CCD or CMOS - you choose

[jonoslack]

I agree, in the sense that it is very much like having a discussion about religion. All the facts can be laid out, but one must be careful to not tread on faith. Faith is the deliberate decision to believe. Faith does not have to concern itself with facts. You just have to believe that something looks better, then describe it as an aesthetic or a feeling that is personal. End of objective discussion. Don't tread on faith.

But, I can tell you as an owner of the M8, M9 and M... and not just borrowing them for a week or looking at photos on the internet, but really working with the files in PP for months and years that they all can be made to look the same. Except, once you leave base ISO where the M retains its color depth much, much better. Or, if you are trying to recover highlights or bring out shadow tone where the M has the advantage.

I also print a lot and I often print huge on a HP Z3200 44" printer. Both cameras are able to produce, under most situations, identical prints that I know for a fact that none would be able to tell the difference in the feel or the esthetic of the sensor type.

But, on the internet people can claim anything they want. Come over to my print office and look at my prints and try and identify the esthetic that you like so much and I'll tell you if you are picking the correct sensor... you won't be able to. In print it isn't even an issue of equal but different. In print they are equal.

Rick :deadhorse:

Hi There Rick - I'm certain you're correct - I also know for a fact that without delving into pixel count etc. etc. I cannot tell the camera which took a properly processed image just by looking at it. . . . .

But telling people this kind of common sense only infuriates them and makes them say that's because You (or Me) have 'cloth eyes'. :chug::ROTFL::facesmack:

 

[Auni]

Hi Jono,

You said something above somewhere (long thread) that I also agree with. It has to do with why come on these threads (like I do from time to time) and argue something that seems so subjective? And, the reason I would give is that folks read these threads and sometimes believe that experts can tell the difference and have a personal preference of CCD over CMOS. Which sounds like that is ok. Why bully them? Can't they have an opinion?

Yes they can. I have actually learned from Marc and have looked at the output from the Sony cameras and I do see a difference in skin tone and I agree with him now. The Sony is more pleasing and much easier to work with in this area. (I believe it is a side effect of the thin sensor and the IR contamination from the Leica sensors having to be made so thin).

But, the CCD and the CMOS issue is just a subjective belief. It can't even be proven when we look subjectively at posted images. Objectively, no one can subjectively tell a difference.

So, propagating this belief does have an effect... to Leica and to those who will miss out on the M. This is the problem with floating someone's belief or perception as fact. As the saying goes, perception is reality. If, people believe there is an objective difference it becomes reality... and that has a knock-on effect.

Rick

 

[jonoslack]

Hi Jono,

You said something above somewhere (long thread) that I also agree with. It has to do with why come on these threads (like I do from time to time) and argue something that seems so subjective? And, the reason I would give is that folks read these threads and sometimes believe that experts can tell the difference and have a personal preference of CCD over CMOS. Which sounds like that is ok. Why bully them? Can't they have an opinion?

Yes they can. I have actually learned from Marc and have looked at the output from the Sony cameras and I do see a difference in skin tone and I agree with him now. The Sony is more pleasing and much easier to work with in this area. (I believe it is a side effect of the thin sensor and the IR contamination from the Leica sensors having to be made so thin).

Yes indeed - there certainly is IR contamination - mind you, lots of other cameras have it as well, it's just not beefed about! The downside of the thicker cover glass on the Sony cameras is, of course, the problems with wide angle M lenses at the corners (no free lunches).

But, the CCD and the CMOS issue is just a subjective belief. It can't even be proven when we look subjectively at posted images. Objectively, no one can subjectively tell a difference.

So, propagating this belief does have an effect... to Leica and to those who will miss out on the M. This is the problem with floating someone's belief or perception as fact. As the saying goes, perception is reality. If, people believe there is an objective difference it becomes reality... and that has a knock-on effect.

Rick

Indeed Rick - I couldn't agree more. . . . . . But you mustn't think Marc and I are at loggerheads - we agree with each other about most things and go back a long way!

 

[aDam007]

IMO it doesnt matter if the difference people see comes from the CCD vs CMOS, or from something else. As long as everybody knows which camera he/she prefers we are all fine.
I prefer the M over the M9, but I also prefer the S over the M
I dont know yet if I would prefer the S007 over the S006, I will wait and look (at the images but also at the price development).

I was going to say this above point word for word... I would have added one thing though in Marc's defence. Something nobody here is mentioning.

Marc shoots a LOT of frames (weddings). One must also take into consideration time spent editing. If Marc decides to us an M9 and gets the images he likes almost right off the bat, then he's saving a lot of time and hassle trying to figure out what to do with the files. Thus the M9 is god-sent in comparison to a M240. His sensor choice doesn't matter, it's how he gets to his end result. So yes, you can make an M240 image look IDENTICAL to an M9 image (CCD vs CMOS debunked) but that doesn't mean you won't be working for it. If you like the base file from a M9, then it's the easier choice, assuming you can put up with the quirks of the M9 body/architecture.

And also to throw a curve ball at all of you.. I've done the testing. My conclusion is that the M9 sensor reflections and M240 sensor reflections are different. Thus different lenses react to the different sensors in a way that affects the lenses ability to handle certain lighting conditions. It's subtle and not always there, but it's there. And no bayer filter or post processing is going to the fix the problems that arise from this issue. Most people chalk it up to lighting conditions, and it slips by most M9 vs M240 CCD vs CMOS sensor testers (simply because most people do basic bare minimum testing). I first noticed this when I was putting my defective 50APO through the ringer. I've since tested all my M lenses, which is almost every M lens in the current lineup. For the 50APO I had one of the first, it worked well on the M9, horribly on the M240 but had great color and contrast. After the fix in Germany, the color and contrast went down noticeably, but it worked well on the M240 with only minor issues. I was mad about the color/contrast going down, so I sent it in a second time, but it came back the same, working on the M240 but with reduced color and contrast. Had I known, I would have put up with the defective lens, and kept the color/contrast high.
I will add, I also prototyped the 50APO before it was announced and never noticed a problem with the prototype, but that was before the M240 existed, so I was using it with an M9. I think out of 400-500 test shots around town over two days I seen flare from the M9 twice, and some ghosting in only the most extreme situations (but maybe the 50APO prototype didn't have problems like my 50APO production model). With the production model on the M240 vs M9 I could get ghosting and flare about 50% of the time on the M240 and still very low on the M9. And no I wasn't using LV on the M240 to coax out the problems.

Anyway, bottom line. CCD vs CMOS is personal choice. Is CCD magically better, no not on it's own. Their are a number of factors at play, but the factors do influence how you get to the same end result. So just choose the camera that's right for you and be happy to shoot with it.

 

[Auni]

But you mustn't think Marc and I are at loggerheads - we agree with each other about most things and go back a long way!

I know. And the same goes with Marc and me. There are several examples where Marc has turned out to change my mind. But, no real surprise here, he has been at this professionally for a whole career. I wish he lived next door. I could sit down with him and try and figure out what he is seeing and I could show him what I see in piles of prints I've made.

But, since he doesn't live next door I just have to believe he is wrong.

Rick

 

[Malina DZ]

How hard is it for Leica to create an M9 color profile for M240? All these CCD vs CMOS debates would not even start if manufactures cared to provide the consistent color output their users/fans have appreciated with the previous sensor generation. Leica and Pentax shooters are lucky to have a choice between CCD and CMOS to mount the lenses on. Canon, Nikon, and Sony don't offer such a luxury in 35mm FF format.

 

[D&A]

I've worked with M9 and M240 files since the inception of each camera. Although I might not be able to reliably tell the difference between the files of each camera in Dave's posted test, I do know when I work intensely with the files from both cameras (especially when printing large), which images I most often prefer. Whether others viewing my prints feel the same, I can't say but as I pointed out in my original post, its the visceral response I get upon viewing matched sets of prints that matters most. This too has happened with some other CCD sensor based cameras. Are there measurable deficiencies in some of the images captured by these CCD based cameras? Absolutely, but nearer to perfection is not what I am after.

The 50mm Lux asph is technically and measurably a far better lens than its predisessor the pre asph. Yet I like some others, more often prefer the look of the files the pre asph produces than that from the Lux asph. Same goes for many other vintage lenses vs. their modern day counterparts.

Doesn't Leica lose sales when some internet chatter often espouces the virtues of these older classic lenses vs. what is being produced by Leica today? How come there isn't the same uproar and debate in the case of lenses? In my opinion, its a similar type of argument and just as subjective as the debate between the preferences for base iso files from the M9 vs M240. There really isn't a right or wrong in subjective opinions and preferences.

My perceived advantages of the M9 base iso images over the M240 is based on a quite narrow window whereby the M240 in so many other ways trumps the M9. Yet its that narrow window thats so powerful and important to me, and therefore what has held me back from switching over. Time and continual changes though may ultimately tip that balance.

Dave (D&A)

 

[Shashin]

The problem is called "confirmation bias."

I have no doubt people have preferences. The problem comes when try to ascribe cause. I had a conversation with a person that swore that a particular Alpa camera made sharper images than other camera, even though the optics, and in this case, film back were identical ( Alpa does not make these things). Apparently. The metal spacer that was the Alpa camera was the secret sauce. That person could not show me the difference in the images except to say that they could see it.

I really do not doubt that folks find a particular camera "better." I do not doubt their experience, nor do I doubt their skill. I simple doubt the ascribed cause, in this case.

I am not sure calling this a religious affliction is quite fair. Many photographers are very intuitive. Intuition can be a powerful driver in the creative arts. The problem comes when trying to rationalize intuition. It is extremely hard to reduce complex intuitive experience. While it is easy to reduce the CCD/CMOS issue technically, it still is difficult to reduce the intuitive experience of those having those system. While I can't subscribe to a CCD look, I don't doubt that photographers can have a real perception of systemic quality.

Sorry for the typos, iPads have drawbacks.

 

[fotografz]

The problem is called "confirmation bias."

I have no doubt people have preferences. The problem comes when try to ascribe cause. I had a conversation with a person that swore that a particular Alpa camera made sharper images than other camera, even though the optics, and in this case, film back were identical ( Alpa does not make these things). Apparently. The metal spacer that was the Alpa camera was the secret sauce. That person could not show me the difference in the images except to say that they could see it.

I really do not doubt that folks find a particular camera "better." I do not doubt their experience, nor do I doubt their skill. I simple doubt the ascribed cause, in this case.

I am not sure calling this a religious affliction is quite fair. Many photographers are very intuitive. Intuition can be a powerful driver in the creative arts. The problem comes when trying to rationalize intuition. It is extremely hard to reduce complex intuitive experience. While it is easy to reduce the CCD/CMOS issue technically, it still is difficult to reduce the intuitive experience of those having those system. While I can't subscribe to a CCD look, I don't doubt that photographers can have a real perception of systemic quality.

Sorry for the typos, iPads have drawbacks.

Well stated, and eye opening for both sides of the debate.

I sometimes also feel there is a bit of "Cognitive Dissonance" at play here ... also on both sides of the discussion.

While there may not be the pure scientific logic to support preference for the M9 rendering, there is plenty of empirical evidence. For me it is direct experience and a lot of it, fueled by intuitive observation ... and by extension that of others who's opinion I trust and value.

My pal Irakly is a great photographer IMO, and soldiers on with his M9, I see Doug of Wildlight Photography continue on with his 10 meg crop frame DMR with a CCD and electronics by Imacon when he could have adapted his long Leica R lenses to another camera long ago. Etc. Etc.

I do admit to feeling that there is a disproportionate hue and cry regarding the M(240) "differences", or supposed lack of them compared to the rendering from the M9 ... to quote: "A key tenet of cognitive dissonance theory is that those who have heavily invested in a position may, when confronted with disconfirming evidence, go to greater lengths to justify their position." thus my: "Me Thinks thou protest to much" post when a retailer posts their "evidence". As an ad man, I've sat through enough blind product comparisons to know how flawed that approach is.

To be fair, the same could be said about others defending the CCD verses CMOS based solely on second-hand web info.

Personally, I am not trying to defend a my personal preference and intuitions ... it is MY choice, and I sure am not trying to convince anyone else it is right or wrong ... just that it is right for me.

Heck, I currently do not have either a M9 or M(240) ... all rangefinder work is done with a MM ... so I have zero motivation to even engage in this debate. However, I do wish I could have smoothly transitioned to a M(240) and availed myself of the camera's functional improvements ... while retaining the look and feel I got from the M9 which IMO was unique amongst 35mm cameras, and a image companion in continuity with my S system.

Actually, it beggars the imagination that Leica didn't see their own unique image properties compared to the myriad other choices, and didn't do what was necessary in the CMOS image chain to retain it.

IF it is the non-sensor aspects that led to the M(240) differences, is it not possible to alter those? Can a M9 Profile be created as suggested by others here? If it is the sensor filtration, can a second version of the M be produced like how other camera companies are doing? I don't want a ME with less functionality, I want M functionality that renders in the unique manner the M9 did/does.

- Marc

 

[KeithL]

Actually, it beggars the imagination that Leica didn't see their own unique image properties compared to the myriad other choices, and didn't do what was necessary in the CMOS image chain to retain it.

IF it is the non-sensor aspects that led to the M(240) differences, is it not possible to alter those? Can a M9 Profile be created as suggested by others here?

As much as I like the out-of-camera look of the M9 files I'm not at all sure I'd want the M240 files to be made to look and or behave the same.

The M9 files are punchier largely due to having less dynamic range but are less amenable to post. I find the M240 files slightly flatter but far more malleable and the DR a real improvement.

I guess the preference depends on individual photographer’s subject matter and workflow. Perhaps I might think differently if I was a wedding photographer regularly having to process thousands of files.

 

[Paratom]

For my purpose I found out some factors where (to me) one has to handle the 2 cameras a little different.
One is exposure, I think with the M type240 its better to expose a bit lower compared to the M9. The M has more DR but it still can happen to blow out one of the color channels in some areas (which you do not allways see in the histogramm) which can lead to washed out colors/areas.
Another thing is the choice of lenses, probably due to the higher DR but also due to slightly different color result of the sensor I seem to prefer different lenses ofr each of the 2 cameras (M9/M). I do like for example the Summicrons a lot on the new M (which do draw slightly colder/neutral) vs my Summilux 50/1.4asph which has a little more red in the color.
Anyways, I think we also do not have to "over-discuss the situation", there are so many factors like sensor, software of camera, raw-converter, lenses and so many different tastes that is not possible to have a better and a worse cameras/concept.
And istnt it cool that in case of the M we even can buy both cameras and decide ourselves.

 

[jonoslack]

The 50mm Lux asph is technically and measurably a far better lens than its predisessor the pre asph. Yet I like some others, more often prefer the look of the files the pre asph produces than that from the Lux asph. Same goes for many other vintage lenses vs. their modern day counterparts.

Doesn't Leica lose sales when some internet chatter often espouces the virtues of these older classic lenses vs. what is being produced by Leica today?

I don't think they do . . .but people buy lots of lenses - but not so many camera bodies!

 

[jonoslack]

IF it is the non-sensor aspects that led to the M(240) differences, is it not possible to alter those? Can a M9 Profile be created as suggested by others here? If it is the sensor filtration, can a second version of the M be produced like how other camera companies are doing? I don't want a ME with less functionality, I want M functionality that renders in the unique manner the M9 did/does.

- Marc

HI Marc
as I understand it - yes, it would be absolutely possible to do this - although it would probably also require diminishing the dynamic range. . . . . It wouldn't even require a new version of the camera to do it - just a new version of the firmware. . . . . . but it would certainly irritate a lot of people who actually prefer the rendering of the M240, and putting it in as an option isn't something they'll do.

as an instance of this kind of problem.

When the M240 came out, you could no longer use Auto ISO in manual mode (ie when setting the shutter speed manually). This was as a result of several pleas from purists who said that manual should be manual - so it was actually switched off (it was there in beta firmware, just as it had been in the M9).

Of course - nobody had been sending messages to Leica telling them how much they liked using Auto ISO in manual mode (great for concerts, shootin. . . well, you know!).

However, when it was switched off they were deluged with messages asking for it to be switched on again (after all, the purists could always pick a fixed ISO).

Anyway - it's there as an option now

So there are 400 signatures on a 'petition' for leica to reintroduce a CCD sensor . . . .but even if they risked irritating those who like the rendering of the M240 and changed the colour in the firmware there's no guarantee that these people would then be happy.

Incidentally I still think the starting point of this problem was that the initial firmware did have an error in the WB calculations - which was fixed quite fast, but started off the whole thing.

 

[Godfrey]

...
I guess the preference depends on individual photographer’s subject matter and workflow. Perhaps I might think differently if I was a wedding photographer regularly having to process thousands of files.

... Although in that case, you'd be using a FF Canon or Nikon DSLR with a 24-75 and 80-200 zoom lens most likely. ];-)

Leica M-P + Summilux 35mm f/1.4 v2
ISO 200 @ f/2 @ 1/45​

All I can tell you is that I made this photo with countless cameras, including the M9 and the same lens, at the same ISO and aperture, and it wasn't until I made it with the M-P that I felt the camera had captured what I wanted.

This is an original JPEG (I'd forgotten to turn on raw capture, it was about my tenth exposure with the M-P...) which has had virtually no image processing other than frame and sizing applied. It is exactly what I have had in mind looking at this scene and light for years.

enjoy.
G

 

[D&A]

1. "and putting it in as an option isn't something they'll do."

2. "Incidentally I still think the starting point of this problem was that the initial firmware did have an error in the WB calculations - which was fixed quite fast, but started off the whole thing".

Regarding the two points you made...

1. Why not put it in as an option assuming that offering a M9 profile in a M240, would truly emulate file output from an M9. This way you satisfy both camps. Its conjecture though at this point that an M9 profile would be an acceptable alternative until implementation is initiated in early trials.

2. I wholeheartedly agree with this 2nd point Jono. Differences between the two different cameras was accentuated during the early release phase of the M240 and like a bad taste in the mouth for some, they never forgot it nor wanted to risk testing the waters again. Things have certainly changed for the better with additional firmware releases (for the M240), neverless differences for some still abound with both current cameras when files and subsequent images from both are examined. Its these differences that many current M9 users don't want to loose, regardless how well the M240 does in so many other areas.

Again we come back to subjectivity and it not being a question of right or wrong but what's best and liked by the individual user of these cameras, especially when their opinions are voiced respectfully on public forums.

Dave (D&A)

 

[Auni]

Jono, I for one would not like Leica to make the M240 render like the M9. I often had to work too hard to get the color right in my M9 files. The camera was too much work.

Other than skin hue, that I sometimes change the red toward orange (thin sensor glass?), the M is fine out of camera. By the way, the M9 had its own problems with skin hue especially in tungsten lighting and moderately high ISO.

Rick

 

[JorisV]

Jono, I still do not understand why the creative preferences of others is so important to you. Why the burning need to discredit anyone's personal preference for CCD over CMOS whether real or perceived? Why the insistence on quantitate measurable proof of a difference?

I quite frankly don't get it either...

The people who shoot M240 like their M240, the people who shoot M9 like their M9, what's not to like?

And objectively, if that is important to anybody, it is probably not so hard to prove that the latest Sony cameras beat them both...

No offense to some people but their efforts to convince other people are becoming slightly tiring and counter productive IMO...

 

[Vivek]

I quite frankly don't get it either...

The people who shoot M240 like their M240, the people who shoot M9 like their M9, what's not to like?

And objectively, if that is important to anybody, it is probably not so hard to prove that the latest Sony cameras beat them both...

No offense to some people but their efforts to convince other people are becoming slightly tiring and counter productive IMO...

:clap:

 

[jonoslack]

I quite frankly don't get it either...

Joris - I have NEVER criticised the creative preferences of others, not once, and Marc misrepresents me here: I never tried to discredit people's preference for an M9 for a second. I would be just as vocal about anyone who proposed that the colour of a CMOS sensor was better than that of a CCD.

The people who shoot M240 like their M240, the people who shoot M9 like their M9, what's not to like?

Nothing is not to like - I'm completely happy with this -

And objectively, if that is important to anybody, it is probably not so hard to prove that the latest Sony cameras beat them both...

Well, tough to be objective in terms of 'look' but DxO would certainly bear this out.

No offense to some people but their efforts to convince other people are becoming slightly tiring and counter productive IMO...

No offense taken - but I'm sad that you've so clearly misunderstood what I've been trying to get across - the LAST thing I want to do is to convince people that either camera is better . . . . .

. . .but seeing as it's so clear that I am not being understood, I guess I'd better shut up.

 

[Vivek]

Hi Jono, I (at least I did not) do not think it was directed entirely at you. Just look at some of the posts, everything about color is falling apart if you take some of the posters at their word.

I agree with Joris that this is a futile discussion that is not doing well for Leica in general (my take based on what I read here).