Manifesto: Time to Kiss the M Mount Goodbye? [long!]

[jlw]

I don't have much intelligent to add to this thread right now, but I found it interesting, and fun to read thus far. So, my 2 cents:

I find one solution proposed by jlw - to have a servomotor-based rangefinder - very interesting; this might even help to solve the problem of misaligned rangefinders. It would even be relatively easy to make this type of system self-healing. Imagine this: mount camera on tripod, set lens to infinity, and then press a button until the rangefinder patch lines itself up correctly. Repeat for 1m, and you're set. Have another button to adjust the vertical, and you're laughing. I would welcome this, as I have two Bessa R that currently need slight adjustment in both directions!

Yup, that type of thing should be feasible. More to the point for manufacturers, it should be possible to do the alignment settings robotically, rather than requiring each camera to be hand-calibrated by a technician.

About replacing the lens mount, I don't know what to say... maybe it's because I'm not sure I see the need to change.

For film-camera use, I agree there's no need to change. For digital-camera use, I think the M mount's short backspacing is going to be a perpetual roadblock, because the lens' "chief ray angle" (angle at which rays from the lens reach the sensor) is always going to be steeper than optimum.

Leica has pretty much conceded this in their approach with the M8 to vignetting, antialiasing, and infrared contamination. They've addressed the vignetting issue with their offset microlenses, but they're on record as saying that a truly effective IR-cut filter behind the lens would have had to be so thick that it would have caused color fringing by displacing the steeply angled rays; instead, they want you to install the IR-cut filter over the lens when needed. They've also said that's one reason why they didn't fit an anti-alias filter to the M8. M8 enthusiasts are willing to accept all this, but there's no denying that it makes things more complicated than with DSLRs that can produce excellent results without resorting to these workarounds -- simply because the lens is farther from the sensor.

Personally, I think that if a new system arrives, it must include compatibility for M (and thus, LTM) lenses if it wishes to succeed with the current customer base. And personally, if it results in a significantly larger camera (not larger like an M5, but larger like a 5D), I'm not going to be interested.

Oddly enough, my train of thought got started on this when (in another thread) I imagined a fantasy "Canon 7sD" digital RF camera. My idea for this camera was that it would be about the size of the original Canon 7s (chunkier than a Leica, but not huge) and that it would contain Canon's EOS 5D sensor and electronics plus an electronically-coupled RF system.

The idea was that the basic body would have an M-mount that would accept current M lenses (using a distance encoder in the body to read out the RF coupling arm.) It also would also come with a "Lens Mount Converter D" (to adopt Canon's old terminology) similar to the ones they used to sell that allowed their SLR lenses to be mounted on their RF cameras. This converter would be of the right thickness to space EOS lenses correctly on the camera, and would contain all the electronic contacts needed to operate an EOS lens.

The idea was that they could add electronic distance encoders (this would be quite inexpensive) to a selected RF-friendly range of their EOS lenses -- say, the 28/1.8, 35/1.4, 50/1.2, 85/1.2, and 100/2. These lenses would still work normally as-is on Canon DSLRs (making it cost-effective to build them; they wouldn't need to make a separate range of limited-market RF lenses) but also would provide full electronic RF coupling on the "7sD" body through the lensmount converter.

How would you like them apples?!?!?

[Incidentally, in principle there's no reason the same couldn't be done with the Nikon lens mount, if it were restricted to their latest lenses with built-in focusing motors.]







From the responses to my earlier posts, I know some people would read this and say, "Why not just buy an EOS 5D instead?" My answer: "Because this would be a rangefinder camera! Rangefinder focusing and viewing just plain works better than SLR focusing and viewing for some types of photography!"

It continues to baffle me how many people here on RangeFinder Forum just don't seem to "get" that...

 

[Trius]

High ISO noise is irrelevant if you choose to make images that do not require high ISOs. Such images represent a significant portion of photographic endeavor. Low ISO photography appeals to many but it is not embraced by all.

And the same people who appreciate the necessity and logic of low ISO photography are those who will buy a dRF. I don't expect Joe Double-Double to appreciate the advantages, elegance and cost of the dRF we are envisioning here.

Otherwise high ISO (800 or greater) noise is a relevant issue. The typical dSLR user has a very slow kit zoom lens and a barely adequate pop-up flash. Consumer-level dSLR cameras employ heavy noise filtering for ISO 400. Digital filtering reduces resolution (always).

Again, those people will not be buying a high-end dRF. The very idea of a prime lens leaves them glazed over.

For a dRF to be successful, it has to produce color images that compare to ISO 800 color film. This is more difficult with APS and 4/3 sensors.

Yes, it is more difficult to achieve with APS and 4/3 sensors, but I don't agree that ISO 800 is the benchmark, nor do I agree that ISO 800 w/ noise alleviating post-processing wouldn't be sufficient. At the very least I expect at least some advances in those sensors that will get us to the point of such a camera being an professional tool.

B&W photography is another matter. Small (noisy) sensors can produce B&W images with the aesthetic of 35mm high-ISO film. The Ricoh GRD is interesting as it captures high-ISO B&W images that compare very well with high ISO B&W film. At the same time its low ISO color performance is very good. But its ISO 400 color images are not competitive.

Good points. I suspect its sensor is < 4/3s, though, so in this context the comparison isn't quite there.

Another real challenge with a 4/3s sensor is the manufacture of very wide angle lenses. The mount is designed for telecentric lenses, and the ZD 7-14 is by all reports quite good. But it is not inexpensive, and in the context of a dRF, we would expect primes ranging from 7mm to 14mm. Leica and Zeiss would be expected to set the pass for performance, with other manufacturers setting the pace for the very good but less expensive.

 

[StuartR]

Perhaps I am being unfair, but the idea of a 50mm lens with the depth of field of a 25mm lens, or a 100mm portrait lens with the depth of field of a 50mm lens just strikes me as "wrong". For my type of photography anyway...
That reason alone leaves me cold with respect to the 4/3rd's system. The 24x36mm sensor size strikes a really nice balance of allowing you great depth of field with wide angle lenses, but very short depth of field with longer lenses. It really strikes a sweet spot here, and that would be something I would be loathe to give up. For me, that is another reason why a pea-sized sensor in a point and shoot could never be a replacement for an RF or even something like a Leica CM or Contax T2. At least at f/2.8 or 2.4 and 35mm or 40mm you can have some background blur. Even wide open, the 5-15mm lenses on many point and shoots have incredible depth of field. So, if we ditch the M mount, can't we at least keep a big sensor?

 

[willie_901]

Stuart,

I agree totally with you about the DOF issue. The last time I bought this up (in a completely different context), there were many who though the crop factor was irrelevant. Most people just said "get over it", implying the convenience of digital was the more important than any other issue.

We may feel the crop factor is a show stopper, but I think we're in the minority.

willie

ps the photos in your web site are dynamite, wow!

 

[jlw]

way to make a point :/

Look, you want something that doesn't exist for a reason. Don't get pissy when people point out that the shortcomings of digital are becoming increasingly obvious, along with the strengths of film.

And the vast majority of buyers have already deserted film for digital for a reason. And the development of new film cameras has come to an almost complete halt for a reason. And the supply of film materials is shrinking, not growing, for a reason. Don't get pissy when I point out that the election is over and the marketplace has voted film out of office by a landslide.

If you guys want to kid yourselves that the befuddled masses will suddenly wake up to "the shortcomings of digital" and come groveling back to film, you're just doing the same thing Herbert Hoover did when he told us all that prosperity is just around the corner.

Personally, I think film photography has a long, long future ahead of it -- but its future is going to be merely an "alternative process" pursued by a relative handful of devoted enthusiasts, the way tintypes and albumen prints are now. People will still be doing beautiful work in it, but in terms of infrastructure support it's just going to be a backwater.

And if we continue in the mindset that the main appeal of rangefinder cameras is nostalgia, we're going to be a backwater within a backwater. Even Mr. Kobayashi isn't going to invest in supporting a shrinking market forever.

If we're going to be able to continue with the rangefinder concept -- which I still feel, based on what works in my own experience as a photographer, is highly relevant to today's photography, not just yesteryear's -- then we have to be open to ways of making that concept applicable to the types of technology that the industry uses today.



Yes, what I want doesn't (yet) exist for a reason. And I suspect that reason is that camera companies are perfectly capable of doing focus groups, that they've already researched this, and that they've already studied current rangefinder-camera users... i.e., us -- to see if we could be the opinion leaders, or viral-marketing "infectors," who could increase interest in rangefinder photography among the photography mainstream.

And they've concluded that we could NOT -- because most of us see a rangefinder camera NOT as a better way to make certain kinds of pictures, but as a relic of the past to which we cling to avoid confronting a present-day reality that displeases us.


Okay, fine. We all need our illusions. But... damn. I see a lot of DSLR shooters who I suspect would make better pictures if they had a camera that works like the camera they enjoy using now, but sees more directly, more clearly, and more responsively -- through a range/viewfinder instead of a mirror and focusing screen.

But they'll never get those cameras, and will never make those pictures, because WE -- the people who know the territory -- are saying, "No new things, thank you very much! We're quite happy living in our little imaginary 1954 time warp!"

 

[jlw]

Perhaps I am being unfair, but the idea of a 50mm lens with the depth of field of a 25mm lens, or a 100mm portrait lens with the depth of field of a 50mm lens just strikes me as "wrong".

A 50mm lens is a "normal" lens on a 35mm film camera. A 25mm is a "normal" lens on a 16mm motion-picture-film camera. A 100mm lens is a "normal" lens on a 6x7 film camera.

At comparable final image magnifications, they'll all produce roughly similar DOF effects at the same apertures.

So what is "wrong" here? Perhaps the prejudice that there is only one "right" format size, and that all lenses must be related to that size...?

 

[anselwannab]

A 50mm lens is a "normal" lens on a 35mm film camera. A 25mm is a "normal" lens on a 16mm motion-picture-film camera. A 100mm lens is a "normal" lens on a 6x7 film camera.

At comparable final image magnifications, they'll all produce roughly similar DOF effects at the same apertures.

So what is "wrong" here? Perhaps the prejudice that there is only one "right" format size, and that all lenses must be related to that size...?

I don't have the time to run the numbers, but I don't think that is right, holding the same f-stop value. If you take a normal 50mm at f2 in 35mm film, and go to a 4/3 sensor at 25mm, you are going to have to change the aperture to something smaller than f2 to keep the DOF the same. Reductio ad abusurdum, or near it, with P&S cameras with small sensors and really short normals, you have a really hard time getting DOF control. On the other hand 8x10 view cameras have shallow depth of field with their normals at f2 (ha, try to get that lens).

With 4/3 and APS-c cameras that has been my major complaint. Like when we went from medium format to 35mm, the lenses got shorter and got faster.

They are trying to get us to use 35mm lenses with smaller sensors. Theire should be newer, shorter, faster lenses. Smaller image circles should make this possible. And I think they should be cheaper, but it seems canon EF-s lenses, except for the kit lens, if anything have a premium on them.

To the people that want to use something the size of a Canon EF lens, have you ever held one? My 17-40L is bigger than my CL. Those barrels are huge! Now I am definately in the smaller, compact group of users, but you'd have to have the RF base so high to use these lenses it would make an M5 look small.

Mark

 

[Trius]

I have seen plenty of images from 4/3s and APS-C sensors with beautiful use of DOF. I think the issue is overblown, but I understand the concern.

BTW, the DOF on my 4x5 lenses is way different than those on my 35s. If I used the DOF tables from the Fujinons as my standard, I'd never use 35mm.

 

[anselwannab]

I think Olympus has it right with the E-1 lenses. They have a 35-100/2, a 150/2, a 90-250/2.8.

Plus if oyu have a faster f-stop, you let more light in, thus negating some of the issues with poorer ISO performance due to smaller photosites. If you give up one stop of noise from a APS to 4/3, maybe you can gain in back in having faster f-stop lenses.

Mark

 

[Hiyawaan]

If it ain't broke dont fix it!

 

[RML]

I'm with JLW on this. I love the idea of an affordable rangefinder by Canon or Nikon. Size does matter but when I compare my rangefinders with my old Eos 3000, the difference is minimal. So is weight. The only real difference is in the size of the lenses, and I reckon those could be reduced.

 

[canonetc]

I would love to see Canon or Nikon make a digital RF for the "public". Heck, take the old Canonet QL and slap a 10.2MP chip in that puppy and I'm there.

On the other hand, I feel the reason they won't enter the RF marketplace is their view on the "average consumer": the "average consumer" is LAZY. RangeFinders are HARD.

THis is why we have several hundred models of Point and Shoots, and less than 20 RF models (Leica, Contax, Konica, Voightlander, Plaubel, RD-1, Lumix (?), Fuji 6x9/6x7, Bronica, Mamiya 6/7 and the older Canon's, Nikons, Zeiss, and Zorki).

RF shooters are a specialized bunch. In truth, I think the "public" is too, but are too controlled by advertising and marketing to look beyond the well-known SLR & Point-and-Shoot world. Unless, of course, they meet an RF user.

Chris
canonetc

 

[Plasmat]

How about a DRF where you focus the M-mount lens to an approximate distance with a VERY easy to see distance indicator. Like a semi-transparent rectangle that turns clear when the lens is APPROXIMATELY in focus. (like a heads up display). You could rough focus VERY fast.

Then, a servomotor moves the whole FILM PLANE (sensor plane) a tiny bit until the focus is perfect, instantly, when you hit the shutter.

My other dream project is to produce a small, very high quality digital camera, with a big screen, crinkle finish paint, and a moving sensor plane as above, which uses C-mount lenses. Millions of these lenses are around cheap. Like Kinoptiks and Angenieuxs for $75.

I have a preliminary design with a Chinese manufacturer. I may even finance production of a couple of thousand first models myself.

 

[rolleistef]

The main pb with a "cheap" DRF is, that it's a niche market. It is an anti-consume society approach. Ok that's great but :
- With a digital SLR you get a correctly focused, correctly exposed picture, that more or less matches with what you framed in the VF.
A digital RF gives the image you saw in the frameline, and you need to focus afterwhile with a hard-to-use-for-a-basic-photographer system
- A digital SLR doesn't need any adaptation time to be used. They're just expensive P&Ss.
You need several weeks, if not months, to be able to take worth-printing pictures.
- People like close-ups, telephotos lenses and motorized camerasthat take 1 billion pictures a second.
with a RF you cannot take flowers, eyes or the toe of your b/f or g/f; nor can you take someone who's very far away because you need to get close and build a basic relation with the person you're takin in photo (blah!! 🙂), and taken 1 billion photos at a time is just nonsensical with a RF camera.

Sorry mates, but it's really unlikely that any major camera companies should once build a DRF. I would have loved to have a DQL17 or a DKiev, but it's not due to happen in that space-time dimension.
Keep using your RF camera though!

 

[rolleistef]

Plasmat, it would be great. I'd be the first buyer!
C-mount is for cinema isn't it?

 

[IGMeanwell]

You know at first ... I kind of dismissed this idea; but it does make sense to produce something like this however it seems like a movable sensor would end up costing more and would be more likely to have the potential to fail.

That being said, if you can find someone to develop this camera you have in mind, can produce it relatively inexpensively I think you would have a legion of people who could get behind such a camera. Especially if the amount of available lenses are wide spread.

To be honest there has been a plethora of good ideas in this thread that may or may not have already been explored by the various manufacturers. I just wish something tangible can come out of all this excellent free flowing of ideas...

How about a DRF where you focus the M-mount lens to an approximate distance with a VERY easy to see distance indicator. Like a semi-transparent rectangle that turns clear when the lens is APPROXIMATELY in focus. (like a heads up display). You could rough focus VERY fast.

Then, a servomotor moves the whole FILM PLANE (sensor plane) a tiny bit until the focus is perfect, instantly, when you hit the shutter.

My other dream project is to produce a small, very high quality digital camera, with a big screen, crinkle finish paint, and a moving sensor plane as above, which uses C-mount lenses. Millions of these lenses are around cheap. Like Kinoptiks and Angenieuxs for $75.

I have a preliminary design with a Chinese manufacturer. I may even finance production of a couple of thousand first models myself.

 

[Trius]

Sorry mates, but it's really unlikely that any major camera companies should once build a DRF. I would have loved to have a DQL17 or a DKiev, but it's not due to happen in that space-time dimension.
Keep using your RF camera though!

C'mon Stephane ... I know you have read all the Fed 8 threads here. The glorious workers will deliver any day soon now! 😀

Earl

 

[rolleistef]

Ahh right Trius. You found me. I'm still trying to imagine how it could be possible to turn a Fed into a DRf and there might be a quite simple solution.
Here it is :
Ingredients : a Fed 2 (20$+shipping)
a C-Mound to LTM adapter (15$)
a c-mount lens (zoom or not)
a cheap digital point and shoot, preferably in Fuji in a couple of years because of their excellent super-CCD sensor, for available photography.
Tools : Xacto cutter, watchmaker screwdrivers.

Dismantle like a mental your cheap DP&S, remove the front of your camera so as to get the bare sensor.
Cut a hole on the Fed back
Connect the two shutter releases
Set the correct working distance
mount the adapter and modify it so as to keep the RF working
Mount the C-mount lens (I guess a 12mm is a normal lens for that kind of sensor)
Set the shutter speed on T
Enjoy!
You can also use your DFed as a regular point and shoot, framing and focusing with the LCD screen.
Now some points I'd like to work out :
Is it possible to have the RF working by modifying the adapter?
Is there any special issue for the working distance?


Here is the first prototype of my Kiev with a Digital ModulR attached. I modified the modulR so as to fit the Kiev back and it turned my 50mm J8 into a nice protrait lens. It's called the DMKC, Digital Modular Kiev Camera. I had thought about F-CRAP, Film Camera Returns to Advanced Process but it didn't sound well at all.

cheers

 

[Trius]

Stephane: I see you have a brilliant career ahead both in design and marketing!

 

[rolleistef]

Really? Did I missed an opportunity by studying litterature then? 😀
But do you think it's concretely feasable? The major problem is the RF cam... it would to be modified a lot because not only the C to M39 adapter changes the lens mount, but also does it changes the working distance down to 17mm or so...