Color Photography and the Leica M (New article at overgaard.dk)

[mirekti]

Advertisement (gone after registration)

...And a thinner body than previous digital M cameras.

 

Do you know something I don't? ;-)

[gpwhite]

How do you find the digital age for color photography, compared to the film days, bocaburger?

 

Thorsten, perusing this thread, I am relieved to read that Jaap picks on other people too,  , even famous ones.

 

Your classes probably are not representative of the larger Leica world, as these are shooters you have guided to take "their pictures" as you say.

 

I work to prints, which remains a very different medium than computer monitors. I can make my own prints from digital today that are more snappy, satisfying and larger than what I could do with Velvia 50 and a local Cibachrome print shop (and at a fraction of the cost). B&W prints, however, remain a project as some prints snap like the old Agfa papers and some do not. In other words, in my hands, digital color is a vast improvement over film, but digital B&W, as good as it is out of the M or MM, yet translates consistently into great prints.

 

P.S. You might remember this is why I asked about a printing workshop a while ago.

[samwells]

Yes, a fascinating and informative article - thank you. As an aside, it's interesting that, in the digital age, there's constant search for 'accurate' colour in photographs. This just isn't so in other art forms - painters, for example, find rich blues in shadows, cinematographers constantly shift the palette (e.g. Antonioni, Powell/Pressburger, Conrad Rooks). Somehow, photographers seek to replicate what their eyes see, rather than what the camera - or their imagination - sees (and yet, that's obviously not the case with b&w, which bears little resemblance to what our eyes see). Sometimes, I long for the infidelity of Dufaycolor or autochromes...

[Peter Kilmister]

Dear Thorsten,

 

I am a happy amateur. You are someone that people regard with great respect. Having read your article I can see why ... although some of it went way over my head. (That's quite normal.)

Hopefully you will be able to answer a simple question that seems quite complex to me.

Would it be better to use Lightroom rather than the M240 to produce B&W or vice versa?

The reason I ask is that the M240 has a setting to take B&W. Lightroom has an option to convert to B&W.

 

Over to you. I hope you can help.

[brt]

Mr. Overgaards statement regarding colorblindness is misleading and completly wrong.

Anybody interested in colorblindness may take a closer look at https://en.wikipedia.org/wiki/Color_blindness

 

Regards

brt

[MT0227]

Would it be better to use Lightroom rather than the M240 to produce B&W or vice versa?

The reason I ask is that the M240 has a setting to take B&W. Lightroom has an option to convert to B&W.

 

Pkilmister.....I believe Thorsten's B&W workflow would be to shoot DNG + B&W JPG Fine; using the B&W JPG Fine as your source file. 

 

Use the JPG file from the camera for black and white

Before we walk together into the beginning of the rainbow, let's talk about black and white for a moment more. The genius of the JPG Fine setting in the Leica M9 was that the JPG was a really, really good black and white file. A little bit of editing on the JPG Fine and you were done.

In fact, when I tried to convert the color DNG to black and white in Lightroom and edit it towards the look I would want from my black and white photography, I (almost) always ended up liking the cameras JPG Fine (that I had also edited towards the look I wanted) better when I compared the two.

 

The DNG is 14-bit and has many layers of information from the sensor, the JPG Fine from the camera is 8-bit and have only one layer of information. Yet, the JPG in black and white from the camera wins in my opinion.

After having tested them against each other for a long while, I finally decided that the JPG Fine in black and white won over the converted DNG file in 98% of the cases. Only in very rare cases where I needed to fix something would the DNG offer something I liked better.

[Rick]

Advertisement (gone after registration)

Mr. Overgaards statement regarding colorblindness is misleading and completly wrong.

Anybody interested in colorblindness may take a closer look at https://en.wikipedia.org/wiki/Color_blindness

 

Regards

brt

 

I know, but if you made it through Maxwell being known most for color vision (forget Maxwell's equations!), Thorsten's understanding of the of the physiology of color vision and the science of perception of color vision doesn't seem as misguided (ok, yeah it is).  I see it more like a poet's view of science.  Sorry, but the rest was interesting.

 

Rick    

[pico]

There is color blindness which is a physiological fact and some persons' disregard for color. A computer scientist I know claimed he was color blind but it turned out he just did not care a whit. A red or green light were the only social reality clues he paid attention to. Nuance was unknown to his binary view of life

.

[Michael Geschlecht]

Hello Mirekti,

 

As per Thorsten's "thinner body" in his Post #21 just above yours: He might have been thinking about the flange to sensor distance which is shorter in a 601 than it is in an "M".

 

Best Regards,

 

Michael

[mirekti]

As per Thorsten's "thinner body" in his Post #21 just above yours: He might have been thinking about the flange to sensor distance which is shorter in a 601 than it is in an "M".

 

Thanks for pointing this out. SL still needs an adapter for M lenses, right? ...so the distance might be the same.

[MT0227]

SL still needs an adapter for M lenses, right? .

Yes it does. Quite nicely I might add.

[Peter Branch]

Sorry - but I just don't get what this article is trying to convey.

 

On a technical level what has the bit depth of the conversion of the luminance recorded by each pixel got to do with resolution, which is here called detail? 

 

On a quite different level one of the interview questions for entry to Cambridge University's Engineering Department used to be: -

 

"Which is more difficult, colour photography or black and white, and why?"

 

The answer that wanted, but rarely got, was black and white because it involves a higher degree of abstraction. 

[Peter H]

.........

 

On a quite different level one of the interview questions for entry to Cambridge University's Engineering Department used to be: -

 

"Which is more difficult, colour photography or black and white, and why?"

 

The answer that wanted, but rarely got, was black and white because it involves a higher degree of abstraction.

Yes, that sounds just like an engineer's approach to photography.

[MT0227]

Sorry - but I just don't get what this article is trying to convey.

 

"Perfect" color is perceived differently; some see it differently, some can't see it at all. Get it close, work with it, if it looks good to You...You're done.

[Peter Branch]

Yes, that sounds just like an engineer's approach to photography.

 

The question is, of course, ironic!

 

Colour photography is undoubtedly more technically complex than B&W so students applying to a highly technical university department might be tempted to assume that the more technically complex process was the required answer.  There is, of course, no "correct" answer however the interviewer is seeking to establish that the interviewee has a broad understanding of the issues involved, which are NOT confined to the technical complexity.

[Peter Branch]

"Perfect" color is perceived differently; some see it differently, some can't see it at all. Get it close, work with it, if it looks good to You...You're done.

 

Precisely - so what is all the pseudo scientific mumbo-jumbo doing in the article? 

 

Seriously - the author is widely recognised as being closely linked with Leica Camera.

This sort of article does the reputation of neither the author nor Leica Camera any favours.

 

Leica Camera need urgently to get a grip of the situation.  It's not the first such article.

At the very least they need to approve the technical aspects of such texts.

 

He is a good photographer, though personally I'm not a fan of all that f/0.95 stuff, but he needs to concentrate on his areas of genuine competence.

[pedaes]

I have to say I found the article both interesting and helpful.

[Mornnb]

Sorry - but I just don't get what this article is trying to convey.

 

On a technical level what has the bit depth of the conversion of the luminance recorded by each pixel got to do with resolution, which is here called detail? 

 

That is not what he was saying here, he was trying to say the advantage of bit depth is that the camera can record a greater number of tonal values in a single scene, one can use the word 'detail' to describe such. Detail does not only refer to resolution.

 

Although the section in this article referring to 'detail' does make an error. When one refers to a 14bit RGB sensor such as the M240, this is the bit rate per colour channel, i.e. you must multiply it by 3. Which is capturing a lot more luminance detail than a 12bit monochrome sensor. 

[Peter Branch]

That is not what he was saying here, he was trying to say the advantage of bit depth is that the camera can record a greater number of tonal values in a single scene, one can use the word 'detail' to describe such. Detail does not only refer to resolution.

 

Although the section in this article referring to 'detail' does make an error. When one refers to a 14bit RGB sensor such as the M240, this is the bit rate per colour channel, i.e. you must multiply it by 3. Which is capturing a lot more luminance detail than a 12bit monochrome sensor. 

 

Using the Leica M (240) and M (246) as examples each camera has the same number of pixels on the sensor.  In the case of the M (240) the software has to calculate 3 values for each pixel; Red, Green & Blue.  This it does by interrogating, in some undisclosed way, 4, not 3, pixels; 1 blue, 1 red and 2 green. (There are twice as many green pixels as there are red or blue on the sensor - the Bayer array).  The mathematics of how this affects the "resolution" is not straight forward as it varies with the colour channel and other factors such as direction.  Suffice it to say that the effect is to reduce the effective spatial resolution relative to the B&W sensor which has no need of such algorithms.

 

The question posed is whether the fact that the colour sensor has three colour channels, each of 14 bits luminance resolution, per pixel makes it capable of conveying more information than the single channel, 14 bit luminance resolution for each pixel on the B&W sensor.  It is not possible to get something for nothing.  Because the colour channel signals are derived from adjacent pixels they are not independent; to some degree they are correlated hence the resolution has been reduced.  So the colour sensor, assuming 100% efficient processes, generates more information in the form of colour differentiation than the B&W sensor at the cost of lower resolution.

 

Given that spatial resolution is much easier to demonstrate than colour resolution it is not altogether surprising that the M(246) is preferred for demonstrating the performance of lenses like the 50mm f/2 Apo-Summicron.   

[Advertisement]

Hallo,

Look here: Leica M

[Sharpdressed]

The question posed is whether the fact that the colour sensor has three colour channels, each of 14 bits luminance resolution, per pixel makes it capable of conveying more information than the single channel, 14 bit luminance resolution for each pixel on the B&W sensor. It is not possible to get something for nothing. Because the colour channel signals are derived from adjacent pixels they are not independent; to some degree they are correlated hence the resolution has been reduced. So the colour sensor, assuming 100% efficient processes, generates more information in the form of colour differentiation than the B&W sensor at the cost of lower resolution.

 

 

You seem to not understand that both M240 and M246 have a b/w sensor (the same actually). The difference is the presence of the bayer filter in the M240.

It's like saying that on b/w film with a yellow filter you have more resolution.