400 Leica photographers agree: we love CCD!

[Jeff S]

Advertisement (gone after registration)

. It will be interesting to see how many people will defend the M240 as passionately as the M9 more than five years after its release and after a new model has been released.

 

And I wonder how many would have so passionately defended the M9 if Leica had released an 'improved' M9.2. The grass is often greener….short memories, this Leica crowd.

 

Jeff

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[zlatkob]

Christian Dowling's review of the M240 has a couple of side by side M9 vs. M240 image comparisons: Leica M (typ 240) Field Test and Review — Kristian Dowling. See the sections titled "CMOS Sensor" and "Image Quality". In particular, click on the photo of the dining room with the red table cloth for a full screen comparison.

[BerndReini]

There is already an MP version of the M240, and an M9.2 is exactly what the OP is asking for, yet everyone that switched to the M240 thinks he is crazy. There were people that preferred the M8.2 over the M9 and I admit that I missed the pixel level sharpness of the M8 with its weak IR filter, but the full-frame sensor outweighed its benefits for me personally.

 

I wish there was an M9.2 with the M240 shutter and LCD screen. I would keep the rest from the M9 and be a happy camper. I have enough batteries and I really don't get through them that quickly, the lag and center zoom makes live view unusable for portraiture, which is what I wanted to use it for. I have never had my rangefinder go out of alignment in all these years, and I love the rendering from the M9.

 

Leica had to switch to a CMOS sensor to compete with other camera makers, and compete they must at this point because they are not the only small full-frame camera manufacturer on the market anymore. And even though others don't make rangefinders, Leica's sensors will be compared to the competition.

[Jeff S]

... yet everyone that switched to the M240 thinks he is crazy...

 

….There were people that preferred the M8.2 over the M9...

 

I switched to the M, and I don't think he's crazy. Whatever floats his/your boat. But I do suspect that there's more to it than the CCD vs CMOS issue, e.g., bayer array and color rendering, IR filtration, etc…each of which is independent of sensor type….along with user techniques and preferences.

 

And, yes, I'm one who preferred the M8.2 to the M9. But I prefer the M over my M8.2…camera and files. Guess I'm the lucky one.

 

Jeff

[jaapv]

One does not compare film on Wallmart prints. For a proper comparison use proper developing.

I don't need raw files, I have played with them. I want to see a picture taken with the M9 unaltered in LR with the standard Adobe Profile applied, and the same picture taken with the M240 adjusted to look exactly the same.

 

I have tried and gotten close, I pointed out the difference I see in the pictures that were posted earlier in this thread, yet so many people here claim it can be done. Please show me how close you can get, because when CCD fans are accused of not knowing how to post-process, then these accusations may just stem from people who can't perceive subtle differences.

 

And as far as IR sensitivity is concerned, yes, I notice much stronger sensitivity to IR with the M240 than with the M9, but filters? Now we are back to that? And this time without automatic correction for cyan corners with wide lenses?

 

The M240 is a first generation camera, like the M8, with all kinds of problems, and what it is trying to replace is too good of a camera for me to be an early adopter again.

[Paulus]

The recent issues with the current CCD sensor notwithstanding, there is a strong demand for CCD sensors in M cameras.

 

My Open Letter to Leica has garnered over 400 signatures by discerning photographers, artists, and general photo enthusiasts.

 

Please Leica, keep CCD alive.

 

—Peter.

 

First of all, thank you for this initiative. Leica will certainly read this and act accordingly. They have listened to this forum many times. Not alway with the expected results but they have a business to attend.

 

Second;

 

I have enjoyed the M9 with ccd very much. Still the decision to go to the M was one of better iso. Not better colours. With the M9 I just hadn't enough light to make the picture I wanted. I also noticed that I did not took so many colour photographs with the M9 because of the limitations I felt, or wasn't able to correct properly ( my fault ) or I did not wanted to spent time on it. With the M 240 I don't feel this problem any more.

 

I took the liberty of comparing some of my shots. I also see a difference in the way I develope my M 240 files. Maybe because it's more flexible. I also notice that I use different colour choises. I do not really know if it's impossible to get the same results/colour, at the same time with the two different camera's because I was never in that position. I do know that my colour choice has changed in time. Maybe because of this camera and its possibilities.

 

See:

 

M 9:

 

Affligem kleur - pauljoostenfotograaf

 

Duik 15 mei 2013 Oostplas. Tilburg. Platform - pauljoostenfotograaf

 

 

M 240:

 

Saint Michaels Castle " Sint Michielsgestel" Brabant - pauljoostenfotograaf

 

Ballonvaart en landing in Moerenburg: "Trying to catch a Virgin." - pauljoostenfotograaf

 

Landscapes around the Baltic Sea. - pauljoostenfotograaf

[luigi bertolotti]

Advertisement (gone after registration)

Leica still has 4 CCD Cameras in their pricelist (2 x M, 2 x S) ... and I honestly think that it's the most they can do now for the users who appreciate particularly the imagery from this technology : I do not enter the discussion in itself, which involves levels of skill, knowledge and experience on raw developing on which I don't feel myself up to.. but I make just some consideration on the industrial side of the issue :

 

- Leica in itself hasn't any control on sensor technology.. their choices are depending on other actors' choices in terms of R&D.

- Leica is surely not so a big company to drive / pay the R&D of those actors : there wouldn't be any economic sense into.

- The fact that Phase One still credits CCD of some pluses over CMOS is far by strange : CCD backs are still a flagship product in their product line : they would be autolesionist to declare squarely a definitive shift to CMOS.

[Lenshacker]

Back-Side Illuminated sensors when available for full-frame cameras will solve many of the problems that CMOSIS had to work around with their CMOS sensor. As stated before, Samsumg has an APS-C BSI sensor. It would be interesting to see Leica come out with a "T2" with the APS-C BSI sensor, it should do well with M-Mount lenses on adapters.

 

It will not be long before a full-frame BSI detector is available, when it is- many of these problems of CCD's requiring off-chip A/D and processing and CMOS sensor geometry will be bypassed. They basically have almost 100% fill-factor, have the photo-sensitive sites close to the surface of the chip, and allow for on-chip a/d and signal processing.

Edited January 31, 2015 by Lenshacker

[Jeff S]

A BS detector might also come in handy on the forum from time to time.

 

Jeff

[Lenshacker]

On this forum, it's difficult to know where a lot of the comments come from, especially rude and dogmatic statements. If you are referring to BSI sensors, do a google search on them. The technology has mostly been used for small detectors, for cell phones. Hit the 1" sensor size not long ago, and already at APS-C with a 28MPixel arrays. Digital sensors don't do well with steep angles of incidence, as are produced by lenses close to the image plane. CCD's have a high fill-factor, meaning most of the surface is light-sensitive. CMOS sensors- do not have a high fill-factor. BSI sensors have the advantage of the high fill-factor of the CCD and the on-chip peripherals of CMOS. I would think this technology would do well in a Digital Leica M. If I am missing a technical issue of why this would not do well with a Leica M mount camera, please state it.

Edited February 1, 2015 by Lenshacker

[Jeff S]

But first I guess we need a humor detector. Oh my.

 

Jeff

[batmobile]

I think the dogma in this thread is a product of a flaw in the logic being applied to deriving a solution from the problem and for those who can see that, its quite a surprise to see it being ignored. It goes something like this:

 

I don't like the the rendering of the M240, which has a CMOS sensor. I do like the rendering of the M9, which has a CCD. I therefore want Leica to produce a future M with a CCD, because there must be a relationship between what I like about the M9 and the fact it uses a CCD.

 

The above is false. It is the difference between correlation and cause. As evidenced by the fact that the OP has referenced several CMOS sensors with an output he does like (unlike the M240), it would seem Leica could produce files that the OP likes very much with the right CMOS sensor, filter glass, profiling etc.

 

While I completely respect and understand the OP's desire to find a solution to his problem, his right to write to Leica and ask for what he wants and everything else, the negativism (IMHO) stems from the fact that the logic chain running up to that act is deeply flawed and this has largely been ignored. The inherent differences between CCD and CMOS may exist, but I think everyone recognises that they are relatively minor, of everything else is done right in terms of output. Having a very slight preference (when the other variables are controlled) does not amount to a burning need. Hating M240 colour or skin tones is a completely different issue.

 

I wish him good luck, but suspect launching a new M with a CCD would amount to a financial disaster for Leica. There is also a distinct possibility that the 400 signatories will not like the output of a new CCD camera any more than then M240 successor.... after all, Kodak would not be able to supply the CCD and it would be a whole new process resulting in what might amount to another beta camera, which is another thing Leica does not need. They need to get the current issues licked, not jump around with new sensors, starting from the beginning each time. But hey, they will do what they feel is right and maybe the OP will get what he desires?

Edited February 1, 2015 by batmobile

[01af]

I think the dogma in this thread is a product of a flaw in the logic being applied to deriving a solution from the problem and for those who can see that, its quite a surprise to see it being ignored.

Exactly.

 

 

It goes something like this:

 

I don't like the the rendering of the M (Typ 240), which has a CMOS sensor. I do like the rendering of the M9, which has a CCD. I therefore want Leica to produce a future M with a CCD, because there must be a relationship between what I like about the M9 and the fact it uses a CCD.

For a more formal phrasing, see post #68 (and don't miss the wonderful reply #69)

[pgk]

I don't like the the rendering of the M240, which has a CMOS sensor. I do like the rendering of the M9, which has a CCD. I therefore want Leica to produce a future M with a CCD, because there must be a relationship between what I like about the M9 and the fact it uses a CCD.

 

The above is false. It is the difference between correlation and cause.

Fair enough, so OK, then try this for logic. There is a difference in the rendering of different Leica M digital cameras. But colour and tonal consistency are both keys in driving products forward. Therefore wouldn't it be logical to ensure that the results from each are at least as consistent as possible out of the camera? The major difference is in the sensors.....

 

And as I have said before, its the consistency that I want - I'm not overly bothered or even interested in the sensor specifications provided they produce the results I want (ie are consistent) and I do like the output from the M9.

 

The flaw in the logic is assuming that the results from different cameras can be made to look identical - sure they can be narrowed, with a fair amount of effort, but not made to look identical, but by the logic being applied by some here, the output could be made identical if Leica desired to do so - and I am quite sure that they would prefer consistency for marketing purposes.

[Lenshacker]

CCD's have been around for 40 years, and is a mature technology. The CCD's used in the Leica M digital cameras were made by Kodak, which has a lot of experience with color film. Their engineers spent a long time making the output look like film, this includes the dye used in the color mosaic filter and the dopants used to make the chips. The IR cover glass -despite the environmental issues- is the best available with regard to optical properties.

 

I wonder if two BSI sensors from Samsung can be joined to make a full-frame sensor. The Kodak sensors are made in halves, then joined to give full-frame. Personally, I think once BSI full-frame sensors are used in cameras that I'll be the only one that reminisces over CCDs. I spent a lot of the 1980s and early 1990s doing the A/D and Image processing portion of the imager. Of course it took 4 racks of equipment and a 28-track High-Density Digital Recorder to do 2FPS. But it was impressive.

 

I don't think the OP expressed that he likes some CMOS sensors, that was me. I stated that if the M240 had a sensor as good as the Df, I would want one. What I did not state: it's probably not possible. The Sensor used in the Df has very high saturation count, meaning it has a lot of material- big and deep pixels. Works with SLR lenses. CMOSIS most likely had to thin the sensor, meaning less material. The Df shows no signs of banding at ISO25,600. The M240 shows banding at ISO6400. I'm not sure why. I've shot the M9 at ISO2500 and -1ev, push in post. No signs of banding. I would expect better performance from the M240 than the examples shown.

 

Just as an example- this is with the M9 at ISO2500 and -2EV, Photoshop used to boost the exposure in Post. ISO10,000 equivalent.

 

Skating_J3_Wideopen_ISO10000

 

Noise- yes. I'm impressed by the sensor uniformity, ie lack of banding.The KAF-18500 is made in halves, the seam is not obvious in this photo.

 

Some 30 years ago my boss asked me what I was going to do when my chosen profession made my favorite hobby obsolete. Told him I had a Nikon F with an F36 motor drive that could be converted to a digital back. On Technology sells development boards for their CCD's and 3D printers will make the job a lot easier now, in the 21st century.

 

Smilies can be used for humor indicators.

Edited February 1, 2015 by Lenshacker

[batmobile]

I'm sure that Leica did try to produce a degree of consistency (as well as what they thought people liked) but did not get it quite right for most people. They didn't really get it right for me either, I might add. I bought a MM instead of a M240 because there was enough that did not win me over on the latter's image quality for my purposes so I went for a B&W only solution that I did like.

 

The major difference is not only in the sensor type, but in who manufactured them, their resources and experience (not to mention possible changes in Leica's direction/input). I would say this is as important as anything to do with CCD vs. CMOS. After all, if the M240 sported a sensor that performed just like the one in the A7R (forgetting technical reasons why this might be hard to achieve), I wonder whether satisfaction levels would be higher? I suspect a 'yes' of monumental proportions: great colour, superb clarity, good skin tones and fantastic file flexibility. Even if some nuances are not quite M9-alike, there would be epic benefits.

 

I'm not sure its all that important to make cameras look identical, as we are not in the business of shooting all our portfolios on two cameras and showing both results to our viewers. If the results are 'very good and attractive' comparisons to another camera (an arbitrary benchmark) is fairly irrelevant, even if in one parameter or other, the other camera has some preferable nuances. Its the same as looking at your portfolio and not liking wonderful images because they might have been 'better' shot on the new APO super super lens, had you owned one at the time (but didn't).

 

While it may not be possible to get the M240 to look precisely the same as the M9 all the time, it is clear that it can often be made 'close'. That may not be close enough for some and may be too much work (or beyond the skill ) of others, but again, this does not encessarily have anything to do with CCD vs. CMOS. It only relates to the (1) CCD in the M9 and (2) the CMOS in the M240. Any future CCD camera would not have the same CCD and processing as the one on the M9 and the M240 successor will not have the same CMOS as the M240, so we end up with a bunch more moles to chase with our mallet. It would therefore seem easier to summarise the requirement as:

 

Dear Leica, whatever you do and however you do it, please ensure the next M produces better colour than the M240, far more acceptable skin tones and has less banding under pressure.

 

As I see it, CCD or CMOS has nothing to do with it, as long as the person making the request recognises that there are some CMOS cameras whose output they like very much. If no CMOS cameras would be acceptable to them, my eyes would bulge..... and I would then ask to see their photographs!

 

Fair enough, so OK, then try this for logic. There is a difference in the rendering of different Leica M digital cameras. But colour and tonal consistency are both keys in driving products forward. Therefore wouldn't it be logical to ensure that the results from each are at least as consistent as possible out of the camera? The major difference is in the sensors.....

 

And as I have said before, its the consistency that I want - I'm not overly bothered or even interested in the sensor specifications provided they produce the results I want (ie are consistent) and I do like the output from the M9.

 

The flaw in the logic is assuming that the results from different cameras can be made to look identical - sure they can be narrowed, with a fair amount of effort, but not made to look identical, but by the logic being applied by some here, the output could be made identical if Leica desired to do so - and I am quite sure that they would prefer consistency for marketing purposes.

[mjh]

There is a difference in the rendering of different Leica M digital cameras. But colour and tonal consistency are both keys in driving products forward. Therefore wouldn't it be logical to ensure that the results from each are at least as consistent as possible out of the camera? The major difference is in the sensors.....

Is it really? As a matter of fact, between the M (Typ 240) and the M9 there is very little that is not different – apart from both being M-mount rangefinder cameras these two models have next to nothing in common. The M9 is much closer to the M8 than it is to the M (Typ 240).

[Lenshacker]

The spectral response of the KAF-10500 used in the M8 and the KAF-18500 used in the M9 is very different, more so than can be explained by the IR absorbing cover glass. The M8 relied on the red dye used in the Bayer filter to help reduce IR. The QE for red on the M8 is ~21%, the M9 is ~30% (edit, latest datasheet). Maybe why the M8 has more noise in the red channel.

Edited February 1, 2015 by Lenshacker

[Advertisement]

Hallo,

Look here: Leica M

[Prosophos]

To re-iterate:

 

In my opinion (and in the opinion of the signatories of My Open Letter to Leica), the M9/M-E/Monochrom/M8 sensors provide superior colours, tonal transitioning, and micro-contrast vs. the M240. The CCD sensors also produce files that are more robust for the purposes of post-processing.

 

To some, the differences are obvious. To others, nuanced. To yet others, non-existent.

 

Some CMOS sensors, like the one in the Sony A7S, come close to performing like the M9 at base ISO (I know because I've owned it), but the edge in my opinion still goes to the M9. Considering the age of the M9 (and M8) sensor, that is quite remarkable.

 

Imagine, if you will, what CCD technology could accomplish if development continued.

 

Though I understand that a 40MP 35mm CCD sensor was created after the M9 sensor (I'm not sure for what application) -- which would be interesting in a future M camera -- I personally would be quite happy if all Leica did was fix the cover glass corrosion issue in the current 18 MP chip and then placed it inside an updated M body. Certainly it would be economically doable for them since they've already committed to supporting corroded sensors anyway. In doing so, they could continue to recoup their investment in the form of additional CCD camera sales (to those of us that are either perceptive or delusional, depending on your point of view).

 

The move to CMOS, let's be honest, was (and is) driven by economical considerations, by the convergence of still and motion (video) photography, and by the need (of some) to photograph in the dark. It had nothing to do with quality still photography performed in daylight.

 

For that one application, I'd choose my M9 (and the CCD chip inside of it) every time.

[Prosophos]

 

Dear Leica, whatever you do and however you do it, please ensure the next M produces better colour than the M240, far more acceptable skin tones and has less banding under pressure.

 

On this, we can agree.