tend to have a stronger central focus. If I stop to actively consider my peripheral vision, it is possible to see the leaning trees and curved lines.


Steve

[W.j.christy]

So I have taken some pictures with this lens and find it to be very difficult to achieve any kind of sharp focus. It also seems that I must use the filter in the back to ensure that I can come somewhat close to infinity focus. Not sure if I am doing something wrong or my technique is bad but this seems like a very difficult lens to use more so than other manual lenses I have used. Also the metering seems off as well.

I have tried the lens stopped down to 11 and 16 and have achieved better results but overall this lens just feels very soft. Is that a typical characteristic or am I doing something wrong.

My technique

Hand held
Shutter speed 120 or above
ISO 120 - 500
Aperture 5.6. - 16 with the best results at 11-16 which I know is common.

Any advice?

---------- Post added 04-09-16 at 08:30 AM ----------

So I have taken some pictures with this lens and find it to be very difficult to achieve any kind of sharp focus. It also seems that I must use the filter in the back to ensure that I can come somewhat close to infinity focus. Not sure if I am doing something wrong or my technique is bad but this seems like a very difficult lens to use more so than one manual lenses I have used. Also the metering seems off as well.

I have tried the lens stopped down to 11 and 16 and have achieved better results but overall this lens just feels very soft. Is that a typical characteristic or am I doing something wrong.

My technique

Hand held
Shutter speed 120 or above
ISO 120 - 500
Aperture 5.6. - 16 with the best results at 11-16 which I know is common.

Any advice?

[Arjay Bee]

It also seems that I must use the filter in the back to ensure that I can come somewhat close to infinity focus.

My understanding is that that is not just a filter but an integral part of the lens structure.

Apologies for my terse reply above as it was from phone input. Back at home now with a proper keyboard and access to samples, both film and digital.

Attached are some samples from the lens:


K20D


K20D


K2 (film)


K2 (film)