Archive | Lenses and Optics RSS for this section

All SLR lenses are Coke® bottles, right?

L1009734

Click to enlarge. The eyes are the focus point. What lens took this?

The advent of digital photography has made a couple of things clear: (1) many pros did not have so much talent as ability to overcome barriers to entry and (2) much of what you were told about lens quality – in terms of SLR versus rangefinder – was (or is now) untrue. This second point bears some examination.

What is the state of play on SLR vs rangefinder lens quality? The perception of SLR versus rangefinder lenses was developed when both shot on film, and there has been a major reversal of fortunes. Film was not sensitive to the angle of incidence of light coming from the back of the lens, and because rangefinders did not have mirrors, lens designers could make symmetrical lenses whose rear elements might sit just a few millimeters from the film surface. This knocked out distortion, incurred a little bit of vignetting (which was largely absorbed by the latitude of negative film, and resulted in a compact package.

SLR lenses, on the other hand, had to design around mirrors. So lenses under 50mm generally had to start with a longer focal length and then compensate it down by introducing a negative element in the front. This retrofocus arrangement generally compromised distortion and sharpness slightly, but it produced a good enough result that SLRs were able to exterminate rangefinders as mainstream cameras. But today, when the imaging surface is a flat sensor with a Bayer pattern, chromatic aberration, angle of incidence, color shift, and vignetting became big issues for traditional rangefinder lens designs. Even Leica’s very expensive wide-angle rangefinder lenses, on Leica’s very expensive bodies, were now capable of returning disappointing results in terms of color shifts and vignetting.

The goal today is sometimes called telecentricity, which is commonly understood to be the situation where light rays hit the sensor parallel to the lens axis. It is still achieved by retrofocus designs. It is telling that many Leica and mirrorless wide angles that avoid color shift and vignetting are creeping up in size to SLR lenses. Witness Leica’s fast wide-angle lenses, which are quite large – especially when you compare aperture to aperture. A 21/3.4 Super Elmar has a 46mm front thread; the 21/2.8 Elmarit-M has 60mm, which is only a hair smaller than a 20mm f/2.8D Nikkor (at 62mm). But nowhere is this phenomenon more stark than in Fuji XF lenses, where the register is shorter, lenses cover an APS-C image circle (much smaller than a 35mm camera’s) nor have to clear a mirror, and the lenses yet are 80-90% as large as SLR versions of the same.

Why do SLR lenses meet our expectation bias? In one sense, it is fair to complain about the quality of SLR lenses because the end result is not what we want – and measured as a system, they indeed underperform. But in an era where SLR lenses are being adapted for use in other things, it is fair to deconstruct what part of this is fairly attributed to parts of the system we are no longer using, such as the traditional SLR itself. And let’s be clear about this: until the advent of mirrorless cameras, the SLR (or DSLR) was the only way to achieve perfect, parallax-free framing and to reliably focus long telephotos and macro lenses.

— Focusing wide-open, shooting stopped-down. All SLR lenses are focused wide-open, which makes focusing accuracy vulnerable to focus shift. This phenomenon, which comes with spherical aberration and “good bokeh,” means that a lens might be perfectly focused at a wide aperture but back-focused when the aperture stops down for shooting. This same thing afflicts both rangefinders and SLRs, only in rangefinders, it is written off as “focus shift” and in SLRs, it is called “being a poor performer.” Aspherics and floating elements help mitigate this – and both are in play on modern lenses of all types.

— Suboptimal focusing screens. You can’t win with a single screen on an SLR. The original SLR focusing screen, a plain ground glass, excelled at focusing telephoto lenses because as the focal length increased, so did the magnification of the subject that the photographer sought to focus. But this screen was dim in the corners and sometimes dim, period. It also failed with wide-angle lenses, where the details critical to focus were actually reduced. Over time, SLRs developed focusing aids like split-image center reticles (actually tiny rangefinders). They also introduced fresnel surfaces to brighten the corners. These made it simpler to focus lenses 50mm and down, but they degraded the ability to accurately focuses lenses 85mm and longer.

— Small viewfinder magnification. A key constraint of camera viewfinder systems is that eye point and magnification are in direct opposition. In practical terms, this means that to be able to see the whole picture through a reasonably-sized viewfinder, especially while wearing eyeglasses, the picture must be reduced. This degrades the focusing abilities of every SLR focusing screen.

— Taste-making. The problem with publications like Popular Photography (and now sites like DxOMark) is that they focus the user’s attention on tests that bear little or no necessary connection to real life.The old-school photo magazines paid little attention to rangefinder lenses, so the tests of SLR lenses were generally focused on the relative merits at huge enlargement factors, and not surprisingly, among SLR lenses, the results favored more expensive glass (the larger advertising budgets of the major companies is always suspicious as well). This did not affect the sales of SLR lenses in general (because at the time no one really liked rangefinders), but it did lead to a perception that anything other than a name-brand Canon, Nikon, Pentax, Minolta, or Konica was garbage. This was an inaccurate and unfortunate perception for three reasons: (1)  Cosina, Tokina, and Sigma were making some of the major brands’ lenses under contract; (2) some of the aftermarket lenses performed adequately for the purpose; and (3) the blanket perceptions about these products, particularly third-party lenses, has landed literally millions of completely usable (if not in some cases very good) lenses in landfills.

— Leica people. Yes, we said it. For all of the doctors, economists, attorneys, CPAs, and engineers who own these and similar rangefinder cameras, there is a widespread misperception that MTF figures for SLR lenses – like home run statistics for Japanese baseball – need some kind of implicit adjustment downward to be comparable to MTF for rangefinder lenses. Not so. MTF is MTF, and it is measured in standardized procedures that do neither the camera body nor care about the lens design itself. It is of some note, conversely, that Leica’s presentation of 5lp/mm (a largely obsolete measure relevant primarily to optical prints) leads to an impression that Leica’s MTF numbers are “higher and flatter” than comparable brands.

Turning the world on its head. Two things changed the picture (so to speak), and quite radically.

— First becoming last. As noted above, he advent of 24x36mm (“full-frame”) digital cameras has exposed just how poorly some traditional rangefinder lenses perform when they project images onto flat sensors. That negative effects are minimized on smaller digital RF and mirrorless platforms (because those corners are effectively cut out of the picture) is immaterial; the only compelling thing about using rangefinder lenses on another camera is killer wides. And frankly, native APS-C lenses – because they are designed correctly for digital sensors – crush adapted rangefinder wides.

— The closed circuit. One of the things that makes mirrorless cameras really, really good is that their autofocus systems can gauge focus from the sensor itself. But this benefit – which bypasses all of the focusing infirmities of SLRs. But the same advantages obtain when attaching manual focus lenses. Not only can the user see the image exactly as resolved by the sensor; he or she can see it at greater magnification or with focus peaking. Getting virtually any lens on to any body never seems to cost more than $30, and there is now plenty of opportunity for exploration on an epic scale.

How do SLR lenses do on digital bodies?  The answer is, “it depends on the lens.” The first place to start is the adapter. It needs to be plane-parallel and to have the correct register. Many adapters are off-kilter and are cut short to “assure” infinity focus. They will need to be shimmed sometimes to achieve correct infinity focus (if you want to scale-focus wide-angle lenses). Once you get past that, this is what you can expect. Over the next few posts, we will explore some favorites, but we will spill the beans on a few “sleeper” lenses here. Caution: be careful with M42 (Pentax Screw Mount) lenses with automatic apertures – you may need to disconnect the stop-down pin to get to shooting aperture.

— Wide angles (<35mm). Because these lenses have a palpable focus point wide-open, an EVF, either at magnification or with focus peaking, is the best way to focus these. Consider also that if you are shooting traditional rangefinder wides and actually focusing them, you have to look first in the camera’s viewfinder/rangefinder window, then switch to an external finder. An EVF kills both birds with one stone (or look). Wide-angle lenses will generally perform best close-up, where errors in infinity register will have the least effect (and you should never be aiming for infinity with shorter than a 35mm lens anyway – since subject details are getting too small to give any impression of sharpness). If your thing is close-up, wide-open shots, the Vivitar 20mm f/3.8 Auto is one of the best and cheapest things going. The header picture for this article is shot with it, wide-open on an M typ 240 (which is way more resolution than any historic 35mm-format lens was ever made to handle). Reasonably low distortion (-5 on Lightroom, if you have any straight lines in the shot), high sharpness (click to get it full-size, then blow it up to check out the eyes, which are the focus point), nice bokeh, and reasonable vignetting. Vivitar lenses should not be ignored; this was a company that often employed its own lens designers in the U.S. and produced many manual focus lenses that were quite good (disregard the autofocus products and recent-vintage manual focus lenses, which can be pretty bad). Did we mention that it often costs less than $60? The Tokina RMS 17mm f/3.5 manual focus lens is also pretty good, though it often shows up a bit overpriced. Adapted wide-angles are not as compelling on APS-C cameras – because they become slowish, semi-wide lenses with huge form factors.

— Normal lenses (50mm-60mm). This is the place where there is not much point to adapting lenses – except on APS-C cameras, where these behave like fast-ish short telephotos. The lens that came with your camera is going to outperform an adapted lens – and focus both faster and more accurately. Plus you already own it. One exception is in super-speed (f≥1.2) normal lenses, which become the equivalent of a 75/1.2 on an APS-C camera or remain an awesome 50/1.2 on your Leica M or Sony A7. If Leica users need EVF to accurately focus the $10K 50/1 Noctilux, you shouldn’t feel bad about using one to focus your 1970s Nikkor. The nice thing about 50/1.2 lenses and 57/1.2 lenses is that they were every SLR manufacturer’s showpiece lens; the optics are almost always great. The other use case for adapted normals is for lenses with “character,” such as Tessars and Sonnars. The Soviet Industar 50-2 (50mm f/3.5) and -61 (f/2.0) (both 50mm Tessar, M42 SLR mount) fit this bill.

— Telephoto lenses (≥75mm). Assuming that you can get a high enough shutter speed to use these (you generally want the reciprocal of 2x the focal length or faster), this is where things get fun. SLR telephotos are often a stop or two faster than rangefinder telephotos, and they often have slightly lower contrast wide-open (which was never historically a problem, since for most of history people used these lenses to shoot high-contrast, low-light pictures). Focusing is less challenging due to the higher magnification, and with many of these, focus peaking suffices (magnification would be absurd). From a quality perspective, even cheap telephotos work really well. Here, we would jokingly tell you to “go big or go home.” A worthwhile lens to try is the Konica Hexanon AR 135mm f/3.2. This is the best of Konica’s SLR 135s, it is the cheapest ($50 on Ebay), and it focuses down to a meter. Make sure it’s the 3.2 and not the 3.5 or 2.5. The Soviet Helios-40-2 (85/1.5) is a cult favorite, but there is no argument that it is cheap at $300-400 these days. It was fun for a C-note, but those days are over. The Soviet  Jupiter-9 (85/2) (Sonnar, M42 SLR mount) is also a solid portrait choice.

— Zoom lenses. There are only three true “zoom” lenses for digital rangefinders: the 16-18-21mm Tri-Elmar, the 21-35mm M-Hexanon Dual, and the 28-35-50 Tri-Elmar. The first two are expensive ($>2000), and the third is kind of ho-hum. And none of them is a true zoom; they are all lenses that have two or three discrete focal lengths. This is an area where the things that are most fun are not intuitive. Wide-angle zooms can be unwieldy when adapted to digital cameras; telephoto zooms can be somewhat challenging to control (but have some merits). The midrange zoom is where your sleepers lie, and if you are a heavy EVF user, a good, compact 35-105mm is not a bad thing to have around. One to check out is the AF 35-105 f/3.5-4.5D Nikkor ($100-150 used). This is a tiny, aspherical, internal-focusing push-pull zoom. It is quite sharp and contrasty, and if you ever get back to your Nikon DSLRs, it is quite a nice lens. It was not a cheap lens when it came out, but selling at around $100 today, it’s one to consider.

— Novelties. Many fun (and very occasional functional) accessories were made for SLRs – cheap fisheye lenses, 90 degree attachments, telescope adapters, and the like. For occasional use, these can be economical and entertaining. Fisheyes in particular are something that are, for most people, not worth investing in. Many of these lenses want 24×36 sensors to reach their full, ahem, potential.

Conclusion. It’s probably not good to counter one generalization (that old SLR lenses are no good) with another (that they are all good). For people who occasionally need a focal length, frequently use EVFs to focus heavy fast lens or telephoto users, or are already zone-focusing wide lenses, older SLR lenses are an avenue that might be helpful. Not every SLR lens is a great performer at a small pixel pitch, but there is value in seeing what can be done more simply and cheaply than forking over another several hundred (or several thousand) to buy a native RF or mirrorless lens that comes out of the bag once or twice a year.

Advertisements

Twilight of the viewing filter

tiffen

Among many other things that are fading away with film is the viewing filter. The Kodak Wratten #90 has long been the standard, though as a discontinued item, it is getting rare and expensive. The Zone VI mounted filter is long gone. If you get moving, you can still pick up the Tiffen Viewing Filter #1 ($40), which is a Wratten #90 laminated in glass and mounted in a phenomenally nice metal holder made in the U.S.A. (you cannot say so much for the velcro pouch). It is also only marginally more expensive than an unmounted #90.

If you read the casual descriptions, a “viewing filter” is something that “converts scenes to black and white.” That’s not exactly true; such a filter uses a dark color so overwhelming that your eye cannot easily discriminate the colors in a scene. The #1 filter, designed for black and white photography, is a very dark brown. It purportedly shows you a “normal” film response, which is something arbitrary (the look of a film really depends on your film and developer).Viewing filters come in other varieties and filter colors: they are (or were) also made for low- and high-speed cinema films and chroma key work.

But at a minimum, the device does show you where certain dark tones get muddy and where the highlights are. This in itself makes such a filter worthwhile – at least as a warning device. You can stick your black-and-white contrast filters in front of it (for example, a green filter to correct incandescent light), but it only works to a point – objects of complementary colors do indeed darken, but your eye quickly adjusts to acquire whatever color information it can, however weak.

As to the ready-made unit vs. unmounted gel issue, you might want the unmounted gel if your goal is to implant this filter into an existing accessory viewfinder. A Wratten gel is optically insignificant in terms of distortion, and because it is moisture-sensitive, it benefits from being inside a viewfinder unit (rather than the outside). A ready-made unit will be more durable and resistant to abuse, though it is just another thing to haul around (though you could attach it to the strap for your light meter).

Are alternatives available? Of course. You could go through a $2 Roscolux swatch book until you found something with a similar effect (though it might be a different color). Or you could find a set of old-school, bottle-brown sunglasses – that though not quite as dark as a #90, are quite helpful for visualizing black and white. And if you want to be truly perverse, you could set your iPhone to its black-and-white filter and use that as a visualizer.

 # # # # #

Zeiss C Biogon T* 4,5/21 ZM – and removing the reds

2145

This lens is perfectly usable on the M240. It doesn’t even take that much work.

The Leica M typ 240 presents some unpleasant choices in terms of 21mm lenses: you can spend $3,000 on a Super Elmar 21mm 3.4 and get the sharpest 21mm ever made for Leica – but suffer complex distortion and red edges. The 21-35mm M-Hexanon Dual (which is not a lot cheaper these days than a used Super Elmar) gives you two focal lengths, awesome sharpness and no color shifts – but it gives you a touch of geometric distortion. Everything else presents varying combinations of bulk, color vignetting, low resolution, and general misery. Here at the Machine Planet, we have a certain inbuilt arrogance to try things that conventional wisdom says should not work. The 21mm f/4.5 Biogon is a case in point. And yes, we made it work with a couple of off-the-shelf tools and less than a couple of hours of trial and error learning the ropes.

The good. If this were the film era, the 21mm f/4.5 would be the champ. It is small (barely bigger than a 40mm M-Rokkor), sharp (testing in some reviews to 3000 lines per picture height), well-made, and has about as close to zero distortion as any wideangle lens ever made (for example, it’s lower than the 35/1.4 Summilux ASPH). It also takes normal-depth 46mm filters common to the rest of your lens collection. Here is basically everything you need to know about its stunning performance:

2145

The bad. In terms of conventional performance, the lens is relatively slow in terms of maximum aperture and has the usual light falloff from the center, often exaggerated by digital sensors. You can see from the chart above that it does not get much better as you stop down.

The ugly. The worst thing is that the lens has color shift at the edges. It’s quite severe at first glance. These are the particulars:

  • The red edge extends a couple of MM into the frame, from top to bottom, green and red on the left and red on the right. In the days of the Kodak DCS Pro 14n, this was called the “Italian Flag” effect.
  • The intensity and intrusion of the edges is dependent on selected lens aperture and focused distance. Closer focus and wider apertures mean that the edges are far less obtrusive.
  • There is an overlay of standard brightness vignetting that is characteristic of any symmetrical 21mm lens.

The variable nature of the color shading – why has no one else noticed this? – may well be the cause of claims that the problem “can’t” be corrected or that conventional tools result in under- or over-correction. Once you understand this, it’s easy to solve the problem. Never declare defeat prematurely!

Fixing things up. All solutions to this problem involve some kind of reference image, which is a test shot you make using a white field. You can shoot a white wall, shoot a ceiling with a flash, or use a diffuser. If you shoot through a diffuser, you need one that lacks texture (at small apertures, the ZM 21 can pick up the texture of the paper, even if you have it pressed right up against the lens. Your resulting references will look roughly like this:

L1005380

One very good diffusion material is Yupo polypropylene watercolor “paper,” which, being plastic, has no grain. You can find this in most art stores.

  1. Layer Masks. Some, like Lloyd Chambers, advocate the use of Photoshop adjustment layers and masks to cancel out color and brightness shading. Although this demonstrably works, its shortcoming is that it needs a separate template and action for every permutation of shading (you can, most of the time, get away with four settings: f/4.5 and f/8, at 1m and ∞). It also presents a clumsy workflow that involves leaving Lightroom, going to Photoshop, and then back to Lightroom (and at that point, with a TIFF and not a DMG). For your most OCD applications, this is a workable solution; it’s just not the most batch-friendly or space-efficient solution.
  2. Cornerfix. Long the go-to solution for Leica M8 and M9 users, Cornerfix was originally designed to address the green shift that occurred when you put a UV/IR filter on an M8. This green shift was generally uniform and radial. Cornerfix takes the reference imageand then computes a mathematical mask. Cornerfix works with DNGs and exports DNGs (suffixed “_cf”)Unform, and it has a tremendous range of settings for addressing color shift, brightness vignetting, and the artifacts that result from correction. Cornerfix also shows you the effect of the selected mask on the current image. It also supports batch processing. The shortcoming of Cornerfix, though, is that because it does correction via equation, there are some kinds of color shading that it struggles with.
  3. Adobe Flat Field plug-in. The strangely named Flat Field plug-in is available on the Adobe Labs site. This plugin has virtually no controls and seems to be an automated variant of the layer mask technique. You select the image you want to correct, activate the plug-in, and then give it the reference image. The only controls are for “Color” or “Color and falloff,” which lets you leave in brightness vignetting if you want. The plugin is slow and kicks out another DNG, stacked with the first one, suffixed “_ff.” It does work very well – much better with the 21mm than Cornerfix – and it does not require you to exit Lightroom, but it’s a black-box solution that requires you to select your reference image carefully (because you can over- or under-correct by choosing the wrong one).

The winner: Flat Field. As the only solution that (a) works and (b) does not require shifting from program to program, Adobe’s free Flat Field plug-in for Lightroom is the best solution. Here is precisely how to use it:

  1. Shoot your profiles. Take your sheet of Yupo paper, hold it right in front of the lens (the easiest way is to sandwich the paper between your lens and the glass of a window). Pick your reference distances. We used 1m, 2m, 5m and ∞, but you could also pick your favorite hyperfocal distance. Shoot a test at one f/stop, all distances. Then switch to the next f/stop, all distances.
  2. When you are done, import the files into Lightroom. Immediately rename these with a designator that shows lens, aperture, distance. This will result in a name like “2145-80-inf” for 21/4.5, f/8, at infinity. Export all of these as original DNGs to a folder that is easy to find (think about “profiles” in your “Documents” folder.
  3. Install the Flat Field plugin.
  4. When you want to do a correction, select the picture(s) you want to fix. All of the ones you do together should have the same shooting aperture and distance (the M240 records a computed aperture value, and you should be able to tell by the composition where the lens was focused).
  5. Go to File–>Plug-in Extras–>DNG Flat Field–>Apply External Correction. This will pop up a Finder or Explorer window to select the profile from #2 (Lightroom does not let you choose from the catalog).
  6. Choose “color and falloff.” Although vignetting may seem cool in theory, symmetrical lenses need all the help they can get.
  7. Run it.
  8. You will then get a new file adjacent to the original with the “_ff” suffix. You can now manipulate this as if it were the original.
  9. If you get too much correction, try a reference photo shot at a closer distance. If you get under-correction, go for a farther distance.

Upshot. It is tragic that so many people started unloading these lenses based on a red-shift issue that is so simple to correct with modern tools. The ZM 21/4.5 is a fantastic optic that can now make the jump to modern digital Ms. And there is no reason why the same techniques could not be used to adapt other wideangle lenses to Ms or wideangle M lenses to things like the Sony A7 series.

L1005374

L1005374_ff

L1005358 L1005358_ff L1005340 L1005340_ff

Fujifilm Fujinon XF 14mm f/2.8 R

DSCF5157-Edit

Beauty and the beast. The Fujinon 14mm f/2.8 encapsulates everything that is good and bad about the Fuji X-Pro1 system. The lens is a solidly constructed, masterful piece of optics, so well corrected that it doesn’t rely on computing power to eliminate distortion. It maxes at f/4 to f/5.6, the range where most rangefinder (and most SLR prime) lenses are wide open and challenged. The mechanical design is mostly elegant. Unlike with a Leica, where you bolt on a distorted accessory finder that may have a level in one orientation only – leaving you to DxO Optics Viewpoint for verticals – the Fuji lets you watch the action and align it on a gridded electronic viewfinder (EVF) (you can, of course, see a gridded distorted wide-angle optical viewfinder (OVF), too…).

But then on the minus side you’re using an X-Pro1, which although fast on its feet occasionally puts some of its buttons too close to your fingers (ahem, Q button, I’m talking to you), very occasionally stutters with AF, and might take a second to start up with OVF on. The optical resolution of the lens and the microscopic details it can resolve mercilessly expose the sorry state of Adobe Camera Raw conversion of X-Trans RAF files. The red smearing is gone, but when you shoot white against a blue sky, inducing any amount of overexposure, or cranking up the exposure in post, or dropping a color channel too much can give you bleed in a b/w conversion. Be careful with bright sunlit scenes. Careful work will be rewarded.

The nice thing, though, is that when/if a new X-Pro comes out with a higher-res sensor, X-Trans or not, this lens will hold its own.

The wide view. Wide-angle lenses are strange creatures. The first thing to remember is that truly wide lenses didn’t  come into their own until after World War II. Until then, most cameras used focal lengths that at the widest were the equivalent of a 35mm lens on a 24x36mm camera.  The postwar period and lens coating enabled a number of complicated new designs such as the Zeiss Biogon and the Schneider Super-Angulon, each of which used a staggering number of glass elements in a symmetrical pattern. The result was an undistorted picture, though the lenses required very small clearances between the rearmost glass element and the film (making them unsuitable for most SLRs).

And even though these existed, they were niche products – most people took Robert Capa’s advice that if one did not get the picture he wanted, he was not close enough – and superwides exacerbated the problem. So until about the 1990s, a 28mm to 35mm lens was the “wide” limit for most people. It is not unwarranted; 35mm in particular represents the central field of human vision with both eyes, does not mangle human faces, even at the edges of the rame, and does not create dramatic converging parallels when the camera is tilted. Many still consider it the ideal “only lens.”

By the mid 1990s, manufacturers had figured out how to make good wide-angle lenses that were usable on SLRs. Although they were retrofocus in design, modern glass, coatings, and (most importantly) computers made it possible to make highly corrected designs. Zoom lenses, which long had been maligned, became some of the best performers. For example, the 17-35mm f/2.8D AF-s Nikkor produced sharper and less distorted wide-angle images than almost any prime Nikkor lens that preceded it in its focal-length range.

When rangefinder cameras came back into fashion, lens manufacturers found that it was relatively easy to use modern technology to design great superwide lenses. Cosina and Kobalux fielded relatively economical and high-performing 21mm lenses. These were followed by Ricoh, which remounted its GR21 lens for rangefinder, and the coup came with Konica’s 21-35mm Dual Hexanon. Not to be outdone in capabilities (or price), Leica launched its 16-18-21mm Wide-Angle Tri-Elmar. Unfortunately evident through all of this, however, is that people do not fully understand that the point of a wide-angle lens is dynamic composition, not simply making the field of view wider and everything in it tinier.

Optical performance. There is plenty of test data out there that substantiates this author’s observation: this beats the stuffing out of even most modern 21mm FOV lenses. Peak performance is at f/4 or f/5.6 – depending on whether you value better corner sharpness at the expense of a little center sharpness. A lens like this might cause you to revise your religious beliefs. Needless to say, it outperforms all of the economical adapted Leica-mount lenses and most Leica M 21mm lenses, too. It’s fairly easy to substantiate that there is no in-camera correction of pincushion distortion. Because this lens doesn’t need it. It is one of the least distorted 21mm-equivalent lenses ever made. That won’t stop you from tilting the camera, though… Vingetting is not as apparent as reviewers would lead you to believe, and it is certainly less than symmetrical wide-angle lenses.

Size/weight. This is not a small lens; it is barely smaller than the 18-55mm zoom. It is still a little bit smaller than an aspherical 21mm Elmarit for a Leica. The weight is very light; these lenses are aluminum and magnesium, not brass.  The lens does not unbalance the camera, and an X-Pro1 with this lens still feels unnaturally light. The lens takes 58mm filters, a size last seen on Canon SLRs. This would not be notable except that Fuji has been all over the place on filter sizes.

The focusing ring. This is a big attractor for a lot of people. So let’s start with the bad news: it’s still focus by wire, and it actually stops past the indicated infinity mark. This is not a big deal. But what is a big deal is that Fuji had to elect a focusing direction. Here, Fuji went with the Nikon direction and abandoned both the Leica and former Fuji MF rotation directions to get to infinity. So although you might set your X-Pro1 or XE-1 to focus like a Leica or your old Fujica, this lens will always be an outlier. But you might not care. This is a “set it and forget it” lens.

The selector. The focusing ring is the main auto/manual focus switch. Here are the modes, and they may not be what you expect:

  • Lens in “manual.”  Regardless of camera mode, the lens focuses via the ring (and by wire). There is no distance indicator in the viewfinder.
  • Lens in “auto.” The focusing ring does not turn at all. With the camera in AF mode (AF-S or AF-C), the lens autofocuses like any other Fujinon XF lens. With the camera in manual, you can use the AE/AF button to focus. Both ways, you get a distance scale in the viewfinder.

Now here’s the catch – if you thought that you were going to set the lens scale for your favorite zone-focus distance and switch AF on and off via the ring, you may lose your distance setting on switching modes. The focusing ring only pushes forward into discrete positions, which put the infinity mark (a) between f/22 and f/16, (b) at f/8, and (c) at f/2.8.  If you are not in one of these positions when you push the ring, it will cam into one of these as it goes forward. There are a number of other positions in the sub-one-meter area, but they are not really relevant.

Some people have complained that the selector is easy to knock out of position. It is actually quite stiff – and the bigger danger is forgetting that the lens is in manual mode (although during testing, I repeatedly and accidentally set the aperture at f/5.6, failed to change the hyperfocal setting from f/8, and everything came out fine).

Distance scale. The distance scale shown by the lens in manual-focus mode is fairly compressed and reflects the very short throw of manual focusing. According to the scale, at f/8, the lens can keep cover 1m to ∞ in focus. According to the viewfinder scale, though, it is 3m to infinity. Actual use suggests that the lens barrel DOF indications are very close to reality. The great depth of field of a 14mm lens puts tremendous pressure on composition – depth of focus will not be a factor in most pictures.

Viewfinder picture. Where the camera is set to the OVF, the low magnification setting will just cover the lens field of view (you will see “corners” rather than framelines). The EVF shows 100% of the field (marginally more than the OVF). Either way, you will want to turn on the gridlines and to activate the digital horizon indicator. There is not much point in using an external viewfinder unless you need a 2-direction level (c.f. Leica Universal Wideangle Finder M) – but even then, these only really work in “landscape” orientation.

AF speed. There are varying accounts of AF speed out there. This lens has the same type of AF motor as the 35/1.4, and it is very fast in most light. If you get into low light and it slows down too much, try switching to EVF and change the size of the focusing zone. The reality is that with this lens, you only really need AF from f/2.8-5.6.

Should I buy it? This is Fuji’s second solid hit after the 18-55mm zoom. The best (if not only) arguments against this lens are (1) that it is an XF-mount lens and (2) it’s expensive. If Fuji never went beyond the X-Pro in terms of sensor technology, it’s not much of a stretch to say that the combination outperforms any 35mm film camera with a 21mm lens (and most digital cameras); moreover, with its easy manual focus, any real or perceived shortcomings of the X-Pro or XE-1 in focus are pretty insignificant. As for the price, it is not very high in comparison to 21mm lenses: in the same range as the ZM primes, a couple hundred higher than the Cosina-Voigtlander 21mm M lens, and at least a thousand less than any modern Leica 21mm.

Should I wait for the 10-20mm f/4 OIS (Optical Image Stabilization) lens? Fuji does have a zoom lens on the roadmap that covers this focal length. It will have a lot of appeal to a lot of people – but not necessarily for the same people who would be looking at a super-performing 14mm. From all available prototype pictures, it will be like the 18-55mm, with a flared front end (it takes 72mm filters).  The plusses are that it will likely have a stepper motor focusing mechanism (=fast), has a constant aperture (albeit f/4)  and will cover the 15-30mm range. On the other hand, it is huge, has no focusing scale like the 14mm, will require software distortion correction, and if the 18-55mm is a guide, manual aperture control will be easy to knock out of the desired setting. OIS is a plus – but the Fuji system draws a good amount of power, it takes a second or so to spin up in the mode where it is only activated for the shot, and can be squeaky in the wintertime.

How about the Zeiss 12mm? This is a tougher question that might be determined more by economics (it’s pricey) and aesthetics (it’s flared in a not-so-attractive way). I will be the one to speculate on this: these lenses are being made with the cooperation of Fuji, possibly in the same plant that makes Fuji lenses. That’s why they keep getting mentioned in Fuji videos and are now shown as an ancillary part of the Fuji lens road map.

# # # # #

20130417_152543

Fujifilm Fujinon XF 18-55mm f/2.8-4 R LM OIS

1855If nothing else, the Fuji X series exhibits good looks. The 18-55mm is no exception. For what Fuji charges for this standalone lens ($699 – or, if you are like this writer, $249 plus a zillion American Express points), the company knows not to disappoint buyers with entry-level plastics.  The 18-55mm is a substantial lens, heavy and (at least to the touch) well-built. The zoom is extremely well damped, the switchgear is good, and it even has a nice ED-style trim ring. The lens extends when zoomed but not when focused.

Be aware that this is a fairly big and heavy lens; it is fatter than an original Tri-Elmar, and it makes the camera more nose-heavy (and not so featherweight). It is not pocketable by any stretch. The lens barrel will partially block the AF assist light, and in the “wide” OVF mode, it impinges quite a bit into your field of view.

Let’s get to the big question first. Yes, the image quality is all that, wide open, at pretty much every focal length and all reasonable distances.

Actually, let’s back up a little bit. Why are these marked “Fujifilm,” when “Fujica” is already in the inventory and more media neutral?

The next big question, and the one that gets too little play, is how fast this lens focuses (Fuji claim: 0.1s for locking, though under some unspecified set of conditions).  It’s easy to ascertain that this corresponds to outdoors daylight, which is, realistically, EV10 and up. And yes, it’s crazy fast.  Indoors, in artificial light, it slows down a little bit, but it is still a lot faster than the 35/1.4 in most circumstances.  The 18mm setting is blindingly fast (in no small part because the lens lets in the most light at that setting).  23, 35, and 55mm slow down progressively (and, as can be expected, the closer the range, the slower the focusing).  Toward the 55mm end, you occasionally see some unfamiliar behaviors (if you are used to the 35).  One is that hunting is inaudible, and its only manifestation is that for a second, it looks like nothing is happening when you press the shutter. Then, without warning, the framelines (OVF) shrink, the light goes green, and the camera fires.  In EVF mode, you see the lens run through its distance range and then locks and fires. The interesting thing in EVF mode is that the frame momentarily seems to pixelate, which might be indicating that the camera looks at a downsampled data set to see when contrast is maximized.  The other interesting behavior is the red AF warning.  A few notes:

  • The red AF warning comes up more often in EVF mode, and after some experimentation, it seems that it happens most often in low light, where the AF spot size is set to the smallest setting. If you press AF and then increase the size of the spot with the thumbwheel, it mitigates the problem. This fix is not suprising; increasing the sample set helps the camera find some contrast to compare.
  • In EVF mode, the X-Pro1 will let you fire an out-of-focus shot with no warning. If the box does not turn green, don’t count on getting a picture that is in focus.
  • The “mash-n-go” technique still works in OVF mode. This is the one where you press the shutter release until the camera fires (and it only does so when things are in focus).

Manual focus runs through the entire useful range of distances in a little over a quarter turn at all focal lengths. You won’t use it much, and even if you do, you will probably use the AE/AF button to do a spot focus.

Note that the focusing is dead silent – making your AF-s or USM lens sound like an agricultural implement. What is not silent is the aperture actuator, which makes a very subtle click when focus completes and the lens stops down.

As to controls, the internet seems a little confused about how aperture setting works. The unmarked ring just to the rear of the focal-length indicator is the aperture ring (aperture is not manually controlled from the body, as some people seem to think). Unlike the primes, which have numbers and a physical stop at each end, the zoom has a ring that turns all the way around. Turn it to 2.8 and keep turning – it still stays on 2.8. The A switch activates auto-aperture (like the red A on the prime lens rings). To tell the truth, this is a lot better than the prime lenses, which inexplicably omit locks on the “A” setting – making it easier to accidentally switch the control from A to 16 (with attendant blur). The aperture automatically compensates as the focal length changes if the ring is set to the widest opening – but if you set the aperture to f/4 or smaller, it does not change with focal length changes.

The viewfinder picture answers the question of why this is normally packaged with the XE-1.  On an X-Pro1, in OVF mode, the framelines shrink or grow (continuously) depending on the selected focal length. If you turn on corrected AE targets you can also see part of the challenge of using the OVF: the distance between the nominal and near-range boxes changes dramatically between 18 and 55mm. Though this condition exists with prime lenses, the zoom introduces a situation where you have to be able to internalize intermediate corrections at many more focal lengths.  In addition, you really have to decide whether you want to shoot wide or long. If the X-Pro viewfinder is in “wide” mode, the 55mm frameline seems impossibly small. If it is in “normal” mode, the framelines become bigger than the viewfinder around 30mm. It would actually be nice to have the camera automatically switch magnifications, but that does not look like it is part of Firmware 2.01.  It will be very interesting to see what happens with the 10-20mm OIS lens that is slated for next year.  EVF mode is easier to manage, with a stable focus point, and it is here that you get a little of the Optical Image Stabilization (OIS)  seasickness: the camera moves but the viewfinder picture moves less.

And on to OIS – this has three modes: off (via the lens), “shooting only” mode (via setup menu), and “continous” (also via the setup menu). The latter two appear to be a choice between having the accelerometers come on instantly or having them get up to speed while frame-finding. Cutting down on the OIS on time probably has a lot to do with conserving power.  The OIS will allow you to shoot below 1/10 sec @55mm and get passable results, which probably comes close enough to the “four stops” claimed by Fuji. It is very important to note that VR or OIS or however you want to trademark it is only effective at compensating for camera movement. It is completely useless for arresting subject movement in low light – and depending on  the interaction could conceivably make it worse (i.e., lens compensates in a direction opposite subject movement direction). So if you are trying to track fast-moving children in low light, well, get a Nikon D700 and a fast prime instead (if for no other reason, the AF is better).

Flash operation with the 18-55mm is fairly predictable (at least with the EF-42, which zooms appropriately with the lens). But do note that as you use more telephoto-like focal lengths, redeye becomes a bigger problem if you use direct flash.

All of this aside, though, the 18-55mm radically increases the functionality of the X-Pro1 and in yet another way makes it more fun than its Leica inspiration. The Leica world has no continuously variable “zooms” – let alone any that gets to the 82mm equivalent of the 18-55mm. And nothing in the M world gets as wide as f/2.8 (the 21-35mm Dual Hexanon does f/3.4, but that is mostly just for kicks).  The 18-55mm is a nicely done lens that should work great for travel, and it is certainly more enjoyable than the Fuji primes for casual snaps. It tends to make the reportage dichotomy a choice  between the X-Pro1 on the one hand and the $7,000 M240 on the other.

We will work on some nice-looking sample pictures over the next month when the weather transitions into winter (rather than grey, dull late autumn…). In the meantime, you might want to order one of these.

# # # # #