Will the M9 encourage or discourage Zeiss to make a Digital Ikon?

[Roger Hicks]

Dear Frankie,

How did film get into it?

Saying 'hardware loses every time' is patently nonsense, or there would never be any advances in hardware.

And -- this occurred to me last night when I was falling asleep -- what are the theoretical or practical differences between 8-bit coding with two bits left blank, and 6-bit coding?

Finally, I don't know about Leica, but I do know that at many companies, patrolling/scanning the internet forums is regarded as a kind of penance, and that others simply don't bother because it's more trouble than it's worth.

Cheers,

R.

 

[Quercus]

The bottom line is that CZ do not make cameras any more - so regardless of what a very small proportion of the camera buying market may or may not want CZ will be dependent upon what their partners can offer. Also unlike say soligor or even tamron CZ are not relying upon the photo market for profitability.

that means their possible entry to the d-body market is tied directly to either sony or cosina: Sony have now have a proven record with high end dslr platforms. cosina have a track record in limited build retro designs (even the zm is essentially a restyled CT1 and lovely with it)

To my mind sony are the most likely to be able to offer a platform that could become a rangefinder like d body maybe using 4/3 but probably badged sony using zeiss glass. Could it be an ikon well that will depend on the deal with cosina and the name license - I am not sure if cosina build anything for sony but they did for minolta so maybe there is a way of making a d-ikon happen

does it matter - not really - certainly not to cz - they have become very good at glazing other peoples bodies - sony, cosina (including zm,ze,zk and zf), nokia, plus of course the mediums that it supplies

Possibly the best hope for a d-rf-ish body is the fact that sony do not have anything as an answer to the panasonic compactish stable - now that is a bit of Japanese market share that sony do not have a toe hold in when a major competitor does. Something that is not very the "sony way"

 

[ZeissFan]

In the end, Carl Zeiss will be intimately involved in a digital rangefinder.

Zeiss Ikon has been gone for nearly four decades, and it's doubtful that there would be any camera fabrication around today. And even if there was, it's clear that Zeiss has no interest in restarting those operations.

Cosina has a good track record of building rangefinder cameras but obviously hasn't delved into the digital realm. Sony has experience in electronics, SLRs, point and shoots and video cameras but not rangefinders. Some training would be required, as well as testing and quality assurance methods.

Hypothetically, if Cosina were to make the shell and Sony were to handle the electronics and final assembly, that involves two subcontractors. That pushes up the price, so perhaps my $4,000 price point that I mentioned in another thread is overly optimistic. Both subcontractors would be expected to profit from the venture -- they're not in it for the good of the craft.

In any case, we'll see what happens next. Or not. In the end, it will be Zeiss' decision to move forward on this and not Cosina's and not Sony's. They have no more right to build a product and slap the Carl Zeiss name on it than you or I would have.

 

[Frankie]

Dear Frankie,

How did film get into it?

Saying 'hardware loses every time' is patently nonsense, or there would never be any advances in hardware.

And -- this occurred to me last night when I was falling asleep -- what are the theoretical or practical differences between 8-bit coding with two bits left blank, and 6-bit coding?

Finally, I don't know about Leica, but I do know that at many companies, patrolling/scanning the internet forums is regarded as a kind of penance, and that others simply don't bother because it's more trouble than it's worth.

Cheers,

R.

Context is the key concept here.

DRF happened within the mega-trend of consumers dead-ending film. [Here I used the "F" word again, may I?]

Hardware solution loses every time [within the context of a competition in the digital domain]. Ever heard of a motorized abacus?

Leica will not be able to take advantage of the current Kodak 6 micron anti-blooming CCD technology because its M8/9 hardware micro lens array was engineered for older generation 6.8 micron pixel spacing...ergo, no FF 24 M-pixel M10 either, which also requires 6 micron CCD.

The Leica self-celebrated 6-bit coding is merely marketing. No one [or should I say few before another pedantic argument ensues] writes 6-bit software within 8-bit word structure. Wasting 2-bits worth of space is far simpler...and cost nothing in price or performance. The fact that only 6 B/W bars appear on the lens mount does not mean the firmware [software stored in read-only memory] must also be 6-bit. What do you care if the firmware files are bigger?

[What do you think the firmware does when legacy lenses are mounted? Would no-bar detected causes the firmware to select "no post-processing"? If so, where did that bit of decision come from?]

Penance might happen when one's bad decision causes others to suffer [job loss]. I had faced such Zeiss management in 1995 after digital [colloquially softcopy] photogrammetry wiped out a century of Zeiss hardware excellence in merely 3 years...and Zeiss had written the original book.

 

[Roger Hicks]

Context is the key concept here.

DRF happened within the mega-trend of consumers dead-ending film. [Here I used the "F" word again, may I?]

Hardware solution loses every time [within the context of a competition in the digital domain]. Ever heard of a motorized abacus?

Leica will not be able to take advantage of the current Kodak 6 micron anti-blooming CCD technology because its M8/9 hardware micro lens array was engineered for older generation 6.8 micron pixel spacing...ergo, no FF 24 M-pixel M10 either, which also requires 6 micron CCD.

The Leica self-celebrated 6-bit coding is merely marketing. No one [or should I say few before another pedantic argument ensues] writes 6-bit software within 8-bit word structure. Wasting 2-bits worth of space is far simpler...and cost nothing in price or performance. The fact that only 6 B/W bars appear on the lens mount does not mean the firmware [software stored in read-only memory] must also be 6-bit. What do you care if the firmware files are bigger?

[What do you think the firmware does when legacy lenses are mounted? Would no-bar detected causes the firmware to select "no post-processing"? If so, where did that bit of decision come from?]

Penance might happen when one's bad decision causes others to suffer [job loss]. I had faced such Zeiss management in 1995 after digital [colloquially softcopy] photogrammetry wiped out a century of Zeiss hardware excellence in merely 3 years...and Zeiss had written the original book.

Well, quite. Are you sure that the internal software is in fact 6-bit?

As for your comment about a motorized abacus, that was precisely my point too. Hardware evolves.

Cheers,

R.

 

[Roger Hicks]

In the end, Carl Zeiss will be intimately involved in a digital rangefinder.

Why?

Cheers,

R.

 

[Frankie]

Well, quite. Are you sure that the internal software is in fact 6-bit?

As for your comment about a motorized abacus, that was precisely my point too. Hardware evolves.

Cheers,

R.

I actually don't care whether Leica uses 6-bits or whatever. Modern software are mostly 32-bits [4 x 8-bits]. 6-bits is not cause for celebration.

Hardware evolves, yes. Yet no one attempts the ridiculous. Imagine the complexity of a motorized abacus, versus a calculator we now deemed disposable.

 

[Roger Hicks]

I actually don't care whether Leica uses 6-bits or whatever. Modern software are mostly 32-bits [4 x 8-bits]. 6-bits is not cause for celebration.

Hardware evolves, yes. Yet no one attempts the ridiculous. Imagine the complexity of a motorized abacus, versus a calculator we now deemed disposable.

Dear Frankie,

First, you set up a straw man my (deliberately?) misunderstanding six-bit coding.It's a way of telling the camera what the lens is and 6 bits is all you need. Then you say you don't care.

Then you set up an even sillier straw man with your motorized abacus.

What IS your argument?

Cheers,

R.

 

[Frankie]

Dear Frankie,

First, you set up a straw man my (deliberately?) misunderstanding six-bit coding.It's a way of telling the camera what the lens is and 6 bits is all you need. Then you say you don't care.

Then you set up an even sillier straw man with your motorized abacus.

What IS your argument?

Cheers,

R.

I have no argument and no interest in Leica’s usage of a 6-bit bar-code [for conveying focal lengths, apertures and formulation of their “lenses between 21mm to 90mm focal lengths made since 1954”]. 64 types are more than enough.

However, the software must handle up to 16-bit image data, hence my remark “modern software is mostly 32-bits”…[while the start command could be a 6-bit or even a 1-bit “word”.]

The 6-bit patent however, is simply a way of garnering the lens after-market for another period…now that the M-mount is no longer exclusive. Selling a couple of new lenses to every M8/9 owner is practically a doubling of their gross sales, and is important to Leica now. Leica simply cannot adopt supermarket type bar-code that anyone can use no matter what…the development is not driven by “leadership in innovation”.

The Leica’s I know (from Leica Cameras AG, to Leica Microsystems, to Leica Geosystems) have always been firmly entrenched in hardware. Their occasional and inevitable software foray is the butt of jokes in my industry.

[The M8/9 firmware is said to have been sourced from JenOptik, formerly Zeiss Jena of the defunct DDR, and formerly head office of Carl Zeiss…as provided in the will of Ernst Abbe…until the forced relocation to Oberkocken after WWII… Zeiss Jena software (1945~1991) was an even bigger butt of our jokes.]

Within the context of a DRF, my bet is on the software and CCD development side. Nikon D3S (chip said to have been sourced from Sony) now supports an un-boosted ISO of 200~12,800, a 2-stop advantage over the M9…enough headroom for post-processing without the aid of the fixed 6.8 micron spaced special hardware micro lens array that got so much ink.

As far as the software “knowing” what focal length/aperture in use is concern, “auto-detect” and “auto-process” is long standard operations in my field…software again [a really long lecture].

Hence “I don’t care”…meaning “I am not impressed”…no matter how some might celebrate the “six B/W dots”


The motorized abacus is indeed the silliest example of hardware development I can think of. An abacus is simply a very primitive adding/subtracting machine…and can be made to do simple multiplication [by shifting decimal place plus adding]. It cannot do much beyond that…so why bother?

Hardware evolves...but not the rediculous.

My sentiments also apply to many RF components. See: http://www.rangefinderforum.com/forums/showpost.php?p=1211058&postcount=21

 

[Pavel+]

I think that what Frankie was saying in the first place was that a motorized abacus WOULD be a nutty, pointless excercise. It fits his argument well. Mechanical, as much as we may like it will get washed away, slowly or quickly but surely - because continuing with mechanical excellence, no matter what its former glory or current virtue - IS making the abacus motorized in the face of digital.

The Leica was a marvelous abacus once. What do they do now? Traditionalists would jump off cliff if the new digital didn't closely follow the old abacus form. But how to modernize it and still keep the virtues (and the fans)

Don't get me wrong ... I am faster and faster starting to see that life may have been better in the abacus, err, I mean film world when cameras were solid, simple and tactile. But still, that does not change the challenges. Six bit coding? Dumb. Why did they not drill one hole instead of six and put in micro bar code sensor? Because Leica seems to be with one and three quarters feet in the mechanical world and sometimes that causes them to shoot a few toes off.

 

[Roger Hicks]

I think that what Frankie was saying in the first place was that a motorized abacus WOULD be a nutty, pointless excercise. It fits his argument well. Mechanical, as much as we may like it will get washed away, slowly or quickly but surely - because continuing with mechanical excellence, no matter what its former glory or current virtue - IS making the abacus motorized in the face of digital.

The Leica was a marvelous abacus once. What do they do now? Traditionalists would jump off cliff if the new digital didn't closely follow the old abacus form. But how to modernize it and still keep the virtues (and the fans)

Don't get me wrong ... I am faster and faster starting to see that life may have been better in the abacus, err, I mean film world when cameras were solid, simple and tactile. But still, that does not change the challenges. Six bit coding? Dumb. Why did they not drill one hole instead of six and put in micro bar code sensor? Because Leica seems to be with one and three quarters feet in the mechanical world and sometimes that causes them to shoot a few toes off.

Actually, a motorized abacus is a completely pointless example, because it would slow the abacus down instead of speeding it up, AND be less reliable.

Answer to the bit in bold: 6-bit coding is actually a damn' sight cleverer than micro bar coding because it's bigger, cruder, and a lot less likely to go wrong with a small bit of dirt.

Cheers,

R.

 

[Pavel+]

Roger, point taken on what you have just said. But you have to understand where I'm coming from. I AGREE that the six point is more what you say - cruder and thus likely tougher. And I LIKE that. But that is not the digital paradigm. The rest of the world is going for slick, trasparent, computerized, efficient and - disposable. I'm not a huge fan of it in some way and a big fan in others.

But you are relating to this as a true Leica-phyle. Perhaps you and we should leave digital alone altogether. THis crude semi-mechanical way is not trully embracing digitals mind set. THis six hole thing also seems to be done with proprietary intentions. Just enough combos to have the Leica stuff covered and no room for the other brands. That in the niche that RF are today may be a shortsighted business plan.

 

[Pavel+]

On second thought ... are there not threads everywhere where the six hole fix is fussy? Or is that only with do it at home fixes?

I doubt that a micro scanner would be anymore unreliable as physical holes. Wipe it and go. We trust cloth shutter curtains and dont' mind that they are not exact. But you don't seem to embrace the non-mechanical. It reminds me of how the first bodies that needed batteries were received. Today we don't care. Photographers have moved past old precepts and trust electrons. (for better and worse)

 

[Roger Hicks]

Roger, point taken on what you have just said. But you have to understand where I'm coming from. I AGREE that the six point is more what you say - cruder and thus likely tougher. And I LIKE that. But that is not the digital paradigm. The rest of the world is going for slick, trasparent, computerized, efficient and - disposable. I'm not a huge fan of it in some way and a big fan in others.

Dear Pavel,

Why does it have to be like that? Who says? Why can't we have the best of both worlds?

When I switched from mechanical to electric typewriters, my first was an Olivetti golfball. It was rubbish, and wore out very fast: a year or two. I replaced it with an American electric typewriter, similar in price, which was bigger, cruder, uglier, and non-golfball. It was still working when I gave it away years later.

And on your second post: I think the 'fussy' complaints apply only to home bodges.

Cheers,

R.

 

[Pavel+]

That is a funny analogy with typewriters. I've been moving and one of the things that I have with me is a 1950 something typewriter. Still works except for the ribbon (can you still get those I wonder) and my 9 year old daughter is fascinated by it, almost as if there were bond.
I'm one that wants the best of both worlds. Perhaps in some way even more that we are getting in the current incarnations of the M8 and M9. All I'm saying however is that it is not Leica being effecient and in the spirit of Digital. Good. Unless that shrinks the market rather than grows it. Unless that costs them margins rather than grows them.

What exactly is wrong with the Zeiss Ikon in Film and the M6? Why do we want to go digitize it? And if we do ... why half-way. It is not fish nor fowl that way. Sentiments aside.

 

[Frankie]

Actually, a motorized abacus is a completely pointless example, because it would slow the abacus down instead of speeding it up, AND be less reliable.

Answer to the bit in bold: 6-bit coding is actually a damn' sight cleverer than micro bar coding because it's bigger, cruder, and a lot less likely to go wrong with a small bit of dirt.

Cheers,

R.

Abacus usage is not about speed, thus no point in improving that, making it less reliable, cost more and be the new butt of jokes...

The simplest way to encode a lens is to mill notches in the mount claw rim. The claw rim does nothing anyway, and a few small round notches won't weaken the strength much. Like a key, it is all hardware, will take a long time to wear out and only usable in one lock... But others could duplicate it.

 

[Frankie]

A new page

A new page

Conceptually, a digital camera starts with a lens projecting an image on a CCD instead of film.

Instead of wet film processing, the sensor data is computer processed…whether the computer is on-board or external. The circuit board associated with the CCD is a [dedicated] computer.

Of course, the computer has to be powered and a device has to be attached to record the data. May the best battery, memory card or whatever is next wins. Samsung, Sandisk and many others will help.

A camera needs to be focused, and we have accepted manual. Ground glass is a way, rangefinder another, EVIL a third… The best is the type that let you see what the sensor sees…before, during and after the [exposure] fact. RF is the worst of the three, it does not even see what the taking lens sees. [However, we are talking about a dream RF here, so please don’t start…]

A shutter is a device that limits the period of exposure. The on/off switching of CCD achieves the same. Even Leica Geosystems (now renamed ERDAS) offers a shutter-less solid state aerial scanner that uses this technique…since 2001. Also the Rollei Electronic...no cocking needed and silent.

The more accurate timing device is not a mechanical watch movement but rather a quartz chip set. That debate was settled 40 years ago…even the Zurich Hauptbahnhof had sported a Citizen clock [in 1991].

Evenness in illumination is a universal problem but deemed worst in a dRF…hence the self-celebrated Leica 6-bit and offset micro lens…

However, there are plenty of credible reports that the 6-bit coding means more in alerting the on-board computer that a [less than perfect Leica-made] super-wide angle lens is mounted…so fix the collective but embarrassing and immediately observable chromatic aberrations first.

Lens mechanical and optical vignetting is common also to film cameras. A tried-and-true fix was anti-vignetting filter...on the lens

Pixel vignetting compounds the problem. The solution is to use a CCD with less hooding, such as the Sony HAD CCD-II…

In a dRF, this anti-pixel-vignetting filter can be simply coated also onto the micro lens substrate…just like that infamous M8 IR coating...so long as the CCD sensitivity can handle all that filtering with a usable range left over [same problem as was in film].

Some CCD now sports ISO 200~12,800 [such as the purportedly Sony-made chip used in the Nikon D3S]; wasting 2 stops still give you 200~3,200, matching Leica.

So Carl, talk to Sony…your business partner contract with them has another 4 years and 9 months remaining.

Considering product pricing:

Leica had set a bench mark of ~$7,000…2X the price of an M7, or 2X a bread-and-butter lens [I don’t care you had found a better deal on eBay].

The Zeiss Ikon ZM price settled at ~$1,500, and a bread-and-butter lens ~$1,000…but made in Japan.

Will that proven contractor Cosina accept a Manufacturing Order to build a ZMd targeting a MSRP of 2X ZM plus...after Zeiss had done the home work?

 

[Pavel+]

Wil cosina accept? I sure hope so Frankie! Great information.

So the basic question after "if" is "when"?

 

[Frankie]

Conceptually, a digital camera starts with a lens projecting an image on a CCD instead of film.

Instead of wet film processing, the sensor data is computer processed…whether the computer is on-board or external. The circuit board associated with the CCD is a [dedicated] computer.

Of course, the computer has to be powered and a device has to be attached to record the data. May the best battery, memory card or whatever is next wins. Samsung, Sandisk and many others will help.

A camera needs to be focused, and we have accepted manual. Ground glass is a way, rangefinder another, EVIL a third… The best is the type that let you see what the sensor sees…before, during and after the [exposure] fact. RF is the worst of the three, it does not even see what the taking lens sees. [However, we are talking about a dream RF here, so please don’t start…]

A shutter is a device that limits the period of exposure. The on/off switching of CCD achieves the same. Even Leica Geosystems (now renamed ERDAS) offers a shutter-less solid state aerial scanner that uses this technique…since 2001. Also the Rollei Electronic...no cocking needed and silent.

The more accurate timing device is not a mechanical watch movement but rather a quartz chip set. That debate was settled 40 years ago…even the Zurich Hauptbahnhof had sported a Citizen clock [in 1991].

Evenness in illumination is a universal problem but deemed worst in a dRF…hence the self-celebrated Leica 6-bit and offset micro lens…

However, there are plenty of credible reports that the 6-bit coding means more in alerting the on-board computer that a [less than perfect Leica-made] super-wide angle lens is mounted…so fix the collective but embarrassing and immediately observable chromatic aberrations first.

Lens mechanical and optical vignetting is common also to film cameras. A tried-and-true fix was anti-vignetting filter...on the lens

Pixel vignetting compounds the problem. The solution is to use a CCD with less hooding, such as the Sony HAD CCD-II…

In a dRF, this anti-pixel-vignetting filter can be simply coated also onto the micro lens substrate…just like that infamous M8 IR coating...so long as the CCD sensitivity can handle all that filtering with a usable range left over [same problem as was in film].

Some CCD now sports ISO 200~12,800 [such as the purportedly Sony-made chip used in the Nikon D3S]; wasting 2 stops still give you 200~3,200, matching Leica.

So Carl, talk to Sony…your business partner contract with them has another 4 years and 9 months remaining.

Considering product pricing:

Leica had set a bench mark of ~$7,000…2X the price of an M7, or 2X a bread-and-butter lens [I don’t care you had found a better deal on eBay].

The Zeiss Ikon ZM price settled at ~$1,500, and a bread-and-butter lens ~$1,000…but made in Japan.

Will that proven contractor Cosina accept a Manufacturing Order to build a ZMd targeting a MSRP of 2X ZM plus...after Zeiss had done the home work?

One of the components that I always forgot to inventorize [and immediately jumped on by many] was the LCD screen...mostly because I consider that component in a different light. In any case, LCD's are widely available in many sizes and resolution and the new OLED are so thin...and would be my choice.

An LCD adds much thickness of a camera [cutaway photos long published, from D2/3 to R-D1 clearly indicate that], adding much bulk but for what?

[Measured at the top plate, the ZM thickness is ~33mm, an M4/6/7 ~35+mm. The M8/9 got fat and ugly at ~39+mm, while the Epson even fatter at 42+mm had some of the thickness not shifted to front via a ~3mm protruding lens mount and disguised via the outrigger-style package in the back.]

Remember the M-mount flange to film plane depth is only 28mm? [OK, 27.95mm if you insist to be exact. ]

A dRF...so far...offers no live view, which an LCD would be important. Operating info on battery/exposure remaining, Raw/JPEG, white balance can be better accessed through direct dials R-D1 style. [I purchased a used R-D1 recently and had experienced that first hand.]

Setting up the camera is easy and won't be much needed during a shoot. Personally, I set the camera into using RAW+daylight and go...just like installing a roll of film. [When daylight become evening, I prefer to eat and drink instead. ]

Histogram chimping after the fact is something I don't do...I prefer bracketing to better insure I had captured the moment. Film/processing has been prepaid.

The auto-bracketing feature on the M9 is a move in the correct direction...but make sure the buffer is large enough and firmware fast enough. [Are you listening Carl?]

If all one wants to do is to review images after the action or entertain the kids, an iPhone App would be ideal. I have looked into it and have determined it can be done...essentially reading the in-camera memory chip as its own. AND, Apple will work hard upgrading the hardware in case you need it...for free.

 

[Frankie]

Wil cosina accept? I sure hope so Frankie! Great information.

So the basic question after "if" is "when"?

Thanks Pavel+ for your support.

My post is about what I would have done if chairing the design committee...I have chaired a few in my professional career, often at odds with reactionary engineers. The engineers who desire continuing employment will then do homework and report back...and then a second meeting where delightful alternative solutions were often proposed.

Manufacturing "hard cost" is third-party components...CCD, LCD; "soft cost" are self-made or outsourced components. The per unit cost will be known only after market assessment is done.

[Epson had initially announced 10,000 R-D1 would be produced...remember? Elsewhere in this RFF, an inventory of camera serial numbers is taking place. My guess is: the RD1x was announced to make up the 10,000 unit manufacturing order as negotiated with Kobayashi-San.]

Manufacturing cost...perhaps become known after a few Sake bottles with Kobayashi-San? Carl Zeiss grossed >$3 billion in '09, suffering only a small loss and able to guarantee employees another year of no lay-off. They can afford a few bottles.

As far as I am concerned, "if" is no longer a question. "When" would be after Carl had decided "in or out". Announcement and mock-up day (Olympus style) has to be Photokina 2010...or way too late.