In the summer of 2009, Kodak announced the end for their legendary Kodachrome slide film. The final batch of Kodachrome 64 carries an expiration date of 11/2010, and major retailers have long since sold out—although a few stray rolls can still be found on eBay, at high prices.

But the crucial point to know is this: There is only one commercial lab left in the world developing Kodachrome, and that is Dwayne’s Photo in Parsons, Kansas. And they will only develop Kodachrome through the end of December 2010. The price to develop and mount a 36-exposure roll (before shipping) is USD $10.

Kodachrome requires a complex and difficult process called “K-14″ —completely different from the E-6 developing used with all other slide films. Other color emulsions include the chemistry to produce color dyes within the film itself; but with K-14, each color layer must have dyes added to it during processing. (This raises the image in slight relief on the emulsion side, which is a useful way to ID unlabeled Kodachromes.) Balancing all the K-14 steps correctly is quite tricky—hence the worldwide consolidation of K-14 processing down to one final lab.

So, this year is your last chance to shoot Kodachrome. Dig around in the back of the freezer; check the bottom of your camera drawer. If you have any rolls remaining, shoot them soon—or give them to someone who will!

One part of the Kodachrome legend is its amazing colorfastness—unmatched by any other film created since. Here’s a Kodachrome “glamour” shot taken by my father in 1942 (scandalously, this is not my mother):

Kodachrome slide after nearly 70 years; scan colors only slightly adjusted

Will our hard disks and flash drives still be accessible in 2076? Kodachrome photos you shoot today will stay colorful and accessible for decades to come—and anyone can discover the images, just by holding them up to the light. If there are people or scenes in your world that you’d like to commemorate for the ages, Kodachrome is your film.

Now, before it is exposed, it’s a different story. Kodachrome that has been stored cold will behave fine, even a few years past its expiration date. But unexposed rolls that have been stored at warmer temperatures, or that have gone many years out of date, can begin to get funky:

Kodachrome 25 expired by 9 years; magenta highlights

Aside from the color shift, this roll also had lost some of its speed and contrast. But the results gave kind of an interesting otherworldly feel, which some might enjoy exploring. (And the color shift was within the range where I could still correct it when scanning.)

If you miss the December 2010 deadline, there’s no way to develop Kodachrome into a color positive after that. However you may be able to salvage some visible image by developing it using black & white negative chemistry. Apparently there are a few complications to doing this, and I’ve got no firsthand experience with it. But I’ll report back if I get around to trying it.

Meanwhile, I leave you with the back cover of the November 1950 Popular Photography magazine. [Click to see a larger version.] This dates from an era when Kodachrome had a sensitivity of ASA 10! It wasn’t until 1961 that Kodachrome II raised this to ASA 25. (The ASA film speed scale wasn’t adopted by the ISO until decades later.)

Don’t be ‘beset by unhappy doubts’—shoot your Kodachrome soon

All the factory equipment was subsequently auctioned off and scattered to the four winds—seemingly spelling doom for Polaroid integral films (meaning the most popular, squarish 600 format; and the rectangular Spectra/Image type).

Why is the Polaroid lady twisting her own head off her body?

Fujifilm continues to produce “peel-apart” instant films which are compatible with some Polaroid cameras and backs. Ironically this means certain 1960s Polaroid cameras will remain usable longer than the ubiquitous recent Polaroid OneStep and One600 models. (The 1960s cameras do require a weird-sized battery, however.)

Fuji also makes its own line of integral instant films called “Instax.” However Instax technology is completely different from Polaroid’s, and none of those films are compatible with any Polaroid camera.

To re-create Polaroid’s 600 film from scratch would be a complex and costly process (remember that each pack also contains a special flat battery). It’s doubtful Fujifilm would have a motive to take on that challenge, when they already makes a competing product. There are not many other players in the market with the technical expertise to revive integral film; so unless some mystery savior appears, we should assume 600 and Spectra are disappearing for good.

Earlier this year, the best deal I could find on 600 film was from OfficeDepot’s online store. However for many weeks there has been no more stock available to my Zip code; and anecdotally that seems to be true for other regions of the US too.

So this week I stopped off at a local Target store—in my area, this is the last bricks-and-mortar retailer with decent quantities of 600 remaining. I got two serious shocks: First, the price had been raised to US $17 a pack (yes, that’s $1.70 per photo); and second, the expiration date on the packs was “09/09.”

09/09 is the mark of the End Times

Why is that date significant? A group of Dutch Flickr members toured the Enschede factory in May, two weeks before it shut down; and the date code they saw on finished packs was 09/09. Note that this is actually later than the “Aug 09″ final expiration date listed in Polaroid’s own phase-out announcement.

Apparently there may have been one final week’s production stamped with the expiration date “10/09.” Packs with that date are now available at the European “PolaPremium” website.

This new operation has made a splash selling small batches of various unusual Polaroid films, produced in the factory’s final days. (Some have questioned whether these special products were simply a way to use up old or substandard chemistry. I have no firsthand experience with these films; and considering the shipping charges to the USA, I don’t plan to try them.)

In any case, the message to Polaroid fans is clear. If you see packs with the date code 09/09, assume it’s your last chance to buy them. Ever.

(Well okay. I’m sure gouging profiteers on eBay will have packs to sell for the next few years—but at grossly inflated prices.)

So look deep within your soul (and your bank balance), and decide what it’s worth to you, to save a few final packs for special occasions.

Remember that the life of Polaroid films can be extended a few years past their expiration date by keeping them in the fridge. (Don’t put them in the freezer, because you can wreck the vital developer goo pods which form the image.)

Expired Polaroid film can lend interesting quirks to an image, due to color shifts and fading; or the image can be streaked or incompletely developed. But once film packs are many years out of date, they begin to fail entirely. Batteries die, or the developer pods dry out.

So don’t go hoarding more film than you would shoot in the next 2-3 years. You would just be taking precious shots away from another Polaroid-lover, who might be able to use it.

_________

Update 19 Dec. 2008: Just as I posted this, the Polaroid corporation announced it was filing for Chapter 11 bankruptcy. The reasons have little to do with instant photography; but the company’s troubles make the future of the Polaroid brand even murkier.

_______

Update February 2009: Yes, I have heard the excited talk about The Impossible Project to restart integral film production. It turns out some of the essential factory equipment was saved after all. I will be curious to see how it pans out. Here’s a good podcast radio interview with the technical head of the project.

Remember, they have numerous engineering hurdles to overcome before any film reaches shelves; once it does, it’s likely to be “quirky” emulsions at high prices.

Most rely on a well-proven Gaussian formula, whose nearly-symmetrical configuration automatically reduces certain aberrations like barrel distortion. The fast maximum aperture provides several benefits: A bright viewfinder image that “snaps” into focus, and the potential for pleasingly shallow depth of field.

Some photographers prefer a slightly wider lens as their “normal,” but on the 135 film format the typical 50mm focal length yields a very natural, neutral perspective.

Fast “normal” lenses—why are they such rarities in digital photography?

When one looks at the lens options for digital SLRs, however, the choices become a little murkier. By default, most DSLRs come equipped with kit zooms, with unavoidable penalties in lens size and maximum aperture. Many of these reveal quite obvious barrel distortion when zoomed to their widest setting.

On the other hand, fast 50mm primes are still offered by most cameramakers—as holdovers from their earlier film-camera lens lineups. Yet on a typical DSLR with a cropped-format sensor, these stop being “normal.” Instead they function as short portrait lenses.

That’s useful in many cases, to be sure; but it’s not the classic “all rounder” lens that 50mm represents on a traditional film camera. True, Sigma does offer a 30mm f/1.4 designed as a normal for cropped-format DSLRs; but its near-$500 pricetag is quite a shocker compared to past film-camera equivalents.

Recently, my eye was caught by an Olympus DSLR, the E-420. With this model Olympus has finally delivered on their original promise of very compact camera bodies through the use of a new “Four-Thirds” sensor format, one about half the width of 35mm film. In fact the E-420 is one of the few DSLRs whose size compares to the classic Olympus OM series of film bodies.

And even better, Olympus was pairing it with a new (non-zoom) lens—an impressively-tiny 25mm f/2.8 pancake. This focal length yields “normal” coverage on the smaller sensor format.

But an f/2.8 maximum aperture is hardly going to set pulses racing among serious available-light enthusiasts. And unfortunately, a recent lens test at DPReview.com shows that the new pancake is only a middling performer. Even this digital-specific, blank-slate design showed significant barrel distortion, as well as (freakishly) chromatic aberration even near the center of the frame.

Olympus is known for making some of the best-respected DSLR lenses today; yet their struggle was to design a well-corrected lens when the flange-to-sensor distance is significantly longer than the focal length. The classic symmetrical configurations don’t work in that case; so various optical shenanigans are required which inevitably degrade performance.

At this year’s Photokina photo trade show, Panasonic caused a great deal of buzz with an announcement of their Lumix G1. Panasonic seems to have recognized that the Four-Thirds format has never gained much traction with enthusiast DSLR buyers; yet it would offer consumers vast improvements in image quality over today’s teeny-chipped point’n’shoots, if only the camera body could be shrunk to acceptable size.

Panasonic’s way to achieve this was to omit the reflex mirror entirely (meaning only electronic viewfinders are possible), then halve the flange-to-sensor distance. This is the essence of a new “Micro Four-Thirds” standard, of which the the G1 is the first example coming to market (shortly).

Olympus showed their own prototype of a Micro Four-Thirds camera, made even more compact by omitting any eye-level viewfinder. This emphasis on small body styles has led some to hail Micro Four-Thirds as the arrival of the “digital rangefinder”—i.e. very compact cameras with interchangeable lenses.

Photo: 1854, the blog of the British Journal of Photography

Even the ribbed lens of the Olympus prototype seems reminiscent of certain 1970s compact RFs, like the much-loved Canonet QL17 GIII. (Note that the BJP article calls the prototype an “SLR,” which it is not: “R” stands for “reflex” viewing, precisely what Micro Four-Thirds models lack.)

But to me the most intriguing footnote to these announcements is buried in Panasonic’s future “Lens Roadmap.” In 2009, allegedly they will introduce a non-zoom, 20mm f/1.7 lens in Micro Four-Thirds mount. If we translate that to its equivalent on 135 film, this would be a fast 40mm lens—actually, just like 1972’s Canonet!—with a “wide normal” coverage that I would personally love. And because of the reduced flange-to-sensor distance, its optical design might even be a simpler, well-corrected symmetrical design.

So will compact, fast, well-corrected normal lenses come to digital, at long last? At a price we can afford? This remains to be seen, though I’m cautiously hopeful. But for the moment, virtually any random 35mm SLR from the past, equipped with its humblest possible lens option, offers something that remains a rarity in the world of digital.

_____

Update February 2009: Might Nikon be listening?

I had been taking photographs for over 30 years before ever hearing the term; and at first it confused me too. However it turns out that bokeh refers to quite a specific aspect of lens blur—calling attention to subtle phenomena that might otherwise be overlooked.

Unfortunately though, the meaning of “bokeh” has been getting rather blurred itself lately. We ought to make a stand to preserve its specific technical meaning, before this useful term degenerates into just another name for “fuzzy.”

A corny flower shot shows nice bokeh from a rotten plastic lens (on a vintage Diana)

Now don’t get me wrong—photographs which use selective focus to give nicely blurred backgrounds can be very pleasing. I like this effect, and have written about how to get it. And because many of today’s digital cameras limit your ability to achieve this look, a photo with shallow focus and a creamy blurred background will often attract many admiring comments about “great bokeh!”

But bokeh is NOT a synonym for “blurry background,” or “shallow depth of field.” It actually has little to do with the amount of blur. The degree of blur you see in out-of-focus areas is essentially a function of geometry—the relationship between the aperture’s diameter and its distance from the subject. Lets say you’re taking a portrait from 4 feet away using a 50mm lens at f/4. Every brand and every design of 50mm lens will render the background with the same amount of blur. But to the connoisseur, two different lenses may yield violently different bokeh.

Bokeh refers to the subjective quality of the blur. Is it “jangly” and busy-looking, or creamy and smooth? Do out-of-focus highlights have odd, distracting shapes, or are they unobtrusive circles? Does the blurred area seem to “swirl” around the center of the photo in arcs? These are some of the factors which might be mentioned as aspects of the bokeh for a particular lens. And these may be the reasons why a serious bokeh geek would chose one particular lens over a different brand with otherwise identical specs.

The word “bokeh” officially entered the English language in 1997, in an issue of the magazine Photo Techniques—whose editor Mike Johnston decided to add the final ‘h’ to make the pronunciation less ambiguous. He tells the story here, and includes some interesting photos showing different subjective effects in various blurred backgrounds.

Where does Bokeh come from?

But WHY might different lenses have different bokeh signatures? Well, there are two effects.

Each point of light from an unfocused area of the subject forms an extended bright patch at the image plane. Conventionally we call this a ‘blur disk,’ as if these were always circular; but really the blur spot takes on the same shape as the lens’s aperture stop. If the diaphragm blades form a 6- or 8-sided “stop-sign” shape (as SLR lenses typically do), so will the blur spot.

A most extreme example of this happens with mirror telephoto lenses, which have a central obstruction: Their blur disks are fuzzy doughnuts. This creates exceptionally distracting bokeh, if there are pinpoint highlights to accentuate it.

This crop from an Olympus XA shot shows busy diamond-pattern bokeh, matching the shape of the camera’s simple 2-blade aperture stop

Also, if a lens’s barrel design obstructs the more oblique light rays, the effective aperture opening becomes progressively more football-shaped towards the corners of the frame. This often leads to a “swirly” background effect if the lens is used at wide apertures.

The other issue has to do with a subtlety of optical design; namely, whether the blurred light ends up more concentrated at the middle of the blur disk or at its edges. A bright rim to the blur disk generally leads to distracting, jangly-patterned bokeh. But note that this effect often reverses depending on whether the subject is in front or behind of the focus point.

Both these effects are discussed in much detail in this excellent article (it is actually one of the original 1997 Photo Techniques articles mentioned above).

A blur disk with the light concentrated more towards its center will generally lead to smoother, creamier bokeh—and ironically one way to achieve this is to create a lens design which leaves some uncorrected spherical aberration. That compromises overall sharpness, so lens designers usually avoid it.

But there have been some specialized soft-focus lenses manufactured that exploit the effect; and it’s the reason why a plastic piece-of-junk camera often gives such dreamily smooth blur where the subject is out of focus, like in the Diana daisies shot I posted above.

————

Update: This page from Rick Denny compares the bokeh from several lenses of similar focal lengths; it illustrates very well how differently each renders out-of-focus highlights (scroll down the page to the photographs).

The museum shows several camera lenses cut open—one of which appears to be a Leica Tri-Elmar-M:

I realize this museum is in Berlin, and thus wants to showcase Germany’s well-regarded optics industry. Still, it’s a bit heart-stopping to note that this particular Leica lens is now out of production, and was hardly an inexpensive item—one currently for sale on eBay has a “Buy It Now” price of USD $3,350!

This is an unusual lens for M-mount rangefinder cameras, that can be switched to 28, 35, or 50mm focal lengths. Note that it is not exactly a zoom lens, as only those three discrete focal lengths are available. (This makes sense on a rangefinder camera, where the viewfinder also switches between several discrete choices of frame lines.)

It’s actually rather impressive that the lens achieves this (and has a good reputation for sharpness to boot) with so few lens elements (8 elements in 6 groups).

If you were interested in my past post about simple camera-lens design, check out the link. The complexity that goes into sophisticated camera lenses can be pretty impressive.

Yet for a classic rangefinder lover, the appearance of this new camera in 2000 was like throwing raw meat in front of a hungry beast: The “R” of the model designation might as well stand for Retro Rangefinder Revival.

For years I have happily bought and enjoyed used equipment; but the Bessa-R and its matching lenses actually caused me to open my wallet and plunk down serious cash on new-in-box photo gear for the first time in decades.

Think Different: Bessa-R versus Canon Rebel DSLR. With the digital’s crop factor, both lenses give equivalent coverage, speed

Despite the umlaut-festooned name, the current Voigtländer lineup comes from Japan’s Cosina. This is a company which has been quietly cranking out cameras for many decades—but with only a tiny fraction sold under their own name.

Over the past 30 years, Cosina ironed out a standardized kit of parts—including shutter, film transport and meter electronics—to build a basic, no-frills SLR. And this “platform” (as the auto industry would call it) was quickly adaptable to different lens mounts and camera styles. So, many entry-level cameras for other brands were really manufactured by Cosina, like the Nikon FM10, the Canon T60, Olympus OM2000, or any of the Vivitar SLRs.

The rest of Cosina’s business is making optical glass and lens components—and interestingly, building computer LCD projectors (again, all sold under other brand names). Anyway, they’ve developed quite a lot of expertise at being the low-cost supplier of optics and opto-mechanical products without compromising quality—for example, making aspherical lens surfaces cheaply.

But the president of the company, Hirofumi Kobayashi (the son of Cosina’s founder) is also a camera nut, and in particular, a lover of vintage German rangefinders. So eventually he had the brainwave, “Let’s launch a ‘prestige’ brand to show people what we can really do.”

After some wrangling, Cosina worked out the licensing to use the trademarks from the long-defunct Voigtländer company of Germany (although Cosina still doesn’t own the name outright). This licensing arrangement is why the cameras are still weirdly marked “Voigtländer Germany” on the top and “Made in Japan” on the bottom.

Cosina’s first bombshell product was an ultrawide, rectilinear 15mm f/4.5 lens of excellent quality (including an accessory wide-angle viewfinder that was impressive by itself). This was something even Leica & Zeiss hadn’t attempted before; and even if they had, it would certainly have cost thousands of dollars. But Cosina priced it at under $400.

The original camera body offered to match the 15, the Bessa L, was an oddball with no viewfinder of its own (scale focusing being fine, given wide-angles’ enormous depth of field). The L was a rather quick-and-dirty adaptation of Cosina’s standard SLR camera platform—but notable for its idiosyncratic revival of Leica’s pre-WWII 39mm threaded lens mount. (LTM or M39 for short.)

Voigtlander Bessa R; at left, the tiny 21mm wide-angle with accessory viewfinder

Cosina’s next body was the Bessa R, their first rangefinder model. This also shocked people, by offering an outstandingly crisp and bright viewfinder that was arguably clearer than that of many “classic” models from Leica, Contax, Canon & Nikon. The R’s film door and top and bottom panels were made of plastic, which caused some rangefinder purists to bristle, but this did allow for a lighter weight body.

And Cosina kept introducing more and more interesting, top-quality lenses. One strength of lenses for rangefinders is that without an SLR’s flipping mirror in the way, a lens can recess more deeply into the body and be made much more compact.

This Voigtländer Ultron 35mm f/1.7 can be much smaller than an equivalent lens for SLRs

Also, because slow lens speed does not reduce viewfinder brightness, rangefinders let you opt for even svelter, moderate-aperture lenses if you prefer that. For an interchangeable-lens travel kit, a Bessa R plus a few lenses saves ounces even compared to the far-from-chubby Olympus OM system equivalents that I own.

I bought and adore Cosina’s laughably-tiny 21mm f/4.0 wideangle (which needs an auxiliary viewfinder, natch). A lens this wide is not something you need every day; but it’s so compact that it’s painless to carry it along for the moments you do:

Compared to the elitist pricing of the Leica M series, Cosina simply rewrote the cost/performance equation for rangefinder lenses. They are not always cheap relative to SLR equivalents, but the build quality is always solid and nicely finished. The silky feel of the half-stop aperture detents on my 35mm f/1.7 is quite lovely.

A variety of companies besides Leica have made 39mm thread-mount lenses over the years, particularly Canon (in their 1950s, pre-SLR days). These all turn up on eBay from time to time, including some cheap, idiosyncratic Russian optics originally made for the Zorki and Fed rangefinder cameras. The quality control with these is quite erratic, but you can get lucky and find an excellent one.

Ratty-looking but surprisingly sharp 1961 Jupiter-8 lens, 50mm f/2.0. Standard equipment on many vintage Zorki Russian rangefinders

After the Bessa R, Cosina followed up with the R2 model. This was essentially the same camera, but with metal panels all around, plus a change to a different lens mount: Leica’s “modern” (post-1954!) M-series bayonet. (There’s an adapter which allows you to mount 39mm Leica thread-mount lenses on an M bayonet body, with no loss of function.)

The R and R2 bodies can be ID’d at a distance by their top plate, which slopes away diagonally from the viewfinder windows. All subsequent Bessa models have a step-down top panel design instead, and all continue the M bayonet mount and metal construction. The current bodies weigh about 10% more than my old plasticky R.

The bayonet attachment is obviously a lot less fumblesome than the thread mount for changing lenses quickly. (Threading lenses onto an M39 body goes easier if you rack their focus out to the closest distance, thus retracting the rangefinder coupling.) And the throat of the M bayonet is a little larger, allowing for wider-aperture designs.

So, while the original Voigtländer lens lineup used screw mounts, the trend is for the recent Cosina introductions to be M bayonet only. (Note that you can’t go backwards and mount those on an R screwmount body.) Cosina’s new M bayonet 35mm f/1.4 looks very nifty—even lighter and more compact than my f/1.7 threaded version. But the later lens releases have been accompanied by price increases, making them less of a bargain compared with the original series.

Cosina’s good work also caught the attention of Zeiss: Today there is a new Zeiss Ikon rangefinder line, manufactured by Cosina (in fact, just a re-engineered version of the Bessa series). This is intended as a showcase for Zeiss’s top-end lens designs. All but two of the lenses are assembled by Cosina, however.

Why Rangefinder?

But to take a step back for a moment… There’s a more general philosophical question, about what rangefinder cameras are good for.

Unlike an SLR, rangefinders allow you to maintain “eye contact” with your subject right through the moment of exposure. Also, the area visible outside the framelines can help you anticipate where unpredictably-moving subjects are headed next. So rangefinders have long been a favorite for people photos and street shooting.

I find that my Bessa is the camera I grab for group events, where I expect to take shots of people wandering around interacting with each other.

Compared to the shallow depth of field of an SLR groundglass, the RF viewfinder sharply renders all of the potentially distracting clutter within the frame. You might find this makes you more conscious of when you need to move around to get a clearer viewpoint.

On the minus side, rangefinders aren’t the best choice for exact, tight framing. Most rangefinders stop focusing at about 3 feet from the camera. Different brands have more or less accurate framelines, but you generally get a bit more coverage on the film than what the viewfinder showed. That’s fine if you don’t mind “casual” framing, or plan to crop anyway—but I’m a no-cropping purist, and this aspect sometimes bugs me.

An RF viewfinder does not change its angle of view when you switch lenses; with the Bessas you need to flip a lever to switch in different-sized framelines matching each focal length. But this means that for a particular viewfinder, there is only one frame size that makes the maximum use of the visible area.

Switch to a wider lens choice, and you’ll need an auxiliary viewfinder (slid into the camera’s hot shoe). Go to longer focal lengths, and you’ll be framing your subject in successively smaller rectangles within the middle of the image. Thus, on my Bessa R the 35mm frame is great to use; but the one for a 90mm lens is really somewhat marginal.

You get used to all this quickly enough—but it’s not quite the seamless experience of swapping lenses on an SLR. And manufacturing an interchangeable lens mount with rangefinder coupling adds a lot to the complexity and cost of the camera.

So if you always plan on using just one focal length, you might consider that there have been many delightful fixed-lens rangefinders made over the years—often quite a bit more compact than the Bessa line, and with leaf shutters bordering on the inaudible. I have long pined to own a 1970s Yashica Electro GX—nice and compact, with a fixed 40mm f/1.7 lens and aperture-priority auto. But they’re maddeningly difficult to find in North America, although apparently less so in Japan.

Anyway, today’s current Bessa lineup has become rather confusing, with autoexposure and manual versions of three different viewfinder variations. Two remarkable new models are the R4A and R4M—unfortunately the most expensive ones—whose viewfinder shows the view of wide-angle lenses up to 21mm without a need for auxiliary viewfinders.

All the Voigtländer “A” models offer both manual metering and aperture-priority autoexposure—although their electronic shutters mean they’re dead without batteries. I have not personally touched any of the auto versions, but the description/diagram of the viewfinder display makes it sound slightly distracting to me.

The Bessa bodies offer the cheapest entry point today into a modern rangefinder system. But as their prices have crept north of $500, I’m becoming less enthusiastic about their value-for-dollar (My own closeout Bessa R was just $250). Ultimately, they are all still derived from the same old econo-SLR platform, so they share a few quirks and flaws:

• Because they’re based on a retooled full-sized chassis, their height and width is really no smaller than a typical film SLR. It’s only in their front-to-back body+lens depth where they have a significant advantage.

• The metal, vertically-traveling shutter provides for a nice 1/2000 sec. top speed, and flash sync at 1/125th. But it is not nearly as quiet as the iconic, cloth-shutter Leica shlurp. Instead, a Bessa makes a pretty sharp “clack.” (Actually, if you want stealth, any leaf-shutter camera is the best alternative.)

• The strap lugs on the R are on the front of the body—unmoved from their original SLR position.

However, given the lack of a mirror box, and the smaller RF lenses, the balance is completely different: With lighter lenses, the camera hangs at an angle, digging its hot-shoe into your ribs. (The lug position doesn’t appear to have changed with the current lineup.)

• With the subjects I shoot, I definitely hit the RF close-focus limit more frequently than I’d like.

• A black Bessa R is finished in black paint over white plastic for its top panel—which makes its inevitable “brassing” look particularly tacky. (Would it have been so hard to mold that part using solid black resin?)

• Most worrying, the film transport derived from their old platform just doesn’t quite have the reliability that you’d expect in a $500 camera. About once every second or third roll, my R has a brief frame-spacing hiccup (though this appears to be getting better, not worse, with wear). And I’ve heard a number of stories of people experiencing total jam-ups. On the other hand, I’ve also heard of people “fixing” the problem by sharply thwacking their Bessa against the heel of their hand (if you try this, I’m not responsible).

Now against all of that, I must end by mentioning the positives: A Bessa is just about the handsomest camera around (especially in black). The viewfinder is simply a delight. (A DSLR-using acquaintance once literally gasped when she looked through it—this fueled at least two full months’ worth of smugness on my part.) And the wonderful Ultron 35/1.7 is on my short list of lenses to grab whilst running from my burning home.

Plus, a camera that flies so completely in the face of today’s camera mainstream is a proud, defiant badge of eccentricity—one I am happy to wear.

Recently I’ve posted my own grumbles about digitals’ lithium batteries, and their excessive depth-of-field. But I’d like to take a moment to discuss another subtle failing of digital point-and-shoots, one I rarely see mentioned: Aperture range.

Usually, the range of available apertures on a compact digital is a scant 2 or 3 f/stops. I’m serious: Go check out the “aperture range” listed in some typical specs …I’ll wait for you to come back.

Now, trading off aperture versus shutter speed—to control motion blur and depth of field—is a cornerstone of creative photography. So how could digital camera manufacturers permit such a crippling flaw?

Remember that f/ numbers are defined as the ratio of the lens focal length to the aperture’s diameter.* So as you change to longer focal lengths, the size of an “f/8-sized” hole must grow larger too. But teeny digital sensor chips demand ultra-short focal lengths; on a digi compact, f/8 could mean an opening only 1 millimeter across.

A hole that tiny almost starts behaving like a pinhole. And as any pinhole photographer will tell you, if your hole’s too tiny, you run into a problem: the limits of diffraction.

Diffraction: The Optical Wild Card

Optical calculations generally assume that light rays follow mathematically straight lines, until they’re bent by some air-glass surface.

But it’s not that simple. Because of its wave nature, light grazing the edge of an obstacle can veer off-course. These light waves diffracting in random directions can reduce the sharpness and contrast of an image.

As the aperture stop of a lens becomes smaller, an increasing proportion of the light grazes the edge of the iris opening—rather than passing unaffected through the middle.

Rather than focusing to a single point, this stray light forms a fuzzy bulls-eye pattern instead.

How all this affects image sharpness gets into some choppy waters, technically. With digital cameras, the diameter of those bulls-eyes can can even grow larger than the spacing between sensor pixels. You might have paid for an 8 megapixel camera—but if each point of light is smeared over several pixels, you might effectively get only 2! (For a more technical discussion of these complications, I recommend this helpful page.)

Note that zoom lenses add another wrinkle: An opening of a given diameter equates to different f/stops, depending on what focal length the lens is zoomed to.

The diameter of the lens elements determines the widest possible aperture; but the corresponding f/number changes as you zoom (this is the reason zooms list a range like 2.8–4.0 as their maximum aperture). And if the smallest-possible iris diameter remains constant, this equates to dimmer f/stops (higher f/numbers) as you zoom towards the telephoto end of the range.

Anyway, the upshot of all this is simple: There is some limit to the smallest f/stop which can be used, before diffraction damages sharpness too severely. And tiny image formats suffer the worst from this effect.

A digital point-and-shoot’s lens might stop down to only f/5.6 at the wide-angle setting; and f/8 at the telephoto end. (Contrast that with with lenses for 35mm cameras, which can close down 2 to 3 stops further.)

What about opening wider?

This explains why compact digitals must limit their smallest f/stop. But could we extend the f/stop range to wider apertures? This would also help with our image noise and depth-of-field headaches too.

But the problem here is not optical; rather, it’s a function of what today’s camera marketplace demands.

You can’t increase a lens’s maximum f/stop without adding to its diameter, weight, and cost. And with compact size being a highly-desired camera feature, that’s a tough sell today. Thus, it’s rare to see maximum apertures greater than f/2.8 (and remember, that only applies at the wide end of the zoom range). Compare this to 35mm film compacts of the 1970s, where affordable models often sported excellent, faster-than-f/2.0 lenses.

Actually, the tiny sensor of a digital compact would permit the design of an ultra-fast lens—let’s say f/1.4—that would be smaller and cheaper than its 35mm equivalent. But the snag is, it would not be a zoom. Zoom lens design has made enormous strides in the past decades; but we still can’t avoid a speed penalty of a couple of f/stops, compared to the best single-focal-length designs.

And zoom range is a major tick mark on today’s camera-shopping checklist. With a few notable exceptions, camera makers consider it commercial suicide to offer a zoomless camera (aside from their cheapest and tiniest models).

Both lenses open to f/2.8—the difference is in their focal length. Smaller image formats allow more compact lenses of the same speed.

Between a rock and a hard place

So lens design for small-sensor cameras is hemmed in at both ends. The smallest f/stop is limited by diffraction. The widest is limited by market demand for zooms—but compact and inexpensive ones. Thus the range of available apertures only covers 2 or 3 stops—seriously limiting creative exposure choices. (With primes for 35mm, the range can be 7 or 8 stops instead.)

As new digital models are introduced, some oft-heard complaints—like autofocus lag and poor viewfinders—are gradually being addressed. Other problems like image noise in dim light might be tamed eventually, using better sensors or savvier image-processing.

But any lens is still bound by the laws of optics. No amount of technological whiz-bangery can change that. Compact cameras imply tiny sensor sizes; tiny images imply short focal lengths.

And with those limitations, using aperture in the way a creative photographer demands becomes impossible.

*Note: This explanation of f/numbers ignores a few optical complications, which do not affect the main point here.

Cemetery sculpture: camera ‘E’

Cemetery sculpture: camera ‘M’

Both show the same subject, from the same camera position, and include the same angle of view. The tonal ranges may not match perfectly, but I’ve adjusted them them to be pretty close. So what’s the difference?

Well obviously, the upper image shows much more detail in the background. The zone of sharp focus—aka depth of field or DOF—seems to be much deeper. This is particularly obvious with the twigs at the right edge: The lower image isolates a few; the upper one shows a whole thicket. So that first image must have been shot at a much smaller f/stop, right?

Heh, heh. Gotcha. Would you believe me if I told you both of these photos were taken at the same aperture—f/4.0?

So what on earth created such a huge discrepancy in the depth of field? Simple: The bottom image was shot on 6×4.5 format film, using a 110mm lens. The upper one was shot on my digital compact, where the corresponding focal length is about 11mm. (The digital image has been very slightly cropped to match the film frame.)

Let’s take a brief optical detour to understand why this happens.

Focal Length Changes with Format

Old-school photographers who jumped between 35mm, 120 and sheet-film cameras learned that as you switched formats, you also needed to shift gears thinking about focal lengths.

For any desired lens coverage (say a “portrait” lens, or a moderate wide-angle), the corresponding focal length scales up in proportion to the film’s dimensions. For example, one well-known guideline is that a “normal” lens is one whose focal length equals the diagonal of the image.

Thus, a normal lens for 4×5 sheet film has a focal length of roughly 160mm. But the tiny sensor chip in my digital compact could fit atop a pencil eraser. So the zoom setting corresponding to a “normal” lens for this format is about 7mm.

And those exceptionally short focal lengths lead to extreme depth of field (just as switching to shorter focal lengths does with any camera).

Fuzzy about Focal Lengths

By the end of the 20th century, most serious amateurs shot 35mm exclusively—so, awareness of how focal lengths related to film size began to fade. It became an easy shorthand to quote 35mm-format focal lengths to express how wide or narrow a lens’s angle of view was. If you said “I’m shooting with a 28 today,” everybody understood what kind of perspective you were talking about.

But the digital-camera explosion spawned a plethora of different chip formats—all much smaller than 35mm. Focal length confusion roared back with a vengeance. Camera marketers despaired that publicizing this crazy-quilt mix of new focal lengths would cause chaos and confusion. What did a “12mm lens” even mean in this brave new world?

So they latched onto the well-intentioned—but wholly fictitious—crutch of specifying “35mm equivalent” focal lengths.

These bogus 35mm equivalents will only serve to confuse us. So please, forget they ever existed! It’s the actual, optical focal length we are talking about here. (If not labeled on the lens, it might take some rooting around in your owner’s manual or the image EXIF data to find out the true f.l. numbers.)

Whither Depth of Field?

Meanwhile, the DLSR world had its own confusion. Photographers could now put their old film-body lenses onto new digital models, with “crop factor” sensors. But if a 50mm lens “acts like” an 80mm on your DLSR, then what happens to the DOF? It’s still the same lens, right? To this day, whenever the issue comes up in online forums, discussions can get quite agitated, with dogmatic assertions and misleading generalizations.

Once again, the DOF issue has nothing to do with digital—it’s simply another effect from scaling between one format size and another.

But confusingly, there are two factors to consider:

As I mentioned, smaller image formats require shorter focal length lenses to capture an equivalent angle of view. And a shorter focal length lens gives deeper DOF.

However, a smaller format must be enlarged further to give an equal-sized image to view (whether printed or onscreen). This higher magnification means any fuzziness becomes more detectable—i.e. the limits on how much blur is acceptable become more stringent. So paradoxically, you’d expect depth of field to get shallower with smaller formats.

in this Optical Steel-Cage Death Match, I’m sure you’re breathlessly wondering—who wins? Well, I won’t keep you in suspense: The focal-length effect trumps the magnification one. The essential rule is:

When f/stop, distance, and lens angle-of-view are all held constant, the larger the image format, the shallower the depth of field.

Of course, shallower depth of field is a double-edged sword. Here’s another pair of photos—also both taken at f/4, again differing only in image format:

Depth of field: small-sensor digital camera

Depth of field: 645 film format

The lower, 645 version helpfully takes all that distracting background junk and throws it out of focus. Yet even with this flat-ish subject, only the raised hand is in good focus; the eyes and the right edge of the sign drift off into fuzziness, just because it was at a slight angle to the camera. Shallow DOF places more demands on the photographer (and here, I blew it).

But an equally vital observation is this: Any camera based on a tiny image format (and this includes virtually all digital point-and-shoots) gives you no chance to use selective focus creatively.

All the digital P&S images shown here use the longest available focal length, the widest aperture, and are shot fairly close to their subjects. Thus, they represent the most favorable combination for achieving shallow, selective focus. Yet in all cases, the backgrounds appear sharp enough to be cluttered and distracting.

If this still seems all rather unintuitive, it helps to try plugging some representative numbers into a depth-of-field calculator, like this nifty online one. Remember—you must change the popup for image format, not just the focal length. (Depth of field calculators make certain simplifying assumptions, and are not the last word on the subject; but they’re a great way to compare what happens as you change one factor at a time.)

But, is this just a minor geeky technicality, or does it matter… artistically? Well, here is my final photo comparison. Same ground-rules as before:

Depth of field: digital point & shoot

Depth of field: 645 film format

Scout’s honor here: I did not move the tripod an inch. Now I’m hardly claiming the second image is timeless art. But the upper (digital compact) version is just a tangled, impenetrable mess. With no way to blur the background, and pull the viewer’s eye towards the cattails, there is simply no photograph at all.

Yet today, millions of people are doomed to use cameras where this effect is unavoidable. Again, this is not a “digital-versus-film” issue per se; it’s simply an inevitable result of the smaller image format.

But is it really fair to compare a cheesy digital point-and-shoot to a serious medium-format camera? Well, there’s one last little detail I haven’t mentioned yet.

The film camera I used here is a Mamiya m645—a recent purchase off eBay. While well-worn and probably 30 years old, its sturdy construction seems ready to keep kicking for a few more decades. The 110mm lens was “bargain” grade from KEH.com. The price? Actually—together they cost $50 less than what I paid for my digital point & shoot.

Lomo is a Russian manufacturer of optics and related products. They’ve used the LOMO name (which in Cyrillic can look like “nomo”) since 1965; they existed under the name GOMZ before that. The “L” originally stood for Leningrad—though the city itself has now restored its older name of St. Petersburg.

Lomo is a sophisticated and diversified maker of optics, particularly for the Russian military (check out the night vision goggles on their website). Actually in the GOMZ era, they created one of the very first 35mm SLRs, the 1936 “Sport.” They also made the inexpensive Lubitel and Smena cameras.

The “Lomo Compact” or LC-A camera was their 1980s low-budget knock-off of the Cosina CX-2. The front panel of Cosina’s camera rotated to cover the lens and viewfinder; Lomo dropped that feature, but amusingly, kept the round-topped Cosina body shape. The CX-2 had a 5-element lens; the LC-A substituted a cheaper three-element lens—with the side effect of very noticeable vignetting at the corners of the frame.

Eventually, a couple of Austrian students fell in love with the quirky LC-A, and founded the now-infamous Lomographic Society. This is not a club, but a business who obtained exclusive rights to sell the LC-A in countries outside Russia.

Their “hip, edgy” marketing annoyed many people—because the LC-A was never all that different from many other cheap autoexposure cameras. However the business has been very successful, and now they use the “Lomography” branding on a bunch of different plasticky cameras–mostly made in China and having nothing at all to do with the Russian Lomo company. Today, even the original LC-A has ceased production; the Lomographic Society has commissioned a Chinese re-creation, the LC-A+.

Thus, it’s incorrect to dub all cheap and zany plastic cameras “Lomo.” It might be accurate to call the cameras marketed by the Society “Lomography” cameras—as long as one remembers that they have no camera manufacturing facilities of their own.

The Lomography Society is one of the major sellers of Holga cameras; but they did not invent the Holga (which originated in Hong Kong), and it is not in any sense a “Lomo.” The Lomography Society simply resells Holgas, in pre-packaged bundles with higher prices.

If the fashion-victim aspect of the Lomography Society turns you off, keep in mind that a plain vanilla Holga is widely available at ~$25 USD. If you’re feeling adventurous, there are many other inexpensive, DIY plastic-camera options like flipping the lens on a Brownie Hawkeye. If you you’d like a more LC-A flavored camera with a decent lens, a good choice might be a used Olympus XA2, which often sell on eBay for under $30 with shipping.

[Originally posted in slightly different form on Flickr's 'I Shoot Film' forum. ]

Postscript

Since first writing this, I’ve mellowed my opinion about the Lomography Society somewhat. They certainly have done more than anyone else to bring new people into film photography; converts who presumably will outgrow LSI’s gimmicky limitations and move on, wiser and somewhat poorer.

The Lomography story is the story of modern commerce, textbook “brand marketing”: Attach an aura of coolness to a particular name, then reap the rewards in recognition, sales, and profits. It’s hardly any different from Nike, Apple, or countless others.

But I still chafe at them hijacking the name Lomo.