Leica M11 -purplish tint ???

[lct]

Advertisement (gone after registration)

There is some magenta indeed but not in the sky. Below pic above with magenta saturation pushed by 100%

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

[Advertisement]

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

[BillCB]

I use Capture One and have never seen a magenta cast. I only shoot RAW.

Bill

[lct]

Seems like some good people here wear magenta eye glasses or have never used another raw converter than Adobe 😄

 

[bcaslis]

3 hours ago, lct said:

There is some magenta indeed but not in the sky. Below pic above with magenta saturation pushed by 100%

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

This was shot on a 35mm Summilux FLE 2 at f1.4, while I would not be surprised by the edges of the tree in the original DNG I don't see any magenta on the benches or the ground.

[Adam Bonn]

The second image is globally cooler than the first. This is often a desirable property because it’s easier to globally (or locally) warm an image up to what one wants to achieve than to cool it down.

C1 et el won’t be tied into the same WB metrics as adobe.

When a profile is created there’s not a limitless well of colour to play with, but a certain amount of R G  and B, often producing a desirable characteristic in one colour channel (or more accurately reducing an undesirable one) introduces problems in the other two.

And that’s what’s basically happening here within the Adobe ecosystem the magenta is effectively incorporated into the image at demosaicing ground zero and to profile it away completely is somewhere between a pipe dream and a fudge.

If the M11 were my camera, which despite the best efforts of the Leica store yesterday it isn’t (only had the R a year!) I’d probably fix it in post where ever possible.

For anyone desperately seeking an Adobe compatible profiling solution, I think I’d be tempted to try and create a profile LUT (not to be confused with HSD LUTs) that replicated tint adjustments applied in ACR/LR

[don daniel]

vor 8 Stunden schrieb lct:

There is some magenta indeed but not in the sky. Below pic above with magenta saturation pushed by 100%

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Please excuse me, but you obviously don't see the problem. You can't prove or falsify a magenta cast in the image by making the color magenta more saturated. In white balance, the camera tries to find a balance in temperature between blue and yellow and in tint between green and magenta. If the tint is shifted too much towards green, all colors are affected. This is also the case with a shift towards magenta, as with the M11. The blue of the sky has a magenta tint, the other colors are also affected. I can see it with my eyes without having to measure it. But it is also measurable. However, it cannot be solved by saturating or desaturating magenta, but only by moving the tint slider in the white balance tool: tint here surely shows more than +20 in Adobe Lightroom if the source image is a DNG. If you move this slider to the left to +10, the result should look much more realistic and therefore much better for me as a starting point or even as a finished picture. But you need the DNG for this and can't take a JPG posted here in the forum and show anything by manipulating individual colors.

In the case of a JPG from the camera, the sliders in the tool for the white balance in Lightroom are set to 0. If you move the tint slider to -7 or -8, the magenta cast also disappears in the JPG. However, this affects the overall image quality, which is a great pity. In the new firmware, Leica has made an attempt to optimize the JPG quality by improving the dynamic range in the shadows. So Leica is also aiming to produce images that can be used directly and shared via the Fotos app, for example. For me, however, this is simply made impossible by the fact that the white balance cannot produce a result without a magenta cast under completely normal conditions, neither in the automatic setting nor in the fixed settings for daylight. 

 

[lct]

Advertisement (gone after registration)

1 hour ago, don daniel said:

[...]The blue of the sky has a magenta tint[...]

I disagree. If such a tint existed you would see it obviously when oversaturating magentas in PP. Seems like some confusions arise between magenta and cyan here.

Edited November 25, 2023 by lct

[lct]

Here is how a magenta tint looks like when oversaturating magenta in PP . By 80% on Adobe converter below.

 

[mosh1]

 

4 hours ago, lct said:

I disagree. If such a tint existed you would see it obviously when oversaturating magentas in PP. Seems like some confusions arise between magenta and cyan here.

Saturating magenta via a color slider only affects the magenta within a small narrow range of hues the slider targets. What everyone else is talking about is a magenta *bias* in white balance, which I agree with Don Daniel is often problematic with the M11. 

Edited November 25, 2023 by mosh1

[lct]

20 minutes ago, mosh1 said:

 

Saturating magenta via a color slider only affects the magenta within a small narrow range of hues the slider targets. What everyone else is talking about is a magenta *bias* in white balance, which I agree with Don Daniel is often problematic with the M11. 

What you call narrow range does not look that narrow on my (admittedly rare) Adobe pics but anyway, here is my last test for today. Not an easy one with a lens reputed for pinkish rendition and +100 magenta saturation. Bit of magenta in tree branches but none in the sky as expected. M11, Summicron 35/2 asph v1, auto WB, DNG converted to TIF with Silkypix.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

 

[jaapv]

7 hours ago, Adam Bonn said:

The second image is globally cooler than the first. This is often a desirable property because it’s easier to globally (or locally) warm an image up to what one wants to achieve than to cool it down.

C1 et el won’t be tied into the same WB metrics as adobe.

When a profile is created there’s not a limitless well of colour to play with, but a certain amount of R G  and B, often producing a desirable characteristic in one colour channel (or more accurately reducing an undesirable one) introduces problems in the other two.

And that’s what’s basically happening here within the Adobe ecosystem the magenta is effectively incorporated into the image at demosaicing ground zero and to profile it away completely is somewhere between a pipe dream and a fudge.

If the M11 were my camera, which despite the best efforts of the Leica store yesterday it isn’t (only had the R a year!) I’d probably fix it in post where ever possible.

For anyone desperately seeking an Adobe compatible profiling solution, I think I’d be tempted to try and create a profile LUT (not to be confused with HSD LUTs) that replicated tint adjustments applied in ACR/LR

Spoiler: If you shift to *L*A*B you have far more control over colour and do not run into the limitations that RGB poses by linking luminosity to colour. 

[Adam Bonn]

42 minutes ago, jaapv said:

Spoiler: If you shift to *L*A*B you have far more control over colour and do not run into the limitations that RGB poses by linking luminosity to colour. 

This doesn’t change the fact that adobe demosaics into CIE XYZ D50 (this happens way before there’s anything to display on your screen)

https://en.wikipedia.org/wiki/CIE_1931_color_space

from the DNG spec (written by adobe)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

https://helpx.adobe.com/content/dam/help/en/photoshop/pdf/DNG_Spec_1_7_0_0.pdf#page97

Edited November 25, 2023 by Adam Bonn

[marchyman]

15 hours ago, lct said:

Seems like some good people here wear magenta eye glasses or have never used another raw converter than Adobe 😄

 

And others ignore that straight out of camera jpegs never touched by Adobe software may also show a magenta tint.  Or are you claiming Adobe supplied the in-camera code to Leica?

[jaapv]

No - but I do know that Adobe and Leica cooperate in the design phase. However, when you open a JPG file in an Adobe product, it IS done by Adobe software,  

[jaapv]

1 hour ago, Adam Bonn said:

This doesn’t change the fact that adobe demosaics into CIE XYZ D50 (this happens way before there’s anything to display on your screen)

https://en.wikipedia.org/wiki/CIE_1931_color_space

from the DNG spec (written by adobe)

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

https://helpx.adobe.com/content/dam/help/en/photoshop/pdf/DNG_Spec_1_7_0_0.pdf#page97

Sure, but I was talking about editing the colour shift in a later phase, 

[Adam Bonn]

44 minutes ago, jaapv said:

Sure, but I was talking about editing the colour shift in a later phase, 

IMHO that's the advice folks need, tbh all that DNG spec stuff is basically aimed at camera manufacturers and post processing app designers.

I would imagine that if one would approach adobe and say I'm making a camera and I want it to shoot DNG they send that doc and say off you go... possibly you pay money and they hold your hand a little... I'm not privy the commercial mechanics of it. (DNG is technically open source...)

[elmars]

vor 12 Stunden schrieb jaapv:

No - but I do know that Adobe and Leica cooperate in the design phase. However, when you open a JPG file in an Adobe product, it IS done by Adobe software,  

Jpg look the same everywhere. Only the monitor calibration matters. Lightroom does nothing with jpg. 

[jaapv]

12 hours ago, Adam Bonn said:

IMHO that's the advice folks need, tbh all that DNG spec stuff is basically aimed at camera manufacturers and post processing app designers.

I would imagine that if one would approach adobe and say I'm making a camera and I want it to shoot DNG they send that doc and say off you go... possibly you pay money and they hold your hand a little... I'm not privy the commercial mechanics of it. (DNG is technically open source...)

The indicator is the software that is bundled with the camera. One hand washes the other.  

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[don daniel]

You can open and view DNG and JPG files from the M11 in Adobe programs, you can open and view DNG and JPG files from the M11 in Capture One and you can open and view DNG and JPG files from the M11 in Apple Photos: The values displayed in the white balance tools of the programs may differ. But the images all have the same magenta cast that the incorrect white balance of the M11 gives them. I have tried everything. It remains the same: I have to correct the white balance a lot for almost all images from the M11. I can't get realistic colors any other way. Now it's really up to Leica to create a way for us to get DNGs and JPGs from the camera that don't differ so drastically from everything we know from other Leica cameras and from other manufacturers.

[Adam Bonn]

1 hour ago, jaapv said:

The indicator is the software that is bundled with the camera. One hand washes the other.  

Yeah more or less… I suspect that these commercial partnerships are more nuanced than that… Didn’t the M8 come with C1? Then for a while you got whatever the current version of LR was, then you got some free months of CC (which many places offer as a deal, eg Flickr), also C1 supports SL tethering (which implies a degree of collaboration)

The DNG is open source, the spec and the SDK is available for anyone (I would theorise that this is the cheapest option for a brand, why design your own RAW format when you can get an oven ready one)

Also Leica’s implementation of the DNG isn’t really to the letter of the (adobe) law… well maybe that’s harsh.. more like if an instruction in the spec reads like we recommend this, but at a pinch fk it that’ll do (for clarity that’s not verbatim 😅) Leica tend to select the latter (or at least used too with the M range)

Plus some other weird stuff.. the DNG of the M9 has a tag telling adobe that it has an anti-aliasing filter 🤷and aperture values are recorded in maker notes rather than the recommended place.

If adobe are giving Leica technical support about the implementation of the DNG spec, Leica aren’t really paying that much attention.

This is with the massive caveat that I haven’t delved into a DNG since the M10 and I’ve only ever looked at M DNGs. Maybe it’s all different now, or done differently on the Q/SL etc

Edited November 26, 2023 by Adam Bonn