Media Technology

This third round of 24mm camera lens testing on the Sony a7R will focus on one item in particular, the Konica Hexanon AR 24mm f/2.8 prime lens. The first priority is to evaluate how much lens curvature interferes with picture quality, by using full width photos from a Sony a7R.

This lens is one of those mythical pieces of Japanese glass, from the halcyon days of film. Internet legend has it that Hexanon lenses were used by the Japanese government, as the standard against which all other lenses were measured. This particular 24mm lens is the latest f/22 version, reputed to have been made by Tokina, for Konica. You can find out more about these Konica lenses at:

http://www.buhla.de/Foto/Konica/Objektive/e24_28.html

Ease of focus at long distances: The focus ring on this lens is well damped, to the point of being stiff, and it has somewhere around 50-55 degrees of rotation. It was fairly easy to focus on the "5610" street number, 3 points.

Lens contrast: Certainly decent, but difficult to compare, since the conditions were cloudy. It's worth at least 2 points.

Sides of the images: f/8 is a disaster, and f/11 cleans up significantly, but none of it is up to par with the Canon FDn 24mm f/2.8, the current points leader. Zero points.

Center of the images: f/8 is slightly sharper than f/11, look just above the roof on the "5610" street number building.

Summary: This lens goes straight from f/8 to f/11, there isn't any half-stop detent point, which makes it feel a bit cheap. The addition of the f/22 aperture stop, that the earlier versions didn't have, doesn't do much to alleviate that impression. If you look at the lens diagram picture at buhla.de, it appears that there is more curved glass on this lens, verses the previous design. Could that have played a role in the blurry sides we are seeing? The third comparison picture, in the group shot below, shows how much the sides clean up with a very slight misfocus on the center "5610" number. Focusing around like that could be difficult, if not impossible, with the optical viewfinder on a DSLR, but it's easy with the EVF in the Sony a7R. The question is, can you make misfocusing a part of your shooting workflow, and is it worthwhile? Given that you don't have to do that with the Canon FDn 24mm f/2.8, probably not, so this lens gets kicked back to the consolation round.

Here is the test image, it's nearly 13mb in size, **be sure and view it at 100% size**:

http://www.codectest.com/images/24mmLensComparisonKonicaFocusPointTest.jpg

Dan Euritt

The Sony a7R and wide-angle lenses don't always play well together, as you might have seen in the 24mm lens shootout. The goal in that testing was to find a 24mm camera lens with a flat image plane, which means minimal lens curvature at the edges of the lens. These 28mm lens tests will have that same goal; find a 28mm lens that's not blurry on the sides, before starting to think about other aspects of lens performance.

The good news is that as the focal length of the lens increases, the more compatible the lens is with the a7R and other full-frame sensor cameras. The 28mm lenses that are being evaluated here tend to bear that out, with one exception, as you'll see in a minute.

These 28mm lenses are legacy film and/or DSLR camera lenses. The generic Albinar ADG 28mm f/2.8 lens was the cheapest. It's labeled as "macro focus 52mm", wherein "52" is the measurement for the filter ring. This glass was sold under a bunch of different labels, not just Albinar. It has the "MACRO 1:10 1:8 1:5 1:4" markings on the side of the lens barrel. It's not clear who OEM'd these lenses out for so many different labels, but that same macro number configuration is also seen on Sigma prime lenses.

Which brings up the second lens being tested, the Sigma 28mm f/2.8 MiniWide II. It also takes a 52mm filter, and it's about the same size as the cheap OEM lenses listed above. However, it handles differently, and the performance is not the same. Most significantly, the Albinar lens has a noticeable vignetting problem on the Sony a7R, even at f/10. You can see the dark corners, while the MiniWide II has clear corners at the same aperture.

The classic Vivitar 28mm f/2.0 lens rounded out the bunch. This glass has a serial number of 228*****, which means that it was manufactured by Kiron. It has a reputation for oily aperture blades, which two out of the three copies being tested did indeed have. However, when shooting with a mirrorless camera in dark conditions, a good WYSIWYG electronic viewfinder means that you don't need to open the lens up all the way to set focus, because the EVF automatically turns up the gain, to duplicate the camera settings you are shooting at. That superior trait of an EVF, over an optical viewfinder, means that oily blades that drag when opened do not interfere with getting the shot, as might be the case with a DSLR. Oily blades only become a problem when the blades stick so badly that they don't fully open up at every aperture setting. All three copies of this glass were made by Kiron, but one of the three copies was rejected because of haze on the inside glass. Both copies took pictures that were virtually identical to each other, so only one result was posted here.

Ease of focus at long distances: The Sigma 28mm f/2.8 MiniWide II probably gets the nod here. The lens is like new, and there isn't any visible side-to-side shifting of the glass while focusing, as seen when looking in the EVF at magnify levels. Both of the Vivitars did that, but it didn't seem to affect the final results. The Vivitar 28mm f/2.0 is a nice big piece of glass, but the wonderful long focus throw on it almost worked against it at these long distances. One point for the Sigma.

Lens contrast: 28mm lenses overall get the nod here, over 24mm lenses. Picking a winner among the three is tedious, because very minor exposure differences seemed to affect the contrast level. No clear winner, but the Sigma, with it's more modern coatings, might have a slight edge.

Sides of the images: Both Vivitars take a decisive win here. A flat image plane is so important when choosing wide angle lenses on the Sony a7R, it makes a big difference. The lens half-stop detent at ~f10 or so improved the sides slightly over F8, but focusing slightly past infinity didn't improve the sides much at all. It was a different story with the Sigma 28mm lens, if you hit just the right past-focus condition, the sides improve dramatically, and the center sharpness doesn't get penalized too badly. But why deal with that, when you can have a lens with a flat image plane. Even the cheap Albinar ADG 28mm lens had cleaner sides than the Sigma; in fact, picture quality on the sides was a complete surprise with the Albinar. Vivitar 28mm f/2.0(3 points), Albinar 28mm f/2.8(2 points), Sigma 28mm f/2.8(disqualified).

Center of the images: Nothing stood out at the long infinity focus distances seen here. YMMV, but no clear winner seen here.

Summary: The Vivitar 28mm f/2.0 camera lens moves forward into the next round, by a big margin. The cheap Albinar had a surprisingly flat image field, but it fell apart on the left side when focused very slightly past infinity, which indicates a possible lens issue. For what it cost, you can afford to take a look at as many of these lenses as you want. Just watch out for the vignetting with full-frame sensors, it's going to keep this lens from moving into the next round. The blurry sides of the image on the Sigma 28mm f/2.8 lens was a big disappointment, because this lens looked great and was a pleasure to use. It's a contender for APS-C format cameras.

Here is the test image, it's about 13mb in size, **be sure and view it at 100% size**:

http://www.codectest.com/images/28mmLensComparisonSigmaMiniWideIIalbinarADGVivitar.jpg

Dan Euritt

The goal of this comparison is to see how multiple copies from two different brands compare. We are looking at how the sides of the lenses compare in particular, since wide angle DSLR lenses are problematic on the 36mp sensor in the Sony a7R.

Since we only have six lenses total to compare, on one camera, using two different cheap adapters, it won't be possible to draw any statistically significant conclusions. The best that can be hoped for is for the viewer to see possible trends, and to get a good idea of what to look for when evaluating an image.

Sigma 24mm f/2.8 SuperWide II: If you have a Pentax background, you'll probably know that this lens is legendary. It was highly rated in the photozone.de crop sensor camera test, and it has a 4.0 rating on the old Photodo testing site. We are comparing two autofocus versions and one manual focus version, all with the Pentax k-mount flange. This prime lens was sold in just about every mount option that was popular at the time.

Canon FDn 24mm f/2.8: You won't have any problem locating one of these lenses, because Canon made a whole lot of them. Since there are so many, the pricing is fairly good, especially when you compare it to the astronomical cost of, say, some versions of the Minolta 24mm f/2.8 lens. All three of the lenses being tested here use the same Canon FDn mount, so they are the later versions, but this glass is also available in the earlier FD mount.

Results: Compare the "5610" street number on the back of the center building. On the left side of the images, look at the street sign hanging on the steet light. For the right side, perhaps the "USED CARS" sign on the side of the building. Canon wins for consistency here, the left side is basically the same in all three FDn 24mm images. The center varies somewhat, while the right side is perhaps the weakest for all three copies of the lens. That kind of consistency begs the question of whether there could be a Sony NEX to FDn adapter issue, or perhaps the a7R sensor isn't perfectly parallel to the lens mounting flange on the camera. Copy "B" will be used for further 24mm lens comparison testing.

The Sigma 24mm lens group generated mixed results, with "B" winning on the left side. "C" wins on the right side, but there was a thin cloud overhead, which cut the sunlight down over most of the image, changing the contrast. "B" advances to the consolation round, but just barely. The Sigma 24mm f/2.8 appears to have more vignetting issues than the Canon FDn 24mm f/2.8 lens does.

24mmLensComparisonThreeCopiesOfCanonFDn24mm.jpg

 24mmLensComparisonThreeCopiesOfSigma24mmSuperWide2.jpg

Dan Euritt

This second round of 24mm legacy camera lens testing will continue to focus on picture quality at the sides of the image, by comparing wide landscape test photos. As we saw in the first round of 24mm lens testing, there can be a huge disparity in picture quality at the sides, with the Sony a7R, to the degree that it overwhelms just about any other aspect of image quality. This issue with picture quality at the sides of the a7R images is far less of a problem at focal lengths of 35mm and longer.

A second copy of the Canon Fdn 24mm f/2.8 lens makes an appearance in this round, mounted on a different adapter, this time it's a black no-name Ebay adapter. The lens has a light haze over about a third of the front glass, and a sticky aperture ring. The other two lenses being tested are the Kiron 24mm f/2.0(Pentax k-mount), and the Olympus 24mm f/2.8. All shots were taken with the Sony a7R, on a monopod, at f/8, iso200-250. Both of those lenses are nearly perfect in appearance; no haze, no fungus, and clean snappy aperture blades.

Ease of focus at long distances: The Canon FDn 24mm lens has a significantly longer focus throw than both the Kiron 24mm and the Olympus 24mm, and it's noticeably easier to turn. Contrast is important for focusing wide lenses at extreme distances like this, and the Olympus 24mm may have a slight edge there. The Kiron 24mm optic is legendary, when you hold it in your hand, and marvel at how big the front and rear glass is, it's hard to be subjective. The diminutive Olympus 24mm lens is perhaps the best match for the small a7R, depending on what your preferences are. In the end, none of them had quite the same pop that the 24mm smc Pentax did in the first round, so it's two points for each camera lens.

Lens contrast: The test target being photographed in this round wasn't as helpful for looking at contrast. The "5610" number from the first round was at a closer distance, and the thin lines of that number, against the light background, made a decision about lens contrast seem easier. In this case, the focus point was on the highway sign, and it's numbers are unreadable, against a green background. We are left with looking at bushes on the hillside to the left of the sign, and the trees far away in the distance. The Olympus 24mm seems to handle vertical telephone and street light poles with slightly better clarity, in the center of the image, so 2 points for it. The Canon FDn 24mm may edge the Kiron 24mm out slightly for contrast, but since there aren't any half points available, 1 point for each.

Sides of the images: This is where the differences are much more noticeable, which is why it's so important. The second copy of the Canon FDn 24mm, with all of it's issues, still clearly wins by a long shot on the right hand side. The Kiron 24mm goes down hard on the left hand side, bringing up the question of whether or not it's decentered in that area. The Olympus 24mm could have issues on the right hand side, but it's fairly solid on the left hand side. If you look at the rocks under the bridge, on the left side side, the Canon FDn edges the Olympus lens out slightly, so it's the winner. Canon FDn 3 points, Olympus 24mm 1 point, Kiron 24mm(disqualified) zero points, it never won on any side.

Center of the images: Possibly the Olympus 24mm, but you'd have to analyze the RAW images to be sure. What you are looking at are cropped then recompressed camera original JPEGs, so it's difficult to be sure. The Kiron 24mm might be the weakest of the three here, but is it enough to hand out points on?

Summary: The Kiron 24mm(3 points) is out, but can it's sides be salvaged at f11? What about slight front focusing? Those are questions for the consolation rounds. The Olympus 24mm(5 points) moves forward, even with it's possible weaknesses on the right side. Out of the five lens designs tested so far, the Canon FDn 24mm(6 points) is showing signs of being the most compatible with the Sony a7R.

Here is the test image for this round, it's nearly 12mb in size, **be sure and view it at 100% size**:

http://www.codectest.com/images/24mmCanonFDnOlympusKiron.jpg

Dan Euritt