Robert White's Imprimatur

[willie_901]

More absurd M8 comments:

The M8 has the worst magenta color distortion per euro/dollar cost of any digital camera.

The M8 sensor has the best resolution per euro/dollar cost of any digital camera costing less than $7,000.

The M8 magenta issue highlights an often ignored fact: Bayer filter technology is a crude, incomplete first generation solution. As an aside, I learned a lot about digital imaging from the M8. I also saved a lot of $$$. I too say, "Thank you Leica."

Practically no one cares about the inadequacies of Bayer-based technology because the convenience of digital-image capture has more value than the image itself. Convenience is more valuable than resolution loss due to sensor-mounted IR filters. Convenience is more important than color fidelity. Convenience is more important than B&W image aesthetics.

Leica's M8 quality control for the initial batch of cameras was well-below what one would expect for a $5,000 camera.

There is nothing wrong with using filters as long as they don't have unintended consequences.

It is likely the unintended consequences of lens-mounted IR filters will be overcome by post=processing software, and this will be true for non-coded lenses as well.

Leica erred when they did not publicize the requirement for IR filters before the first cameras shipped.

Leica will eventually ship a M8 that is not flawed.

willie

 

[saxshooter]

Only negative report they have had concerns a shoot at Rememberance Day Parade London.
Guards uniform correct in scarlet red - trousers magenta [should be black].
Difficult to reshoot. Queen unavailable.

I gather the photographer will have the filter with him next year.

PRICELESS. 😀

Unfortunately, Leica does have a reputation to live up to. Words like solid and dependable come to mind. And the Leica M8 is a Leica product. Not a Leica/Panasonic product, or a Leica/Fuji product. Even though the sensor is Kodak made, the buck stops with Leica. When a pro buys a Leica it is expected to deliver the goods. Perhaps it was the false security by having something so M-like in the hand. I guess they were under pressure to get the product out, despite it not being "finished" yet.

 

[Simon Larbalestier]

Just spoken to them today.
M8 link restored.
Received 3 cameras so far.
40+ deposits taken. No cancellations.
.

Actually i cancelled my order about 10 days ago and was no 2 on the current list.
Purchased the Fuji TX2 off them instead. I've dealt with Robert since the late 80's and he has always offered 100% first class service.

 

[Gabriel M.A.]

Socke and Jaap: what are the German and Dutch equivalents for "oy vey"?

No amount of technical explanation is going to make the Nattering Nabobs of Leicabashism hear anything but blah blah blah blah blah. ::shrug::

I am upset at Leica about one thing: they shipped the thing to a lot of people who cried bloody murder at a problem which didn't have the technical competence to grasp and tackle.

 

[fgianni]

Even though the sensor is Kodak made, the buck stops with Leica.

Exactly, Leica choose the sensor, and if the technology behind it is incompatible with top quality IR free images in a RF camera, well there are other manufacturers and different technologies available, if a CCD does not do the trick, then try CMOS, or if it is the Bayer technology the culprit, try Foveon.

 

[Rhodie]

Actually i cancelled my order about 10 days ago and was no 2 on the current list.
Purchased the Fuji TX2 off them instead. I've dealt with Robert since the late 80's and he has always offered 100% first class service.

As both Robert & Liz were away, I spoke to Ken.
Their service is always impeccable. 🙂

But clearly many have a lot invested in the M8 - and its not just Leica.

 

[Gabriel M.A.]

Who wants to ride a ballon? We've got enough for a big big big basket.

yyeeeeeeeeee!

 

[visiondr]

It seems "all" we want is for someone to develop a sensor that ignores IR... that seems to be the best solution to all of this. There should be a way to develop a sensor that ignores wavelengths longer than around 750nm. That would solve the problem, end of story.

Ron

 

[fgianni]

What I don't understand is why Leica made the conscious decision of using insufficient IR filtration, that requires an additional IR filter on the lens.
If the IR filter in front is the correct engineering answer, while a filter on the sensor causes vignetting and softening of details, why not leave the hot mirror in front of the sensor completely out, thus nullifying all the possible negative effects of the IR filter on top of the sensor?

Using a filter that requires the addition of another one seems the silliest solution, either put in an adequate one, or leave it completely out.

 

[rvaubel]

What I don't understand is why Leica made the conscious decision of using insufficient IR filtration, that requires an additional IR filter on the lens.

Look, its pretty obvious that a solution that doesn't involve a front filter could have been implemented. My RD1 is much better corrected (to the point no one every noticed the magenta blacks) and yet its only advantage design wise was a 1.5 crop factor. Leica wasn't involved in a plot to screw the long time users. Maybe a better compromise could have been chosen, but thats water under the bridge now. The workabouts that are being developed, will work although its not clear yet what combination will be optimal. I am waiting for the dust to settle in this regard.

In any case, will waiting 98% of the pictures come out just fine.

Rex

 

[jaapv]

Exactly, Leica choose the sensor, and if the technology behind it is incompatible with top quality IR free images in a RF camera, well there are other manufacturers and different technologies available, if a CCD does not do the trick, then try CMOS, or if it is the Bayer technology the culprit, try Foveon.

Sorry to be the descending duck, close but no cigar... Indeed you are right that a sensor shoulld be developed that does not exhibit these problems, but all current sensor designs are equally sensitive to all light, including IR. The difference is only in the filtration. CMos, apart from being a Canon monopoly, requires heavy filtration to produce the images it does. The consequence is that Canon RAW files require more sharpening than other brands, to compensate for the loss of edge contrast. This is also the reason that they do not realize the theoretical advantage in resolution the 1DsII or 5D should have over any other camera in its class. Indeed they are industry standard, but other makes should not even come close, looking at just the numbers. In reality, the differences are marginal. The filtration renders them less suitable for rangefinders. The Foveon sensor, a very interesting concept, again, needs IR filtration. Because it is a layered sensor its narrow acceptance angle renders it unsuitable for rangefinders. New technology is needed. According to recent statements by ZI there is not even an indication that those technologies will be introduced in the forseeable future. So, for the time being, what we see is what we have.....

 

[jaapv]

Exactly, Leica choose the sensor, and if the technology behind it is incompatible with top quality IR free images in a RF camera, well there are other manufacturers and different technologies available, if a CCD does not do the trick, then try CMOS, or if it is the Bayer technology the culprit, try Foveon.

Sorry to be the descending duck, close but no cigar... Indeed you are right that a sensor should be developed that does not exhibit these problems, but all current sensor designs are equally sensitive to all light, including IR. The difference is only in the filtration. CMos, apart from being a Canon monopoly, requires heavy filtration to produce the images it does. The consequence is that Canon RAW files require more sharpening than other brands, to compensate for the loss of edge contrast. This is also the reason that they do not realize the theoretical advantage in resolution the 1DsII or 5D should have over any other camera in its class. Indeed they are industry standard, but other makes should not even come close, looking at just the numbers. In reality, the differences are marginal. The filtration renders them less suitable for rangefinders. The Foveon sensor, a very interesting concept, again, needs IR filtration. Because it is a layered sensor its narrow acceptance angle renders it unsuitable for rangefinders. New technology is needed. According to recent statements by ZI there is not even an indication that those technologies will be introduced in the forseeable future. So, for the time being, what we see is what we have.....

 

[Toby]

Sorry to be the descending duck, close but no cigar... Indeed you are right that a sensor shoulld be developed that does not exhibit these problems, but all current sensor designs are equally sensitive to all light, including IR. The difference is only in the filtration. CMos, apart from being a Canon monopoly, requires heavy filtration to produce the images it does. The consequence is that Canon RAW files require more sharpening than other brands, to compensate for the loss of edge contrast. This is also the reason that they do not realize the theoretical advantage in resolution the 1DsII or 5D should have over any other camera in its class. Indeed they are industry standard, but other makes should not even come close, looking at just the numbers. In reality, the differences are marginal. The filtration renders them less suitable for rangefinders. The Foveon sensor, a very interesting concept, again, needs IR filtration. Because it is a layered sensor its narrow acceptance angle renders it unsuitable for rangefinders. New technology is needed. According to recent statements by ZI there is not even an indication that those technologies will be introduced in the forseeable future. So, for the time being, what we see is what we have.....

I believe what you mean to say is unsuitable for current rangefinder lenses and mounts. There is nothing stopping leica or zeiss making a fully optimised digital RF, but they would have to dump the m mount. Canon's current advantage is based on their very hard nosed adoption of the EF mount in the late eighties. It is still the last fundamental change of mount by any mainstream manufacturer (at 20 years old!). Leica and nikon have both suffered by their refusal to turn their backs on obsolete lens mounts. A truly sucessful RF would almost certainly need to have a larger diameter mount with retrofocal WA lenses. But that would take a bucketload of money (Leica don't have it) and balls the size of melons (Leica definitely don't have them)

 

[jaapv]

I believe what you mean to say is unsuitable for current rangefinder lenses and mounts. There is nothing stopping leica or zeiss making a fully optimised digital RF, but they would have to dump the m mount. Canon's current advantage is based on their very hard nosed adoption of the EF mount in the late eighties. It is still the last fundamental change of mount by any mainstream manufacturer (at 20 years old!). Leica and nikon have both suffered by their refusal to turn their backs on obsolete lens mounts. A truly sucessful RF would almost certainly need to have a larger diameter mount with retrofocal WA lenses. But that would take a bucketload of money (Leica don't have it) and balls the size of melons (Leica definitely don't have them)

It has nothing to do with the mount at all. It has to do with the fact that the lenses of a RF can be and are much closer to the film/sensor plane. So if one would want to change that it would mean a completly new range of lenses and vastly thicker camera's....BTW most modern rangefinder WA lenses are of the retrofocus type.....

 

[Toby]

It has nothing to do with the mount at all. It has to do with the fact that the lenses of a RF can be and are much closer to the film/sensor plane. So if one would want to change that it would mean a completly new range of lenses and vastly thicker camera's....BTW most modern rangefinder WA lenses are of the retrofocus type.....

That's exactly what I mean, for maximum image quality you'd have to accept a thicker digi-rf (at the moment) and lenses that were further from the image plane and larger. The trade off if you don't want that is an IR filter. There is no free lunch, I just wish leica had been more upfront about it .. as Scotty said "Ye cannae change the laws of physics!"

 

[jaapv]

I would not have bought the camera you propose. Nor would many others.

 

[fgianni]

I would not have bought the camera you propose. Nor would many others.

But probably many more that did not buy it or sent theirs back would.

 

[fgianni]

The Foveon sensor, a very interesting concept, again, needs IR filtration. Because it is a layered sensor its narrow acceptance angle renders it unsuitable for rangefinders.

The foveon sensor has much bigger photosites than a bayer (4 times the surface per photosite given the same density), so the acceptace angle is actually wider, that's why the only company that has so far managed to make a compact digicam with an APS size sensor is using foveon technology (Sigma DP1)

And the foveon sensor does not need a CFA, this would have provided some slack to the Leica engineers for additional IR filtration.

And again, if everything fails, and the best engineering compromise is IR filtration on the front of the lens, then just leave the hot mirror out, it does not make sense to have an IR filter on the lens, and an inadequate hot mirror whose only effect is to reduce image quality, since it has nothing to filter, given that the IR has already been taken care of in front of the lens.

Whichever way you look at it they got it wrong, unless, like many Leica apologists, you think that there are only two way to do things: the Leica way and the wrong way.

 

[jaapv]

The foveon sensor has much bigger photosites than a bayer (4 times the surface per photosite given the same density), so the acceptace angle is actually wider, that's why the only company that has so far managed to make a compact digicam with an APS size sensor is using foveon technology (Sigma DP1)

And the foveon sensor does not need a CFA, this would have provided some slack to the Leica engineers for additional IR filtration.

And again, if everything fails, and the best engineering compromise is IR filtration on the front of the lens, then just leave the hot mirror out, it does not make sense to have an IR filter on the lens, and an inadequate hot mirror whose only effect is to reduce image quality, since it has nothing to filter, given that the IR has already been taken care of in front of the lens.

Whichever way you look at it they got it wrong, unless, like many Leica apologists, you think that there are only two way to do things: the Leica way and the wrong way.

Not an apologist- I would much prefer Leica to make a coherent statament than to speculate myself...But I go by results, and when I can get a camera, that according to the worlds leading print maker, only needs a 50$ filter stuck in the front of the lens to produce prints that are the best in the 135 class and are rivalled only by MF film (Davis Adamson), I, as a lowly amateur, am a happy customer, and consider my 4000 Euro well spent, considering that other top performers cost up to 10.000 Euro.. Having said that, my prime concern was getting a digital rangefinder that adressed most, if not all, the things that made me decide not to go with the RD1. In that this camera would have been fine already had it performed just like a 30D, which it in the event clearly surpasses by a fair margin. So call me an apologist, but in reality I am just a happy customer who gets slightly annoyed at " it could have been and should have been" type of posts.

 

[fgianni]

So call me an apologist, but in reality I am just a happy customer who gets slightly annoyed at " it could have been and should have been" type of posts.

I have already stated in another post that if I was not lucky enough to own a good RD-1 I would most likely gone with the M8, and I will buy it when my R-D1fails beyond repair, so I don't consider the camera a lemon.

When a camera costs $5000 and exhibits hard to explain unwanted features (that some would define as faults), I think "the should have been posts" are more than reasonable, people wants to understand why some compromises have been made and also like to know if really there were no better alternatives and why, sorry if this annoys you but I don't think it is reasonable to stop people asking questions.

It is also annoying that Leica refuses to answer questions like: if the filter in front is the right engineering answer, why putting a redundant and inadequate IR filter in the back, the experts tell us that an IR filter reduces image quality, so why not doing away with it and give us even better quality, expecially since with the proposed solution there is no IR light to filter sensor wise.

You are happy with the results, and I agree that they probably are good enough to justify the purchase, but good enough is not the same as as good as they could have got it, this sounds unreasonable to you? Well sorry but I see it differently.