Leica M11 -purplish tint ???

[CDodkin]

Advertisement (gone after registration)

3 hours ago, M11 for me said:

Thank you for that. It is a bit aggressive though. I do not know why you have to switch into that mode. Here its just about magenta not about life or death.

Note that the original image was ooc. It had barely magenta cast (which per se is nothing bad at all). Remember that your corrected image with greenish cast was in no way any better. 

You appeared to be in denial - how else would you like me to dress up the facts?

[Advertisement]

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

[Ray Harrison]

At the end of the day, a raw file is luminance data (not color) passed through a color filter array and stored with metadata to aid in decoding these values. RawDigger is a great tool to look at something like that if you're ever interested.

Raw converters like LrC or C1 turn those luminances into color files, interpret the white balance from the metadata and so forth. The CFAs can be bespoke-tuned to enhance response to different color frequencies—think Phase's Trichromatic tuning on some of their digital backs using Sony sensors. I'm unsure if Leica has done any specific tuning on their CFA, and I'm not sure anyone here has attempted to measure it. I'm guessing that Leica did because someone here in the bowels of this thread sent their body back into Wetzlar, and they no longer have the sort of bias they'd seen previously. Whatever they did must have nothing to do with public firmware, presumably. 

Regardless, much of what we see on screen is down to the raw converter and the profiles applied to the file. I primarily use Capture One and, as a personal rule, rarely use AWB. Capture One's ProStandard profile for the M11 biases slightly towards a green tint whether I use a fixed WB or AWB; for me, tint values are usually between -1.6 and -0.2. Capture One's scale is -50 to 50, and Lightroom's is -150 to 150. C1's settings tend to be just to the left of 0, and Lr's tend to be more magenta-bound but only slightly to the right of 0, like 12-15. Capture One's profiles also tweak even specific settings of WB to being slightly cooler (say in the 4900 range for "daylight" rather than what LrC picks (5100)), and is camera-dependent (or profile-dependent).

I don't ever use in-camera JPEGs, so I don't know the story there, but should I note a magenta or green tint bias that I don't personally find pleasing, I can fix it 100% of the time, just like I fix any issues with Fuji, Nikon or Phase color cast or other color concerns. I rarely look for a "neutral" image regardless of my camera. I'm looking for a personally pleasing one. Or if I'm doing art repro/cultural heritage, I build a specific scene-based profile if needed (not that I use a Leica for that).   

Edited January 4 by Ray Harrison

[Luke_Miller]

While I don't have an M11 my M8.2, M9, and M240 tended towards magenta.  This was particulary noticeable in shots of a family member who has a lot of small veins close to the surface of the skin on his face.  My Leicas would produce shots that made him have a ruddy complexion, which was not accurate.  In response I profiled my M bodies using a color checker card.  Problem solved.  From this thread it seems this magenta bias continues in current bodies.  Should I get an M11 this magenta characteristic will not be an issue.

[jaapv]

That is probably IR contamination, Luke. Blood, specifically Heamoglobin,  reflects far more IR than fatty tissue. Try a few shots with an IR cut filter. I will always use one on any M camera when I want maximum colour fidelity or when I expect IR problems.

[hdmesa]

6 hours ago, M11 for me said:

...It is a bit aggressive though...

Note that the original image was ooc. It had barely magenta cast (which per se is nothing bad at all). Remember that your corrected image with greenish cast was in no way any better. 

Did you provide a link to the DNG? It's hard to fully correct the magenta cast on a JPEG without it looking too green. Correction to the DNG would look fully neutral.

[hdmesa]

3 hours ago, Ray Harrison said:

At the end of the day, a raw file is luminance data (not color) passed through a color filter array and stored with metadata to aid in decoding these values. RawDigger is a great tool to look at something like that if you're ever interested.

Raw converters like LrC or C1 turn those luminances into color files, interpret the white balance from the metadata and so forth. The CFAs can be bespoke-tuned to enhance response to different color frequencies—think Phase's Trichromatic tuning on some of their digital backs using Sony sensors. I'm unsure if Leica has done any specific tuning on their CFA, and I'm not sure anyone here has attempted to measure it. I'm guessing that Leica did because someone here in the bowels of this thread sent their body back into Wetzlar, and they no longer have the sort of bias they'd seen previously. Whatever they did must have nothing to do with public firmware, presumably. 

Regardless, much of what we see on screen is down to the raw converter and the profiles applied to the file. I primarily use Capture One and, as a personal rule, rarely use AWB. Capture One's ProStandard profile for the M11 biases slightly towards a green tint whether I use a fixed WB or AWB; for me, tint values are usually between -1.6 and -0.2. Capture One's scale is -50 to 50, and Lightroom's is -150 to 150. C1's settings tend to be just to the left of 0, and Lr's tend to be more magenta-bound but only slightly to the right of 0, like 12-15. Capture One's profiles also tweak even specific settings of WB to being slightly cooler (say in the 4900 range for "daylight" rather than what LrC picks (5100)), and is camera-dependent (or profile-dependent).

I don't ever use in-camera JPEGs, so I don't know the story there, but should I note a magenta or green tint bias that I don't personally find pleasing, I can fix it 100% of the time, just like I fix any issues with Fuji, Nikon or Phase color cast or other color concerns. I rarely look for a "neutral" image regardless of my camera. I'm looking for a personally pleasing one. Or if I'm doing art repro/cultural heritage, I build a specific scene-based profile if needed (not that I use a Leica for that).   

C1 is no better at preventing a magenta cast when using AWB or Daylight in camera with the M11 when imported. They always need adjustment to avoid magenta cast, at least in the environment in which I shoot (desert scenes at higher altitude).

Might be worth noting that if you use the Pro Standard profiles in C1, those are very skewed toward a more suble color representation. They will completely wash out vivid yellows from a sunset for example. Always use the generic profile if you don't want the creator of the Pro Standard profile's interpretation of your scene. The generic profile will give you something much closer to the Leica JPEGs. Pro Standard is great for portraits but terrible for landscape.

Edited January 4 by hdmesa

[3D-Kraft.com]

Advertisement (gone after registration)

vor 3 Stunden schrieb Ray Harrison:

At the end of the day, a raw file is luminance data (not color) passed through a color filter array and stored with metadata to aid in decoding these values. RawDigger is a great tool to look at something like that if you're ever interested.

Raw converters like LrC or C1 turn those luminances into color files, interpret the white balance from the metadata and so forth. The CFAs can be bespoke-tuned to enhance response to different color frequencies—think Phase's Trichromatic tuning on some of their digital backs using Sony sensors. I'm unsure if Leica has done any specific tuning on their CFA, and I'm not sure anyone here has attempted to measure it.

To be honest, I am not 100% sure, but I think Leica uses "Linear DNG" on the M11 and SL3 in order do allow to store it in three different resolutions. This is already demosaiced, so a color interpretation already happened partially inside the camera.

If you have such a "raw" file from the M11 and a tool that can show the photometric interpretation tag, you can check if it is set to 32803 (CFA, then it contains truly RAW data) or if this tag value is 34892 (linear DNG, partially demosaiced). For a deeper discussion, see https://forum.dxo.com/t/dng-files-should-all-work-out-of-the-box/37454/10

This means, there is less room for interpretation by the raw file converter and would explain, that this color cast is shown with LR/ACR as well as with C1.

 

Edited January 4 by 3D-Kraft.com

[Ray Harrison]

4 hours ago, hdmesa said:

C1 is no better at preventing a magenta cast when using AWB or Daylight in camera with the M11 when imported. They always need adjustment to avoid magenta cast, at least in the environment in which I shoot (desert scenes at higher altitude).

Might be worth noting that if you use the Pro Standard profiles in C1, those are very skewed toward a more suble color representation. They will completely wash out vivid yellows from a sunset for example. Always use the generic profile if you don't want the creator of the Pro Standard profile's interpretation of your scene. The generic profile will give you something much closer to the Leica JPEGs. Pro Standard is great for portraits but terrible for landscape.

Interesting info. I tend to use my Phase gear for landscapes / architecture work here in Colorado and the pro standard profile does work well on the color digital back I use, except for certain sunrise/sunset scenarios. Regardless, most of what I do gets adjusted, Leica, Phase or otherwise. I’ll try the generic profile for the m11, though, thanks for the tip.

[Ray Harrison]

4 hours ago, 3D-Kraft.com said:

To be honest, I am not 100% sure, but I think Leica uses "Linear DNG" on the M11 and SL3 in order do allow to store it in three different resolutions. This is already demosaiced, so a color interpretation already happened partially inside the camera.

If you have such a "raw" file from the M11 and a tool that can show the photometric interpretation tag, you can check if it is set to 32803 (CFA, then it contains truly RAW data) or if this tag value is 34892 (linear DNG, partially demosaiced). For a deeper discussion, see https://forum.dxo.com/t/dng-files-should-all-work-out-of-the-box/37454/10

This means, there is less room for interpretation by the raw file converter and would explain, that this color cast is shown with LR/ACR as well as with C1.

 

The photometric interpretation tag is CFA for Leica DNGs (at least for the M, not sure about the SL-series). 

[hdmesa]

35 minutes ago, Ray Harrison said:

Interesting info. I tend to use my Phase gear for landscapes / architecture work here in Colorado and the pro standard profile does work well on the color digital back I use, except for certain sunrise/sunset scenarios. Regardless, most of what I do gets adjusted, Leica, Phase or otherwise. I’ll try the generic profile for the m11, though, thanks for the tip.

The Pro Standard profile doesn't just hide yellows and allow you to bring them back via adjustments, they are desaturated from the image and irrecoverable without changing the profile to Generic.

[Tseg]

I understand Leica’s newer close focus M lenses provide correct color.  This is all a lens issue…. I think we are gonna need a new thread.

[Cattoo]

On 1/3/2025 at 8:17 PM, CDodkin said:

it were a Leica choice, then I'd expect the in-camera custom WB option to still show the same bias in tint. Given that they were choosing to do that.

Instead we see a calibrated target in-camera WB as neutral, and the AWB as magenta - indicating that the offset is indeed a function of the AWB. 

Or the fact that it’s a function of the AWB indicates that it is indeed a Leica choice.

I read in a couple of reviews three years ago that claimed that Leica told them that they wanted to reproduce a Kodachrome or M9 look specifically and only in AWB. Who knows whether reviewers claims are true or not, but my dealer coincidently told me the same thing when I bought my M11 three years ago. Ok, you might counter that it’s after the fact rationalization, but they said this long before it became a topic of controversy in this and other forums. 

Here language from one of these reviews:

Image quality from the Leica M11 is part of what makes this camera so exciting. In our meeting with Leica, they said that the colors are supposed to mimic Kodachrome. That told me that the Leica M9’s colors are back. And in reality, yes, I can confirm that this is true. However, it’s only valid in the auto white-balance mode. With manual white balance, my preferred method, you won’t get the same look. Though at times and with the right lens, you can get a similar halation look that you would from CineStill film.

He could have been making this up but he was incredibly prescient if so.
Here’s the link: https://www.thephoblographer.com/2022/01/13/three-great-cameras-in-one-leica-m11-review/

So for me personally, I expected that there would be a color bias in AWB because I read and was told there would be and that it was intentional, part of the design. Do I like it? Not in every instance but I accept it since it’s an easy correction. 

In any case, I’d wager that it would take most members of this forum less time to correct this color bias in-camera or in Lightroom than they’ve spent complaining about it. 

Edited January 5 by Cattoo

[don daniel]

I think this is complete nonsense or pure marketing talk. The M11’s white balance isn’t just magenta-biased in the AWB setting, but also in the sun, shade, and cloudy settings, as well as in all manual settings with a specific Kelvin value. Of all the settings relevant to daylight, the only one free of a magenta bias is the immediate white balance on a neutral surface (e.g., a gray card). This phenomenon is not a deliberate decision by Leica, but rather a mistake they refuse to admit—and never will. That doesn’t mean they won’t fix it. They likely will, but then they’ll dress it up again with marketing talk: specifically, when they introduce a way to preconfigure the white balance to suit your needs, as is possible with other camera brands.

[Cattoo]

13 minutes ago, don daniel said:

I think this is complete nonsense or pure marketing talk. The M11’s white balance isn’t just magenta-biased in the AWB setting, but also in the sun, shade, and cloudy settings, as well as in all manual settings with a specific Kelvin value. Of all the settings relevant to daylight, the only one free of a magenta bias is the immediate white balance on a neutral surface (e.g., a gray card). This phenomenon is not a deliberate decision by Leica, but rather a mistake they refuse to admit—and never will. That doesn’t mean they won’t fix it. They likely will, but then they’ll dress it up again with marketing talk: specifically, when they introduce a way to preconfigure the white balance to suit your needs, as is possible with other camera brands.

Why would Leica alert reviewers to this color bias before even releasing the camera to consumers if it was a mistake? Why wouldn’t they just fix it if they thought it important enough to cover up? 

i think it was intentional.  Perhaps not well executed, but intentional nonetheless.  

[charlesphoto99]

"The M9 colors are back..." Complete nonsense. The M11 looks nothing like the M9. Sure, it can be made to come close in post, but only if you know what you are doing (so can the 240, 10, 10-R).  

[fil-m]

I never "suffered" from the M11 purple tint since I generally use C1's auto adjust as a first action, and then manually tune the exposure and dynamic range settings until I reach the desired objective for the pictures that I like. But after reading this thread, I noticed that C1's auto adjust also plays with the "tint" setting and on most of my M11 images, it tends to automatically tune it with a green bias. I've never given much attention in the past to this "tint slider", but by resetting it to zero on many pictures, I am starting to notice a slight magenta tint. So I revert to C1’s choice.

Regarding the M9 (answering the above comments)... Honestly the "out of the box" color signature of the M11 and M9 have nothing in common, independently of the dynamic range which is way broader for the M11. I've resisted for so long the temptation of acquiring a M9 until I bought a used M9-P a few months ago. I can tell with full confidence that both RAWs and jpegs are different in terms of color rendering. Not that the M11 colors are inferior or superior, but different, very different. Maybe Leica's intention was to bring them back, but I can't see how they did. 

As a matter of fact, I don't like C1's auto adjust on my M9 images, I’ve stopped using this feature. I barely tune the M9 raw files except for exposure compensation and some shadow/highlight recovery, if any. I spend way more time adjusting the M11 files. As for JPEGs, I have a tendency to erase the M11 JPEGs as they bring nothing new to the table IMHO, keeping only my heavily adjusted M11 RAW files while for the M9, I keep both the RAW and JPEGs as each variant produces a different "feel".  

In any case and to make it short, yes I see the purple tint, it was hidden due to my usage of C1’s auto adjust with its tendency to neutralize it and make it disappear, and no, I don’t see at all how Leica believed that they brought back the M9 colors with the M11, in fact if anything, my wife immediately detects any pictures that I’ve taken with my M9 due to its… unique color rendering.

[Cattoo]

You’re missing the point, which is that well-executed or not, Leica did advise this reviewer even before the camera was released to the public that there’d be some color bias, an effort to mimic Kodachrome.  I also got the same line from my dealer.  To me that  indicates that there was some intentionality. 

[charlesphoto99]

6 minutes ago, Cattoo said:

You’re missing the point, which is that well-executed or not, Leica did advise this reviewer even before the camera was released to the public that there’d be some color bias, an effort to mimic Kodachrome.  I also got the same line from my dealer.  To me that  indicates that there was some intentionality. 

I'm sure the phoblographer is a really nice guy, but as far as photography goes he's just another GAS hack. Too much constant switching of gear, and not enough actual photography beyond drinking with friends or people walking with their cell phones to their faces. Nothing of substance there, or any of the other articles I've read by him. So hard to take anything serious he says or what Leica might have said to him. Sorry. I think it was just some PR hyperbole from a rep at Leica. 

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[fil-m]

7 minutes ago, Cattoo said:

You’re missing the point, which is that well-executed or not, Leica did advise this reviewer even before the camera was released to the public that there’d be some color bias, an effort to mimic Kodachrome.  I also got the same line from my dealer.  To me that  indicates that there was some intentionality. 

I see your point. Their intention could've been to mimic Kodak, and I can't argue with an intention. But the end result is not in line with the initial intention (IMO). And what matters at the end is the result, not the intention... 

[CDodkin]

3 hours ago, Cattoo said:

Or the fact that it’s a function of the AWB indicates that it is indeed a Leica choice.

 

Exactly the same magenta tint issue beset the original Sony censored 50MP Fujifilm GFX cameras - so they'd have had to be in on the same 'film look' idea - which seems highly unlikely - especially as they're Fujifilm, and wouldn't be trying to replicate a Kodak film look!!!

This kind of speculation just confuses the whole discussion - and is complete nonsense I'm afraid