Home Search Gallery How-To Books Links Workshops About Contact
Zenitar 16mm Fisheye
Zenitar 16mm f/2.8 Fisheye. enlarge
I borrowed this from a friend. I've never bought one myself. You can't get them through the usual channels. I believe you can get them here for about $175 in the Canon mount or $225 in Nikon. You also can look here. It's made by Krasnogorsky Zavod somewhere in Russia.
It's designed for full-frame and film cameras. You lose the effect on smaller sensor cameras. You'd only get something like this on a DX camera, which is only mildly distorted and covering nowhere near 180 degrees:
instead of this, covering a full 180 degrees from corner-to-corner:
See also Crop Factor.
4.) Free soft focus effect at f/2.8.
1.) Hopelessly fuzzy at large apertures.
2.) No autofocus.
3.) Manual diaphragm: you have to open and close it by hand for each shot if you want to focus or compose.
3.) No manual focus confirmation indication on Canon cameras.
3.) No communication with your camera, which means manual guesstimate exposure on Canon.
Name: Zenit calls this the MC Zenitar-M2,8/16 Fisheye.
Focal Length: 16mm fisheye. It fills the frame of a film, FX or full-frame digital camera. It doesn't do all that much on smaller frame cameras because they only use the center of the image. See also Crop Factor.
Maximum Aperture: f/2.8.
Optics: 10 elements in 7 groups.
Diaphragm: 6 blade manual. Stops down to f/22 in half-stop clicks, which is a pain trying to count them without taking my eye from the finder.
Filter Size: Rear-mount M26.5 x 0.5mm thread.
Close Focus: 1' or 0.3m from the image plane (the back of the camera).
Size: 2.554" diameter x 1.954" extension from flange (64.87 x 49.63mm), measured maximums. The measured nominal diameter of the front engraved section is 62mm, too big for the 61mm Nikon fisheye cap.
Weight: 10.385 oz. (294.4 g), measured, naked.
If you stop this lens down to f/16 and are prepared for 1950's style manual operation on a full-frame or film camera you can get better results than I get with my Nikon 10.5mm fisheye.
Used at larger apertures this Russian lens is softer, and regardless of aperture it is always a relative pain to use, with no automatic anything.
The Nikon 10.5mm is very slightly wider on a D80 than the Zenitar is on a Canon 5D.
Full image of Zenitar 16mm on Canon 5D. Mouse over: 10.5mm on Nikon.
The focus turns with one finger.
It's solid and sturdy, but sloppier and grittier than Canon and Nikon manual lenses.
I find the easiest way to focus, especially since the diaphragm doesn't open automatically after your shot, is to focus by scale as we did in the 1940s.
Guess at the subject's distance and set it on the scale.
The distance scale is bizarre. It's in meters only, which is fine. The bizarre part is the arbitrary distances: Infinity, 1.2m, 0.6m, 0.4m and 0.3m. It's as if they had an English version marked at 4,' 2,' 1.3' and 1.' If I were the lens czar I'd mark it at Infinity, 2m, 1m, 0.5m, 0.4m and 0.3m.
Better than my Nikon 10.5mm! There is none with this Zenitar, and lots with the Nikon.
Here are the full guide images from which the crops are taken:
And here are 100% crops from the far left side. I enlarged the Nikon image slightly to make up for the the higher resolution and very slightly wider angle. This way they both match in size.
I see no differences from my other lenses. It's multicoated so this shouldn't be an issue.
Exterior: Enameled aluminum.
Focus Ring: Metal with rubber band.
Markings: Engraved nomenclature and focus scale. Painted index, apertures and depth of field scale.
Filter Threads: Metal (rear).
Noises when shaken: None.
Made in: Russia.
It's not bad, even at f/2.8. Fisheyes don't have the illumination issue of conventional lenses.
If you deliberately look for it, it's gone completely by f/8.
It's got an almost imperceptible darkening of the last millimeter of the corner at f/5.6.
It's got just a tiny amount of falloff at f/4.
It's got some falloff at f/2.8
It's almost impossible to find something evenly illuminated across 180 degrees with which to illustrate this, other than a photometric sphere or the overcast days we've heard about in California but have never seen.
As an example of the futility of this, here's some California sky. The dark band you see is the band of naturally polarized light at 90 degrees from the sun. The sky gets brighter naturally on the left and right sides of these images.
I only can see this on my 30" screen and looking in the corners.
If you want to avoid falloff, stick with f/4 or smaller.
If you want to avoid any detectible falloff whatsoever, stick to f/8 or smaller.
Remember that if you want a image sharp in those corners you need f/16 anyway.
Filters are weird little M26.5 x 0.5mm threaded ones that screw into the rear of the lens.
The lens is supposed to come with some, probably yellow, orange red and green for B/W film.
Zenitar Flare, f/16. Roll mouse over to compare to Nikon 10.5mm @ f/8.
These are at the usual shooting aperture for each.
The Nikon has a violet ghost in the lower left corner which, for the same exposure, becomes smaller and brighter (more noticeable) at smaller apertures.
The Zenitar had no obvious ghosts, but does have some veiling streaks radiating from the disk of the sun throughout the entire image.
The Nikon has a superior 7-blade diaphragm, which leads to magnificent 14-point stars if you stop it down. The 6-blade diaphragm of the Zenit leads to fewer (six) but bigger lines radiating from the sun.
The serial number is engraved and filled with white paint on the front side of the lens barrel.
It's awful at f/2.8 and improves as its stopped down.
It's sharp in the center at f/4, but isn't sharp in the corners until f/16 (!).
Diffraction softens it at f/22 as expected.
Let's have some fun and do a direct comparison to my brand new Nikon 10.5mm fisheye on my D80.
No one buys a fisheye because they want the center. Read on, and remember that fisheyes and wide angles' only reason to live are the sides.
There's not much going on in the center. The Nikon is sharp from f/2.8. The Zenitar is very soft-focus (spherical aberration) at f/2.8. It's not that obviously in this crop, but believe me, the Zenitar is comically soft even in the center at f/2.8.
By f/4 both lenses are sharp in the center. The Zenitar has the advantage because of the higher resolution of the Canon 5D. Sorry Nikon.
This is the last sharpest aperture for the Zenitar. It's been this sharp since f/4. The Nikon is starting to soften from diffraction. The Crop Factor of the Nikon's smaller sensor means these images are shown at 1.5x the size of the Zenitar/Canons, so diffraction is more of an issue. Math tells us that at f/16 the Nikon has as much of a diffraction issue as the Zenitar has at f/22.
Diffraction has softened the Zenitar, too., The Nikon is even worse, since diffraction remains the same on a lines per mm basis, and we enlarge each mm of the Nikon's image more due to the Crop Factor.
This is the fun part. We have a real horse race.
The Nikon images look almost identical for each f/stop. This is how the Nikon performs. I didn't get lazy and use the same clip each time.
The Zenitar is abysmal. The Nikon is as sharp as ever.
The Zenitar is better, but still abysmal. The Nikon is as sharp as ever.
The Zenitar is better, but still abysmal. The Nikon is as sharp as ever.
The Zenitar is better, but still worse than the Nikon was at f/2.8.
Now the Zenitar is better than the Nikon at any aperture. The Nikon's corners are crippled by its chromatic aberration. I can see the rope's winding in the Zenitar image. All I see are rainbows in the Nikon's image.
The Zenitar is at it's best. This is the side. I haven't been showing you the far corners, where the Zenitar was still pretty awful at the other apertures. The far corners are now sharp on the Zenitar.
The Zenitar is still a little sharper than at f/16 in the far corners, but it's softened a little in the center. The Nikon is softer, but its chromatic aberration has hidden the sharpness all along anyway.
Easy: This lens is a cheap, crappy lens. Like all cheap, crappy equipment, it you're willing to bend over forwards or backwards, you can get great results at f/16, and if you don't mind fuzzy, results are fine at larger apertures.
If you don't mind guessing at manual exposures and can shoot at f/16 all the time, get the Zenitar for a full-frame camera. The Zenitar looks great in these greatly enlarged torture tests, but in practice is a royal pain since you can't just shoot away.
No one prints fisheye images as big as I show here, which is the equivalent of 44" (1.1m) wide. Fisheyes are used for comic effect, and you'll lose the punch line if you're busy fumbling with the Zenitar.
If you want really sharp images, or want auto diaphragm, auto exposure or basically don't want to take 50 steps to make each shot, then save yourself the trouble and just get the correct Nikon or Canon lens for your camera. Get the 10.5mm Nikon for DX cameras, and the 16mm AF Nikon for Nikon film and FX cameras, and the 15mm Canon for Canon film and Full-Frame cameras. There is no fisheye for Canon 1.3x or 1.6x cameras, sorry.
Using the Zenitar 16mm Fisheye on the Canon 5D
Metering and Exposure
I found metering went bananas since the camera had no communication with the lens. I got different readings as I selected different modes!
Using the Av (aperture-prefered) exposure mode was an exercise in futility. I got greatly different exposures at different apertures.
I gave up and regained my sanity in manual exposure mode. I use successive approximation, which is looking at the LCD and adjusting the next shot accordingly.
Apertures are well calibrated. 1/60 at f/16 works great in full sunlight if you're guessing, or use a hand-held meter just like the 1950s.
The diaphragm is completely manual. It doesn't stay open while you compose and close down as you make each snap. You're back in the 1950s with this lens.
If you want to focus on the screen you'll have to open the aperture the hard way after each shot, and stop it back down before the next.
There is no electronic focus confirmation in the Canons, and fisheyes have such huge depth of field that there isn't much to see on the ground glass.
I prefer to guess the distance and set it on the scale.
For whatever reason, probably the lack of electronic communication, the auto white balance of my 5D went berserk.
Use any setting other than AWB. These shots were made in the Daylight setting.
If you find this as helpful as a book you might have had to buy or a workshop you may have had to take, feel free to help me continue helping everyone.
Caveat: The all the ads below come from third parties. I don't see them before they appear on your screen. See more at my Buying Advice page. Personally I get my goodies at Ritz (the store, not the hotel gift shop), Amazon and Adorama.