dante stella stories photographs technical guestbook

Smithers: "They're fighting like Iran and Iraq!"

Mr. Burns: "What?"

Smithers: "Persia and Mesopotamia."

Leica M: that distinct feeling of ennui

April 16, 2012. All over the world, there are provincial towns believed by their residents to be equal to New York City, Tokyo, or Paris. In a way, Leica M might be such a town. Leica (the company) is not so myopic in terms of technology, but for whatever reason, digital M arguably has become both a technological and a cultural backwater. As Carlo Levi would have put it, Cristo si fermato a Solms.

The duality of Leica

As this author has observed it over a 15-year period, M culture is basically drawn from two groups (a) people who put up with Leica's quirks and price due to a belief (often justified) that the resulting image quality is better and (b) a group of photographers cool to modern technology and suspicious of the idea that in spending tens of thousands of dollars on a system, someone might want features that make an M look more like a practical "only" camera. We can call the first group the Opportunists and the second group the True Believers.

Surmising what you can surmise about them, the Opportunists are fairly mobile between camera systems. In fact, given Leica's cyclical appeal, this group largely abandoned Leica's system in the early 1970s and abandoned it again when Leica was dragging on a digital body in the early 2000s (recall how lens prices fell back then). Despite claims that demand for M9 cameras and lenses outstrips the ability to produce them, production is small - and even so, the market price for used M9s has now drifted to 60% of new prices. Even new cameras are being discounted by designation as "demos" (no camera that was really used as a demo has five or fewer exposures on it). We know from this that there are definitely fair-weather fans and that they are starting to head for the doors again. When things change precipitously, we know the Opportunists are on the march. And some Opportunists march by keeping their M8s and simply supplementing their missing capabilities with D700s and X100s. The effect, however, is the same - that they stop buying brand L and begin experiencing the forbidden fruits of other manufacturers.

The True Believers - a smaller group but more influential with Leica's management - hold that the world stopped producing useful new camera features in about 1986 (or, alternatively, in 2002 with the M7). For that reason, they believe, Leica M must be locked into a world of vestigial and functionally-useless removable baseplates, frameline preview levers, and ergonomics lazily whittled from a bar of Ivory soap. The True Believers deny that any feature a Leica M currently lacks is significant, desirable, or valid. Their faith is strong despite the fact that Leica itself has proven them wrong by introducing the very things True Believers claimed were nonessential to the M system: film backs that opened, lever winding, combined rangefinder/viewfinders, TTL metering, electronic shutters/autoexposure, TTL flash, and ultimately, digial imaging. For this group, the M9 - which emerged years behind technologically - is "enough." In fact, it is already too much (one dares not speak of the D-Lux, the Digilux, the S1, the S2, or the DMR - all of which were actually cutting edge when released).

When things run their commercial course, we can call them effectively obsolescent. Obviously, nothing actually stops taking pictures (or anything else) when it is superseded by newer, flashier products - or even products with better specs. But new products often do the same thing with more speed, better efficiency, or fewer avoidable annoyances. The world is littered with well-built, well-designed items that should have lasted forever in the market but were passed up by things that were simpler, cheaper, or more appealing to the masses. Fountain pens, for example, a durable, perfected designs that are largely ignored for cheaper, less messy Bic Biro ballpoints.

When it comes to cameras (or anything), this author would take it a step further and point out that that is not fair to judge an older product for lacking features that had not been invented when it came out (and this is being charitable where Leica did not, for whatever reason, implement technology that was available at the time). So talking about digital M, let's leave aside things like live view and video. Let's even forget about DSP speed, screen density, and frame rate. But it is fair to compare apples to apples: to take the core features (or selling points) of an old product and examine their uniqueness in the marketplace and whether they are necessary or desirable solutions to problems.

The five points of Leicas

Leica cameras have five big selling points: high sensor quality; high lens quality, a great synergism between the two, a superb optical viewfinder and a superlative mechanical rangefinder. Take them in turn:

1. Sensor image quality. Image quality is really the reason why serious photographers buy Leicas. The Leica magic (at least at the body level) comes from two things: (1) lack of an antialiasing filter, which gives a perception of an additional 25% in resolving power (or the ability to up-res by a like amount) and (2) image processing algorithms that build a unique look. These huge determinants of quality do not depend on the overall build quality of the camera body; they reside entirely in a CCD sensor and a couple of hundreds dollars in electronic parts. As long as the same glass formula were put in front of this sensor, the end result would be identical, even if the body were ABS plastic and even if the lenses operated by autofocus.

A lot of things have happened in the 7 years since the basic digital M technology arrived. First, other manufacturers have caught up to the filter-free sensor (Kodak actually preceded Leica with many models in which the AA filter was absent or removable). Sigma has the Foveon sensor, which omits it. Fuji has the X-Trans CMOS sensor - which in addition to lacking an AA sensor, has a randomized color pattern that obviates the anti-moire processing that bogs down Leica's cameras. Nikon put a weak filter in the D3 and D700, and the D800E effectively has none (as well as twice the pixel density and much better low-light performance than current Leica sensors). Ricoh is making GXR modules that take Leica lenses and have no AA filters. And the Leica "look," while challenging to replicate, can indeed be achieved in relatively cheap software like Lightroom.

But backing up a little, the world has also moved away from CCD in favor of CMOS chips for lower power consumption, higher sensitivity, and live view capability. Sticking with CCDs constrains Leica's sensor choices for any future digital M (unless Leica changes the imager completely) and puts Leicas at a long-running disadvantage in higher ISO performance. CCD chips do have great color, but so do a lot of CMOS chips. In the end analysis, slipping behind the sensor speed curve is a big issue; the number of megapixels, not so much.

2. Lens image quality. Leica was an early participant in the Lens Speed Wars that started in the 1920s and 1930s. Back then, you needed superspeed lenses because film was rated at a blistering ISO 12. And let's be clear here: from about the 1960s onward, Leica was pretty much unchallenged in terms of lenses, in the build, quality control and resolving departments (and in many ways still is). But a few funny things happened on the way to the 21st century.

  • When the world went digital and addressed low-light situations by upping sensor capabilities, Leica instead focused on simply making faster lenses. Although this technically gets to a correct exposure in a lot of situations without upping sensor performance, it also locks users into what could be called the "Noctilux Aesthetic," shorthand for pictures where there is a razor-thin plane of focus and often heavy shading of the corners. Some people prefer to do things with higher ISO sensitivity (rather than wider apertures) so as to have more things in focus. And if it's the aesthetic that appeals, there is always Instagram.
  • Leica's drive to make faster lenses made lenses for a compact system heavy, large, and insanely expensive (a 24mm f/1.4 Summilux, for example, costs $7,000). An M9-P and a 24/1.4 will run you approximately $14,000. A D800E with a 24/1.4 Nikkor runs about $5,500 - and can either shoot in a quarter of the light with the same noise or the same light with four times the depth of field. Sometimes it is nice to have the luxury of choosing the method of taking low-light pictures. Although the expense is typically met with the refrain of, "it's expensive because it's good," or "it's not for everyone," it is worth pointing out that many of the nouveau riche who buy things like Leicas did not get there by spending money just to spend money- cost/utility analyses go on all the time (albeit among much more expensive products). In units produced annually, Leica M9 production is about equal to the M6 - though the number of eligible buyers in the world has increased radically. Leica's sales are up in China, but with flat overall volume, that means that they are diminishing in other parts of the world.
  • Leica M lenses have very limited options for addressing focus shift. All lenses exhibit focus shift when stopped down, and this can make rangefinder focusing more inaccurate than it should be. Digital has less tolerance for error, and the only ways to mitigate focus shift in fast lenses is to use floating elements and aspherics, both of which - when executed to Leica standards - cost a mint. Closed-loop focusing (in the guise of contrast-detection AF) allows things like the $600 35/1.4 Fuji X lens (for the X-Pro1) to perform comparably to the $3,500 35 Summilux ASPH. But even before that, the lowly Hexar AF was able to keep up with the legendary 35mm Summicron ASPH by adjusting its focus to account for the selected aperture.
  • Leica's 20th-century lenses hold the digital M system back. Users often fixate on speed, but older, high-speed lenses are not world beaters (though many people pay those types of prices for them). The 75mm Summilux command prices that are more driven by rarity than its relatively humdrum performance on a flat sensor (or the somewhat provincial appeal of shooting a portrait with just the eyelashes in focus). Even some of Leica's more innovative designs like the 28-50-35 Tri-Elmar are fairly unremarkable performers on a Leica digital. The standards required to make a good digital lens are far more exacting than what made superlative film lenses in the past. There are always third-party lenses, but sometimes it seems silly to attach a $300 lens to a $7,000 body.

None of these are show-stoppers, but they tend to paint Leica M into the corner of being a very specific solution to any given problem. And getting to the place where a Leica M optically outperforms the competition requires very expensive gear.

3. The synergy. One thing about Leica M was that for a long time, you had to use a Leica body to get the Leica M lenses. This was due in part to patents on the lens mount. Even where other manufacturers made M-mount cameras (like the Minolta CLE, Hexar RF, Bessa R, Zeiss-Ikons, and Rolleis), Leica always had a little bit of an edge due to its huge and wide pressure plate. Today, though, the entire synergistic advantage of using a Leica lens with a Leica body lies in the microlens pattern on the Leica sensor glass. It is not a perfect solution, but it is currently the only way to get the Leica resolution all across the board - and on a 24x36mm sensor. All of that said, the synergy between Leica lenses and bodies really only matters if you assume a Leica M lens to be an essential part of the equation. Where other cameras are built as a cohesive unit (lens and sensor), the 80/20 rule kicks in (80% of the performance at 20% of the price). Only here, Leica's pricing now pushes that toward a 90/10 proposition.

4. The optical viewfinder. One of the big points of excitement about the Leica M is its big, clear viewfinder. Though Leica fields the brightest and least-distorted finders in the industry, those finders are expensive to produce and, given the mechanical nature of the framelines, are incapable of showing accurate framing except at one arbitrary distance. This tends to make shots frame looser than they should be, thereby wasting real estate on the sensor. Japanese manufacturers have not surpassed the Leica clarity, but they have managed to produce close equivalents for much less money. But the bigger issue came with the rise of hybrid viewfinders that use LED overlay displays to (a) show instant playback; (b) project a digital level and composition gridlines; (c) display a computation of the depth of field based on focal length, aperture, and focused distance; and (d) show field-corrected framelines appropriate to any focal length. This is to say nothing of allowing an instant TTL lens view as well. These features - which can universally be shut off - add a considerable amount of utility for people who want them. They don't take away from the beauty of the Leica version, but one line of 8-segment LEDs provides no warning about running through an SD card or a battery, two conditions that did not really exist when the viewfinder was last redesigned, 10 years ago. In the end, the major compelling feature of that Leica view is...

5. ...The rangefinder - part of what makes the Leica M is the rangefinder. Leica Ms will always have rangefinders, because the "M" actually stands for Messsucher (rangefinder). When the Leica II was developed, there were no small SLRs. Leica and Zeiss based their competing 35mm cameras on coupled prism rangefinders. This was, at the time, the only technology that allowed a compact camera to focus accurately, particularly with high-speed lenses.

Even when 35mm SLRs came into the mainstream in the 1950s and 1960s, rangefinders persisted. Rangefinders were smaller in general, and it was easier to make wide-angle lenses for them. Back then - and now - rangefinders also did a better job of focusing those wide-angle lenses. Where a rangefinder system has a constant magnification and starts running into problems with longer lenses, SLRs benefit from assuming the magnification of telephoto lenses they use.

Many competitors have made runs at matching the Leica rangefinder, and the common vendor to Fuji, Mamiya, and Konica almost managed to do it. The Leica mechanism is a wonder of precision and high-end manufacturing. Today, though, it seems like a precisely engineered, laser-engraved, CNC machined, hand-honed ... typewriter. The rangefinder's competence is in focusing wideangle-to-normal lenses - but run-of-the-mill autofocus is just as good at doing that.

The weight

Aside from struggles with relevance to Opportunists at a core technology level - i.e., creeping effective obsolescence - Leica M carries a lot of baggage. The weight (all apologies to Rick Danko and Robbie Robertson) goes beyond simulating the size and weight of a camera of 1953 (the weight is, in fact, simulated - the brass covers of a digital M account for almost 25% of ite weight). It goes beyond doing things they way they have always been done - in the name of tradition. It goes beyond being accosted in public by weirdos who recognize your M8 as "an M4." To this author, the most perfidious part of it is the cognitive dissonance that arises when one carries $10,000 in gear around his neck but fancies himself to be a photographic Zen Buddhist.

Leica used to think outside the box - not only did it popularize 35mm film photography, it also invented things like phase-detect AF, made innovative cameras like the M3, and otherwise kept up with the world (even Leica's current S2 is technological light-years ahead of the M). Had this progressive philosophy carried over into the M series (or an updated successor), the M8/M9 would not have slavishly copied film cameras in looks, live view would have been added to stand in for the Visoflex, and it would have been Leica to introduce hybrid viewfinders. Maybe this will change on May 10, 2012 with some huge product announcement.

But in our hearts, we know it won't. The world of Leica is somewhat frustrating. The products are high quality, the resulting images are excellent, and the general solidity of the system makes all of us keep our lenses as we repetitively upgrade digital bodies (and upgraded film bodies before that). We always want to think that some vastly improved new M is around the corner, yet ultimately, we just end up settling for something that is behind the curve, for a lot of money. One could get the sense - reinforced by the rapid pace of the rest of the photographic world - that this bubble of IR filters, color vingetting, bottom-plate loading, and black paint is going to burst.When you look at things like the Fuji X-Pro1, you begin to think that perhaps it already has. Maybe the better thing would be for Leica to declare victory in 2013 after 60 years of M - after all, it outlasted Contax, Alpa, and everyone elese's film rangefinders (and even outlasted Polaroid, Kodak, Agfa, and Ilford...) - and reboot with something as earthshaking as the M3 was in 1953.

Disclosure: the author has been a Leica user for the better part of two decades and was an early adopter of the M8.