dante stella stories photographs technical guestbook

Uh oh...
Konica Hexar RF: The final frontier?

Hexar RF, Leica 35/1.4 Summilux-M ASPH; Konica Hexar case

Before you read this, bear in mind that I have used both a Leica M3 and a Konica Hexar RF for the last two years. I don't think there's anything wrong with the M3, but I do have to say the Leica M was a great camera... the first time it was made, 50 years ago.

1. Antecedents to the Hexar RF

Shutters and differences in camera architecture lead to holy wars. Like it or not, a camera is basically a box with a lens on one end and a shutter on the other. The story of the rival rangefinder factions begins in the early part of the 20th century, with the miniature focal plane shutter found in the Ernemann Ermanox. The Ermanox was basically a 100mm f/2 Ernostar lens with a rubberized-cloth focal plane shutter in front of plate film. No rangefinder, just a folding glass viewfinder. Erich Salomon used this discreet camera to devastating effect.

Leitz, a noted microscope maker, seized upon this idea in the mid-1920s and made a version of the Ermanox that shot movie film. Its frame size, 24x34mm (later changed to 24x36), or two movie frames, would become the standard-size negative the world over. The Leica (LEItz CAmera) shutter traveled left to right, as opposed to the Ernemann up-and-down. In 1932, Leica installed a coupled rangefinder on this camera, and the Leica II emerged. Not to be outdone, Zeiss Ikon (an offshoot of Carl Zeiss, another microscope maker) created a new coupled rangefinder (shortly afterward combined with the viewfinder), and the Contax I was born that same year. The Zeiss shutter was made of metal blinds that rolled from top to bottom. It also purported to hit 1/1250 sec. Both the Leica and Contax cameras were built around the shutter mechanism, which required great mechanical force to operate.

The basic designs of Leicas and Contaxes were static until 1953, when Leica adopted a Zeiss-style bayonet lens mount and combined viewfinder/rangefinder in its M3. The M3 was still using the basic shutter from 1925. The Zeiss shutter had not changed its basic design, either. Both shutter mechanisms took up most of the camera body. At this point, the basic design constraints were still present: shutter fatigue, vibration, and slow synch speeds dictated by the fact that the shutter curtains have to travel across 36mm of film gate. Despite these limitations, Leica-style shutter, whether in cloth, titanium foil, or stainless-steel foil, would continue to do yeoman duty with Leicas, Nikon rangefinders, and Canon rangefinders. It even went on to work in some Canon and Nikon SLRs until the mid-1980s.

As Zeiss-Ikon was folding in the early 1960s, the Japanese photo industry was beginning to experiment with modular camera assemblies. One product of this was the fan-blade metal focal plane shutter, which owed nothing to the historic focal-plane shutter design. The benefits of the fan-shutter were that covered less ground (allowing a higher synch speed), had lower moving mass, required less force to operate, and would not burn through if pointed at the sun. And best yet, it was a modular part which could be replaced wholesale if it failed. The first successful model, the 1960 Copal Hi-Synchro (which sported a synch of 1/125 sec and a top speed of 1/2000), represented seven years' work for a rogue engineer underwritten by Konica. Unfortunately, the Hi-Synchro was impossible to produce economically and was replaced by the slower-running Copal Square I and Square S. The Konica Autoreflex T and the Cosina Hi-Lite were two of the cameras using the Square S. Interestingly, the Copal shutter was licensed to a consortium that did not include Nikon (the Nikkormats with these shutters were actually made by Mamiya, not Nikon). The Copal SE was an electronically-controlled version of the Square S.

Over time, the Square S would be tweaked into the 1/4000 sec mechanical shutter on the Nikon FM2 and the SE would become the 1/4000 sec electronic shutter on the FE2 and FA. The final development of the SE would be in to a shutter that was both electronically timed and driven by a stepper motor instead of a spring. This was the version found in the Konica FS-1, the first 35mm SLR with an integral motor drive. The motor-driven Copal shutter would find its way into the Nikon N2020 (and all successive Nikons with integral motor drive), the Contax G, other cameras, and ultimately, the Hexar. The Hexar's version is the same as the one in the Contax G: it has carbon blades, a top speed of 1/4000 sec, and a flash synch speed of 1/125 sec. The motor-driven shutters have an added benefit for mirror-free cameras like the Hexar and the G2: they reset themselves instantly, obviating the need for a second set of curtains (as the Cosina Bessas have). Incidentally, the Bessa shutter is very similar in specs to the original Hi-Synchro shutter.

The slow march of progress. During the development of the camera concept that would give birth to the Hexar RF, Leica remained very stable. Other than changes in finder magnification, framelines, takeup spools, and flash synch terminals, the Leica M cameras underwent virtually no development between 1953 and 1985. The only technological standouts were the M5 (which Leitz essentially disowned in favor of the M4-2) and the Leitz/Minolta CL. Both included built-in light meters, but neither was in the mainline of Leica M development. Minolta later made the CLE, which incorporated an electronic shutter and TTL flash metering, but this too was something of a special product, since it was incompatible with a large quantity of Leica gear. Leitz's major improvement in the 1980s was the M6, which now incorporated a match-diode meter system. The last Leitz (well, now Leica) addition before the Hexar RF was TTL flash metering.

2. Emergence of the Hexar RF

Konica started as Konishiroku in 1877 as a supplier of photo materials, making it a bit older than Kodak. It made its first camera in 1918, about seven years before Leitz made its first 35mm camera. The name "Hexar" was actually the name of the first lens Konshiroku had made under the Rokuo-shah brand. When Konica made its first postwar camera, the Konica I, it had a 50mm f/3.5 Hexar lens and a coupled rangefinder. In many ways it resembled the Leica II, except that the viewfinder/rangefinder was one window. In the mid-1950s, the 50/3.5 Hexar lens was made in Leica thread mount, along with a 50/1.9 Hexanon lens (drawn from the last models of the Konica I, which featured 6-element planar-type lenses). The Hexar name fell out of use with Konica at some point in the late 1950s in favor of Hexanon (which with the SLR line would not only be applied to double-gauss lenses but also retrofocus wideangles).

The Hexar name was resurrected in the late 1970s, when "Hexar" was the name given to Konica's more economical SLR lenses. The first camera to bear that name was the 1992 Konica Hexar autofocus, which combined a very close copy of the Leica 7-element 35/2 Summicron with a leaf shutter and a silent motor drive. According to Konica lore, the Hexar autofocus project was a reward for engineers who had worked on the SLRs and were now demoted to making point-and-shoot cameras. Sometime you should compare the top deck of a Konica FT-1 and a Hexar Silver.

Konica resumed making Leica thread-mount lenses in small runs in 1996, as they had in the 1950s. They started with the 35/2 lens (which was redesigned in 2001) that was in the Hexar autofocus model and then made a 60/1.2 and a 50/2.4 collapsible. Konica and Ricoh began experimenting with prototypes of M-mount cameras in the mid-1990s, probably in anticipation of the Leica M-mount patent's expiration. Numerous rumors began to circulate about a Hexar II. Some said it was autofocus; others said manual. Some said it had a brass body; others said more modern materials.

When the Hexar RF sprung fully-formed from the head of Konica in late 1999, it had exactly two things in common with the Leica M: a lens mount/register, a rangefinder mechanism, a 51.6mm normal lens, a cable release socket and a hot shoe. Not a single other vestige of Leica design remained. Where Leicas were being covered with oxidized zinc, the Hexar was fitted out with flat black epoxy-coated titanium. No one knows the exact origin of the Hexar's design, but it is probably fair to say that it was the same design house that created the Hexar autofocus, the Contax G2, and the Hasselblad/Fuji X-Pan, all of which the Hexar RF strongly resembles.

Why wasn't the Leica M7 more like the Hexar RF?

... Jun Ikuyama and Taku Wagatsuma filed for and received a utility patent (06,389,234 7/6/2000 in the U.S.; 7/7/1999 in Japan) on the Hexar RF body as a photographic system, which they then assigned to Konica. The principal innovation was the integrated shutter/lightmeter module, which is related both to the ease of magnification and to reducing measuring field error. While I am not sure how well this would hold up if challenged, one thing that is clear is that it would be a stab in the eye to anyone trying to make a similar Leica product.

Interestingly, some things that were in the patent that apparently did not get into the Hexar RF were:

— An integral date projector (a la Canodate E) for imprinting the date on the film;

— An error controller to make some type of correction when no x-synch signal is sent by the shutter. It may well be the case, reading on, that this is the failed shutter error code on the top LCD; and

— A variable metering pattern (spot to full-frame) (Fig. 10) with in-viewfinder indication (Fig. 11)

Top: the variable coverage metering; bottom, the view through the viewfinder.

3. Holy War

A lot of people in the Leica community were not ready for a product like this. Long priding themselves on photography through suffering (and if that doesn't work, match-needle metering) and telling themselves their photography sprang from the lack of technological bells and whistles, there was no way they could have been. Probably in shock at the idea that you could suddenly use Leica glass while jettisoning a lot of the historical baggage, they immediately derided the Hexar, speculating that it was unreliable (while all the while sending their Ms for mandatory service every couple of years). What could they do when faced with a machine that could shoot Leica glass wide open with 1/10 of a stop film-plane exposure accuracy?

The initial shock gave way to what could be described as the Kellog-Briand pact of photography. Some Leica people spoke in more conciliatory tones — that the Hexar would serve as a good "second body." This was a politically-correct shibboleth; you could have your Hexar as long as you shelled out for a Leica. It's kind of like saying it's okay to have your hot mistress as long as you keep your ugly wife.

The final stage of Hexar rage came in the form of an oft-repeated tirade on the film-flange distance being different, with the Hexar being 0.2 mm longer flange to film. This somewhat apocryphal story is the combination of a couple of sources. The first is a story about a preproduction model sent to B&H not being able to focus Leica lenses (no way to verify). The second was a widely publicized problem with one example of a Hexanon lens that focused *past* infinity on a Leica. If the lens were collimated properly, it would have indicated a shorter film-flange distance, not a longer one (if it indicates anything at all). At last report, the Hexar and the Leica the same flange to film distance, which is 27.80mm from the flange to the film surface (although they create this distance by slightly different methods). Maybe the most telling thing about the "incompatibility" issue is that it is voiced most loudly by people who own Leicas and Cosinas.

The advent of the Leica M7 added even more speculation to the mix: that the Hexar RF was the M7 but Leica decided it wasn't up to snuff; that the M7 would be like the Hexar RF but say would Leica on the cover; and on. What the M7 discussions indicated was a hope that Leica would endorse some type of technological change. When the M7 did arrive, in February 2002, it was... an M6 with an electronically-controlled shutter and autoexposure. Otherwise, there were no changes whatsoever. Once again, Leica successfully avoided new tooling, new technology and rocking the boat rowed by its diehards.

4. Can't we all get along?

Even now there is some emnity between Leica and Hexar users. There are some good reasons we can expect the Leica/Hexar holy war to continue forever:

The film-flange distance issue is insoluble. Film-flange distances are notoriously difficult to measure. Part of the problem is that measuring to the pressure plate can cause the pressure plate to move. Measuring to the film rails ignores the behavior of film, and here the differing behavior as between the two rival models. Collimating from a first-suface mirror ignores the frontward curl of film that is exposed to ambient humidity. Any practical focusing test is going to be problematic with lenses whose plane of focus is curved. True flange-film measurement is chimeric, and it would require a lot of test samples. For a more detailed discussion of the film-flange distance with the Leica and the Hexar, click here.

Most users do not know what they are looking for. Film-flange distance issues arise with fast wideangle lenses, generally of 21mm focal length and less. This is due to the fact that at infinity with these lenses, the lens travel front-to-back is much shorter. If the flange-film distance is a tiny bit longer, focus will fall well short of infinity. Problems with fast telephoto lenses close-up, where small changes in distance require large front-to-back lens movements, are more likely the result of rangefinder alignment problems. On any Leica-type rangefinder, whether labeled "Leica," or "Konica," accuracy decreases dramatically inside 1m, to the point where even Leicas do not focus Summilux 75s correctly at 0.7m.

Leica users are too emotionally invested in their equipment to believe that anything could be better. Leicas are expensive pieces of hardware. Although some pros use them, the market is has a fair number of other professionals and everyday people who feel like they have worked themselves silly, have been dreaming about Leicas for years, and want some leisure. They lack the time, training or inclination to muck around with testing, and they want to buy something that they can be confident will take the best possible pictures. The whole point for them of buying a Leica is that there is nothing better.

Konica users are too emotionally invested in their equipment to believe that anything could be better. Konica has always enjoyed cult status among its users (especially of the SLRs or the autofocus Hexar). Hexar users (which have their share doctors, lawyers, academics and hobbyists) largely suffer from the same lack of time, training or inclination to do a lot of testing. They are largely fixated on the utilitarian, to the exclusion of enjoying the fine workmanship of 1950s and 1960s Leicas or seriously testing their gear.

Ford vs. Chevy. Well, this should be self-explanatory to most Americans...

Mercedes vs. BMW. for the Germans...

And now, if the two sides will get into their respective bunkers...

5. The Hexar RF in actual use

Enough religion and politics! Because the Hexar RF is primarily a functional tool, I'd like to discuss some of its salient functional aspects. This is my 2 year, 250-roll report.

Hexar RF with 35mm f/2 4th-generation Summicron, wide-open

Ergonomics. The Hexar RF is a very boxy camera, and it does not curve into the palm of the hand the way the M3 or even the Hexar AF does. This, I believe, is intentional and due to the placement of the shutter speed dial. That dial sits close to the back edge of the camera, and if you gripped a Hexar RF the same way you gripped a Leica M, you would not be able to turn the shutter speed dial with your thumb (bravo for this idea). The camera also feels different in the hand because it has grippy neoprene surfaces that are easy to hold onto. The camera is heavy, as heavy as a Leica M. This helps handholdability. Although to my knowledge the reason why the Hexar is so heavy has not been discussed, I suspect that the body casting is heavier-weight than the one in the M3. The titanium covers on the Hexar are definitely lighter than the zinc or brass ones on the Leica Ms, so something inside must be pretty dense.

The control arrangement is well thought out. The shutter speed dial is on the back right corner (looking from the top). This allows the shutter speed to be changed without dislodging you grip on the camera. Your thumb turns the shutter speed dial. Your index finger turns on the camera via a switch coaxial with the shutter release. You have to look at the top to change the exposure compensation.

A closely allied topic is fit and finish. Normally, this is not a functional aspect. However, the Hexar's finish is very utilitarian, being a flat epoxy. This makes the camera warm to the touch and also makes it impervious to scratches. The only real wear spot is the metal shutter speed dial, which has a lot of sharp edges that, when the dial is locked in AE, tend to lose their paint. All of the controls are positive and difficult to dislodge.

Loading. The loading of the Hexar RF is pretty straightforward — pull the leader to the red mark and close the back. Press the shutter release. The film will wind to frame 1, and if it doesn't load correctly, the counter will flash. The only loading problems I have experienced are with ultracheap Kodak Gold 100, which due to the rough handling it must get on the way to Target, ends up with crimped leaders. Of course, if you can afford any M camera, you should probably be shooting something better than the cheapest film — I find that for quick 'n' dirty scanning, Gold 100-6 is pretty hard to beat. The camera counts frames optically, so there is no sprocket wheel to break.

Focusing. The focusing regime on the Hexar is pretty straightforward: it is just like any other M camera. There is a coincident rangefinder spot that has sharp edges for vernier focusing. The Hexar has a relatively low magnification finder to accommodate eyeglass wearers and wideangle lenses. This works two ways with 50mm lenses: either you get a lot of outside-the-frame space to work with or it looks too small; it's probably more a matter of taste. While a lot of people say the finder is dim, I think it's about as bright as an M3 finder (if not brigher), and where you can't focus with this finder is where you wouldn't be able to shoot anyway. You can get an inaccurate RF reading but you have to get a bit off center from the eyepiece to do so. Konica has promised diopter lenses, but I have not seen any. The only other issue of note is that with low-magnification finders, it seems easier to focus using lenses that go from infinity to close focus in a short turn of the ring (see below).

Exposure control . This part is easy. The recommended (or in AE, the actual) shutter speed appears on a vertical scale on the left side of the viewfinder. In manual mode, you make the blinking and solid readings line up. The scale goes from 1/4000 to 1/4, and there is a +/- indicator if you have activated exposure compensation. A lot of people tend to complain about the lack of indication for slower speeds than 1/4, but let's face it: handholding is about useless down there, and if you're using a tripod, you won't care much about whether it's 1/4, 4 or 30 seconds. The meter samples about 30 times a second from a white center shutter curtain. The meter will clear lenses with deep rear elements. The 35/2.8 Jupiter changes the metering pattern slightly, but it will still work ok with the meter.

Shooting. Press the button, and it shoots. If you set the camera to continuous, it will shoot a little over 2 frames per second (assuming that you shutter speed is 1/15 or faster). The shutter button, like any other electromagnetic shutter release, has two stages. Light pressure turns on the meter; heavier shoots the camera. There is a little bit of a lag, but I don't attribute this as much to autoexposure as I do to the fact that a small motor is running the shutter. The noise level from the shutter is low; the camera has some whine from the motor, but you can attenuate the effect by shooting in continuous mode (firing and winding are nearly simultaneous).

Flash. The included HX-18 flash is small and powerful, but not a real substitute for a Nikon SB-20 for those big jobs. The SB-20 is a great flash for the Hexar, because the Nikon dedication pin for the ready light automatically sets the Hexar for 1/125 sec if you have the camera in AE. The SB-20 is even better because it's powerful, has multiple auto modes, and can bounce. But who are we kidding? Matsushita makes all camera flashes out of Japan.

Rewinding. The camera will extract about 38 pictures from a roll of 135-36. When it hits the end of the roll, it will rewind (you can interrupt this by switching the camera off). You can also manually rewind by pressing the rewind button with the little nub on the strap buckle. The camera will pause before sucking the leader in.

Malfunctions? I have had very, very few. Aside from the aforementioned loading issue with Gold 100, I have had one other problem. Once the shutter was on B and it closed immediately. And it never happened again.

Some people have complained of rangefinder focusing past infinity with Leica or Konica lenses, but from what I can tell that was an early-production problem. From what I understand, the material of a pivot was changed from stainess steel to brass (or vice-versa). As a practical matter, the rangefinder going past infinity by a little bit is not a huge problem. First, the rangefinder shows misalignment well before actual focus is affected. Second, if the rangefinder is focusing past infinity, you are likely focusing in front of the where the rangefinder says. This is not all bad, since you get twice the depth of field behind the point of focus as in front. But the past-infinity problem is easily remedied by sending the camera body in for warranty service.

Battery use and checker. The only LCD display on the camera shows the frame counter and the battery indicator. The battery indicator is on full-time. The camera will do about 120 rolls of film between sets of batteries, and it goes into sleep mode if you leave it on for too long without shooting a picture (10 minutes). Once the battery indicator shows low battery, you can belt out an entire roll of 36 frames. How they get this much out of two CR2 lithium batteries is amazing.

Lens compatibility. Many electrons have been spilt over this issue. The issue of lens compatibility has not been helped much by pronouncements from Leica and its subsidiaries, which range from the uninformative to the confusing to the conciliatory. It is hard to tell what to make of this, since no one can translate Konica back-focus specs into meaningful terms for Leica and vice-versa.

An oft-repeated comparison is a 28.00mm back focus distance of the Hexar vs. a 27.80mm back focus distance of the Leica. This is an apples-to-oranges comparison based on two different measurements: the rails versus the film position. If it were the same measurement, a Leica wideangle could never focus anywhere near infinity on a Hexar RF.

I think that as a practical matter, potential incompatibility is an overblown issue since flange-to-film distance is easily changed and so is rangefinder adjustment. So even if you manage to turn up a problem, in most cases it can be corrected. Maybe I was lucky with mine (14462xx) in that it worked well out of the box. I do not tend to regard focal plane error to be any more of a problem with the Hexar than it is with any older Leica M, since as Ken Ruth reminded me today, an M3 will often be up to 9/1000 of an inch smaller after 50 years (0.222 mm, greater than the greatest difference ever alleged between RF and M back-focus distances).

It is worth noting again that focal-plane distance problems will manifest themselves at infinity with fast, wide-angle lenses. If you are experiencing problems with normal or telephoto lenses, especially close-up, it is very likely the result of a rangefinder adjustment issue. The Hexar RF has exactly the same type of adjustments (1m to infinity) that Leicas have. As discussed above, you should also remember that close=inaccurate with any rangefinder camera.

These are my results to date. MNT = mounts; FCS = focus; CPS = collapse; MTR = metering. Generally, I use a 15x loupe and a negative scanner (3600 dpi) to look at the negs to verify focus.

These are all the lenses that I could count from my records. Most lenses are well, unremarkable in use, which is good. If you have experience with other lenses, let me know and I will put your comments in.

21mm Kobalux
3rd generation
        4 samples tested
28mm Kobalux
3rd generation
        3 samples tested
35mm f/1.4 ASPH Summicron-M          
35mm f/1.8 Canon         2 samples tested
35mm f/2 Canon         2 samples tested
35mm f/2 4th
        2 samples tested; one of the two uncoupled from rangefinder closer than 0.7m.
35mm f/2L
        Konica version of above; made for Leica bodies.
35mm f/2.8
        Can change the metering pattern subtly; interesting that the shutter chamber appears to have been fitted precisely to the rear element of this lens. (3 samples tested)
50mm f/1.2
        This lens requires a collimation check when it gets a CLA (2 samples successfully tested).
50mm f/1.4
SC Nikkor
        Uncouples from rangefinder closer than 0.9m. There is a detent in the lens to remind you.
50mm f/1.5 Canon/Serenar         Focusing cams on these lenses may not always be perfect at 1m (2 samples).
50mm f/1.5
        Close focus error and overall performance on this lens makes it difficult to assess, but a good J-3 will work properly (8 samples tested).
50mm f/1.8 Canon          
50mm f/2.0
        Uncouples from rangefinder closer than 0.7m
50mm f/2.0 Summicron C         Collapses into body. Just don't jam it in.
50mm f/2.0
        Well, it had better work! Also works on M3.
50mm f/2.0
HC Nikkor
         Uncouples from rangefinder closer than 0.9m. There is a detent in the lens to remind you.
50mm f/2
50mm f/3.5 Industar         My sample of the Industar hung up on the frame preview lever. Collapses into body. Just don't jam it in. Stops very close to the shutter assembly.
85mm f/2
PC Nikkor
85mm f/2
        The lens does not have very good close focus capabilities (7 samples tested). Black lenses tend not to focus properly close-up on RFs or M3s.
90mm f/2.0 Summicron          
90mm f/2.8
90mm f/2.8
         Also works on M3.
105mm f/2.5
PC Nikkor
        The tripod socket on this lens is dead level with the bottom of the Hexar RF. Hmmm. It doesn't line up on Leicas...
135mm f/3.5
        Some variation in Canon lens cam grinding but works properly (2 samples tested).
135mm f/3.5
QC Nikkor

Others have been kind enough to report the following (I am still compiling what people have sent in so far, and when I get a minute, I will make a giant table):

25/4 Snapshot
        This lens lacks a rangefinder coupling. Thanks to Nick Merritt
28mm f/2.8
        Thanks to Jeffery Smith; lens also works on Bessa-R2 (thanks to Scott Stewart).
35mm f/1.7 Ultron ASPH         Thanks to Joop Mes
35mm f/2
        Thanks to Jeffery Smith
35/2.5 Skopar
        Thanks to Nick Merritt
50mm f/1.2
        Thanks to Jeffery Smith
50mm f/1.4
        Thanks to Scott Stewart
50mm f/2
        Thanks to Jeffery Smith; lens also works on Bessa-R2 (thanks to Scott Stewart).
90mm f/2.8
        Thanks to Jeffery Smith; lens also works on Bessa-R2 (thanks to Scott Stewart).

90/2.8 "Skinny"

        Thanks to Nick Merritt
135/2.8 Tele Elmarit-M         Bugeyes darken finder corners, but otherwise ok (thanks to Francois P. Weill).
135/4 Tele-Elmar         Thanks to Francois P. Weill
        Thanks to Nick Merritt
28-35-50 2nd
        Thanks to Jeffery Smith

Features and benefits vis-à-vis Leica. I have been able to identify the following functional aspects that make the Hexar more useful to me than a traditional M.

Motor drive. I am not as interested in keeping the film moving as I am in keeping my right thumb out of my right eye when I wind the film. Left-eyed shooters rejoice!

Multicoated finder. How did we live without this until the M7? No rangefinder flare.

Electromagnetic shutter release. I find that I move the camera less shooting with a short-travel switch than a long-travel mechanical relase.

Lower moving mass shutter. The Hexar RF has palpably less vibration from the shutter than the M does. There is some vibration from the motor drive after the shot.

Higher top speed. Comments to the effect that a 1/4000 sec shutter speed is meaningless are to me suspect. With 100-speed film, a 1/4000 sec speed means the difference between shooting at f/5.6 and f/2.8 in bright sunlight. Many Leica rangefinder lenses are designed to be best under f/5.6, and having two extra stops means that you can minimize depth of field with your teles. You can, of course, buy ND filters for use with Leica bodies, but it is something of a clumsy workaround. Additionally, a 1/125 second shutter synch also has, in my experience, less possibility of ghosting.

Scale exposure readout. While it is nice that the Leica M7 finally has a shutter-speed readout, it does not tell you how far you are off the meter reading in manual mode, nor does it help you understand the contrast range of a scene (unless you can average fractional shutter speeds in your head). Being able to see a scale of shutter speeds helps you understand how many stops away the brightest object in a scene is from the dimmest. This is something that arrows simply cannot do.

Places where the M3 (or even M7) is still better. There are some places where the 50-year-old M3 is still better than a Hexar RF:

Noise. While the Hexar RF does not attract much attention, even indoors, the M3 is still less obtrusive noisewise. It's not a huge issue, but it is there. But the Leica M3 is more visually obtrusive unless you have a black one.

Ease of telephoto focusing. One place where the M3 has never been beaten is in focusing 90 or 135mm lenses. It is not impossible to focus these on a Hexar RF, but it is not as much fun. The Hexar RF can sometimes be difficult to focus with slow-twist telephotos, because you agonize about lining up the image perfectly. The M3 does not put you in that position, because the RF spot is so big and clear you can't miss the mark (unless your rangefinder is maladjusted). The M6/M7 0.85x finder is a good substitute, but not the same thing.

Film flatness. While the Hexar RF can blast out the shots two at a time, what suffers is the film flatness. The Leica Ms have a different rail arrangement and bigger pressure plates that put more force on the film. The Hexar, with its need for more friction-free motor-driven operation, needs you to slow down to one frame per second with fast, wide lenses so the film can settle down.

And so... On balance for most people, I think that the Hexar RF is a pretty good buy, and being only $500 more for the camera, lens and flash than the lower-end Cosina Bessa-R2 body (alone), it is pretty hard to turn down. It is also hard to turn down when you stop to think that an M7 with a (bulky) motor drive, a 50mm Summicron lens and a Leica flash costs about (adding this up...) $2,400 more.

There will always be a place for my M3, but if anyone in Tokyo is listening, where's that 0.9x Hexar?!