The vast difference in out of the box SL2 and SL2S colors

[Stuart Richardson]

Advertisement (gone after registration)

I happened to be in the same spot with both cameras at the same time. Same lens, same exposure settings, same Adobe Standard profile, both set to Daylight.

It just reminded me of Freud's statement, "the narcissism of small differences". The general idea being that the closer things are to each other, the more hypersensitive we become to small differences. The colors are indeed slightly different, and I have noticed that the SL2S tends to be a bit more saturated out of the box, but I think that is probably because my standard SL2 profile is Cobalt Neutral, and the SL2S is still Adobe Standard for me (did not feel like buying another set of profiles!). The biggest difference I see in this photo is that the SL2S is a bit more magenta when both are set to daylight. It is most visible in the sky. But who are we kidding? They are so close! Adding 3 or 4 points of magenta and they are almost indistinguishable. It's actually surprising to me how close they are given totally different sensor manufacturers. Leica did a really good job at keeping things consistent. I don't usually shoot them directly side by side, but I thought it was funny how, especially when the SL2S came out, so many were talking about how much better its color was than the SL2. I did not really see a big difference at the time, and I don't see one now either. No offense to any who find it in their own pictures, but I honestly think it is mostly related to selection bias. Each camera is great and has its own strengths. Obviously the SL2 has much higher resolution, while the SL2S is better at high ISO. 

 

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!

Edited yesterday at 11:38 AM by Stuart Richardson

[jaapv]

It is indeed marginal but the second has a better tonal differentiation in the darker greens. Look at the wood on the hillside. However for the yellows it appears to be the other way around. But which one is which? I kinda think the first is the SL2. 

[Stuart Richardson]

Hmm. I don't really see it. I think it is maybe a slight difference in contrast. That might even be the light...not sure if a cloud went by. Or it could even be a slight difference from the yellow tone. I have not noticed a difference in DR at the capture stage to be honest, though I know there is supposed to be one. I do see it in having more push/pull flexibility though.

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!

[Stuart Richardson]

16 minutes ago, jaapv said:

It is indeed marginal but the second has a better tonal differentiation in the darker greens. Look at the wood on the hillside. However for the yellows it appears to be the other way around. But which one is which? I kinda think the first is the SL2. 

The first is the SL2S. It has a V number at the beginning, and the SL2 has an L number. 

[jaapv]

I’ll have to get back to my home computer with a decent Eizo screen. In any case, you are right. The difference is of no significance at all. 

[pgh]

Thanks for this.

Whatever difference there is, what I can see is that it would make absolutely makes no difference in my work itself, and I would be a fool to get hung up on it before purchase.

I wonder if it’s similar with the new X2d2 compared to X2D which has people opining about colors not being as good. For what purposes?

The only meaningful difference I’ve seen in color, really, is when I go back to images shot on my old CCD sensor cameras - Leica M8, Pentax 645D, etc. Still, even there, I’m hard pressed to say they have a meaningful impact on the final image. I’ve showed my wife photos, for example, asking if they had a different sort of rendering to them, and she’s like what are you talking about?

[Photoworks]

Advertisement (gone after registration)

The color temperature is warmer in the first one, but it could be the profile you are using.

[Stuart Richardson]

15 minutes ago, Photoworks said:

The color temperature is warmer in the first one, but it could be the profile you are using.

They were both shot at the same settings and then set to daylight in Lightroom, both using the Adobe Standard profile, so that they are consistent. The SL2 is a bit yellower when they are both set to daylight. This was not the "ideal" color balance for this photo to my eye, I just wanted it to be the same and there was no obvious 18% gray to click on in both photos to equalize it. 

As I stated in the beginning, there IS a difference. The point was that it was exceedingly minor. 

[Stuart Richardson]

1 hour ago, pgh said:

Thanks for this.

Whatever difference there is, what I can see is that it would make absolutely makes no difference in my work itself, and I would be a fool to get hung up on it before purchase.

I wonder if it’s similar with the new X2d2 compared to X2D which has people opining about colors not being as good. For what purposes?

The only meaningful difference I’ve seen in color, really, is when I go back to images shot on my old CCD sensor cameras - Leica M8, Pentax 645D, etc. Still, even there, I’m hard pressed to say they have a meaningful impact on the final image. I’ve showed my wife photos, for example, asking if they had a different sort of rendering to them, and she’s like what are you talking about?

I don't want to imply that there are no differences between cameras and their standard output. I think there definitely is. My point was that between the SL2 and SL2S, the differences are very minor. This is actually fantastic, as it allows them to be used interchangeable without any extra work to have them match. 

My favorite camera of all time for out of the box colors was the Leica S006 (and S2...color was the same). The M9 came next, then SL2/SL2S, then the Panasonic S1. Everything else I feel like I have to work a bit to get what I want out of them. I have not shot the X series Hasselblads, so I cannot comment on them. I really liked Camera Natural in the S1 though. It did have a gentle, natural feel that made it look great for the kind of photography I like to do (which is generally to represent natural colors faithfully). 

While most cameras can be profiled to achieve standard colors (like with an Macbeth chart), I think that is really only a halfway measure. I also don't think the standard Macbeth has enough color samples and enough saturation/contrast to fully achieve the kind of profiling that companies like Leica and Hasselblad (or Adobe and C1) seem to manage in their products. While they can be useful, they generally underperform in deep shadows, highlight gradation and colors that are almost out of gamut. I suspect that the practicality of achieving those extreme colors in a durable, low sheen paint is probably part of the challenge. I know that when making my own custom profiles for my printing studio, shadow gamut is artificially restricted by the fact that the i1 Profiler illuminates the patches...you can see that the bright light being shone on the blackest patches has a certain level of flare that artificially raises Dmax. You can even see it in profile viewers as compared to calculated profiles like Adobe RGB. I have experimented and found that the printer can achieve a better black using Adobe RGB vs custom profiling, and I think it is this measuring error that is causing it. In any case, I bring it up only to stress that out of camera results are important. Also because I suspect Jaap might come in and say that all cameras can be the same with a Macbeth profile, haha. And I disagree! I think it is a bit more complicated that just that. It also does not account for the spectral response, the default response curve and so on. It has its uses of course though. 

Edited yesterday at 02:07 PM by Stuart Richardson

[frame-it]

do u have a DSC-Lab chart in your studio?

 

Edited yesterday at 02:22 PM by frame-it

[AdjusterBrett]

no technician here - but considering the difference in the sensors, you would expect some differentiation from the amount of information acquired by each pixel.  I'll use a FUJI example to highlight this, as I lived through it.  At one point I had all the XT series.  Got the 2 first, then the 3, then 4, then 1, then 5.  Exceedingly disappointed with the 3 and 5.  Of the bunch - I kept the 4 (IBIS) and the 1 & 2.  1 has by far the best color, 2 is the compromise, and 4 is a modern camera even by today's standards.  I shot a pop can in a tree with shaded light with ALL five of them.  I shot the same pop can on a wooden post in sunlight, and then the can on a hillside a hundred feet away.  Shot with the XF90 (at the time Fujis sharpest lens - may still be).  Shot with MD 50/1.7 (my best colorfast), and with a thoriated TAK 50/1.4.  Did all this at the same park on the same day.  

Lightroom - unprocessed raw and SOOC Jpeg.  There was no questions.  1 had most details in the shadow, and by far the better color.  2 was good, 3 sucked, 4 was decent (balance, IBIS kept it around), and the 5 was simply a disappointment.  

NON-TECHNICAL, weekend hack - but that was how I culled the herd a bit (god knows I need more culling).   Same settings, same lens, same scene, then used my eyes.  Although I could go to town with the processing - it was the unprocessed that I judged upon.  I was rather proud - thinking I had become a PURIST - and then I watched a YouTube video (several - all DUMBED down to my level) and it was explained about the depth of the information each pixel acquires (fixed sensor size - less pixels = more information per pixel).  This does explain why I liked the output of the XT-1 & 2 more than the newer sensors.  

Perhaps (stated with trepidation) the smaller pixel count is exposing more information vs the larger, and that detail in the shadows could account for the color variation.  HOWEVER, you really gotta look to see a substantial difference.  Like to see the UNPROCESSED shots in B/W - it tends to give more 'detail' in the difference of the capture/rendering.

 

[jaapv]

1 hour ago, Stuart Richardson said:

I don't want to imply that there are no differences between cameras and their standard output. I think there definitely is. My point was that between the SL2 and SL2S, the differences are very minor. This is actually fantastic, as it allows them to be used interchangeable without any extra work to have them match. 

My favorite camera of all time for out of the box colors was the Leica S006 (and S2...color was the same). The M9 came next, then SL2/SL2S, then the Panasonic S1. Everything else I feel like I have to work a bit to get what I want out of them. I have not shot the X series Hasselblads, so I cannot comment on them. I really liked Camera Natural in the S1 though. It did have a gentle, natural feel that made it look great for the kind of photography I like to do (which is generally to represent natural colors faithfully). 

While most cameras can be profiled to achieve standard colors (like with an Macbeth chart), I think that is really only a halfway measure. I also don't think the standard Macbeth has enough color samples and enough saturation/contrast to fully achieve the kind of profiling that companies like Leica and Hasselblad (or Adobe and C1) seem to manage in their products. While they can be useful, they generally underperform in deep shadows, highlight gradation and colors that are almost out of gamut. I suspect that the practicality of achieving those extreme colors in a durable, low sheen paint is probably part of the challenge. I know that when making my own custom profiles for my printing studio, shadow gamut is artificially restricted by the fact that the i1 Profiler illuminates the patches...you can see that the bright light being shone on the blackest patches has a certain level of flare that artificially raises Dmax. You can even see it in profile viewers as compared to calculated profiles like Adobe RGB. I have experimented and found that the printer can achieve a better black using Adobe RGB vs custom profiling, and I think it is this measuring error that is causing it. In any case, I bring it up only to stress that out of camera results are important. Also because I suspect Jaap might come in and say that all cameras can be the same with a Macbeth profile, haha. And I disagree! I think it is a bit more complicated that just that. It also does not account for the spectral response, the default response curve and so on. It has its uses of course though. 

I am not going to say that. I have always said that the specification of the Bayer filter makes a difference. Not the sensor itself. That is a monochrome analog device. 

[Stuart Richardson]

1 hour ago, frame-it said:

do u have a DSC-Lab chart in your studio?

 

No, I have only used the standard Xrite/Macbeth Charts for camera calibration. I have used a Hutcheson chart for scanner calibration.

[Chris W]

To my eyes, very little difference indeed.

[Stuart Richardson]

1 hour ago, AdjusterBrett said:

no technician here - but considering the difference in the sensors, you would expect some differentiation from the amount of information acquired by each pixel.  I'll use a FUJI example to highlight this, as I lived through it.  At one point I had all the XT series.  Got the 2 first, then the 3, then 4, then 1, then 5.  Exceedingly disappointed with the 3 and 5.  Of the bunch - I kept the 4 (IBIS) and the 1 & 2.  1 has by far the best color, 2 is the compromise, and 4 is a modern camera even by today's standards.  I shot a pop can in a tree with shaded light with ALL five of them.  I shot the same pop can on a wooden post in sunlight, and then the can on a hillside a hundred feet away.  Shot with the XF90 (at the time Fujis sharpest lens - may still be).  Shot with MD 50/1.7 (my best colorfast), and with a thoriated TAK 50/1.4.  Did all this at the same park on the same day.  

Lightroom - unprocessed raw and SOOC Jpeg.  There was no questions.  1 had most details in the shadow, and by far the better color.  2 was good, 3 sucked, 4 was decent (balance, IBIS kept it around), and the 5 was simply a disappointment.  

NON-TECHNICAL, weekend hack - but that was how I culled the herd a bit (god knows I need more culling).   Same settings, same lens, same scene, then used my eyes.  Although I could go to town with the processing - it was the unprocessed that I judged upon.  I was rather proud - thinking I had become a PURIST - and then I watched a YouTube video (several - all DUMBED down to my level) and it was explained about the depth of the information each pixel acquires (fixed sensor size - less pixels = more information per pixel).  This does explain why I liked the output of the XT-1 & 2 more than the newer sensors.  

Perhaps (stated with trepidation) the smaller pixel count is exposing more information vs the larger, and that detail in the shadows could account for the color variation.  HOWEVER, you really gotta look to see a substantial difference.  Like to see the UNPROCESSED shots in B/W - it tends to give more 'detail' in the difference of the capture/rendering.

 

I think if anything you would expect the opposite. A larger pixel might have a better signal to noise ratio than a small pixel, but many smaller pixels will record more information (both tonal and chromatic). A larger number of megapixels means a larger number of total samples, which means more accurate tonal discrimination and superior tonality. The bayer demosaicing relies on "estimating" the surrounding colors...the more guesses it has, the more accurate the color (or at least, the more likely that color corresponds to what the color is in nature, even if the exact hue is slightly off). Since both of these are high end sensors shot at base ISO, I would expect that the quality of their pixel level sampling is very high. I should also state that I think at this level, you are not going to see dramatic differences in tonality between these sensors. 24mp is already a lot of pixels. The biggest improvement in moving to 47mp is in large prints, reduction of moire and in the finest details. I don't think it is something that is going to be as visible as, for example, the difference in tonality between 35mm and medium format film. 

Edited yesterday at 04:27 PM by Stuart Richardson

[jaapv]

I’m have seen no authoritative articles on the subject, but would think that the better noise behaviour and higher dynamic range of lower resolution sensors  rather suggests the opposite. 

[Stuart Richardson]

At least as far as I understand the term, tonality is the number of potential tonal variations in a given image. Just as 16 bit is a smoother curve than 8 bit, having twice as many samples (24 vs 47mp) gives you twice as many discrete tonal variations. I would agree with you it we are talking the extremes (the deepest shadow or brightest highlight, or at the highest ISOs), but within normal lighting situations and a properly exposed photo, I think one can expect better tonality from a higher resolution sensor. This has been my own visual experience at least....

But anyway, this is a bit of a digression. Honestly the main point of this post was to say that everything is great, hahaha. The SL2 series cameras are really just so well done, particularly if you are not someone who really lives and dies by the fastest AFc etc. 

[jaapv]

I have a strong suspicion that this is a subject that does not lend itself to simple explanations. 

[AdjusterBrett]

9 hours ago, jaapv said:

I have a strong suspicion that this is a subject that does not lend itself to simple explanations. 

Considering I BARELY understood the last couple responses - I'd second your thought.   HOWEVER - let me clarify what I mean by tonal because I'm likely using the term wrong.

50mm shot in whatever settings the body is on (matched between those I'm trying) not going into anything beyond the triangle and perhaps set W/B to the same card.  Then when I view on an LG Ultrafine (fairly decent, albeit aged) - and without blowing it up - I can see variance in the darker tones.  I will - in the coming days - do this between the SL and S1 on a vintage on something and post (In hotel tonight or work).  Then you guys can tell me what it is I'm seeing and why, as after watching the YouTubeVERSE I 'kinda' felt I had a handle on it, but now - NOPE...

[Advertisement]

Hallo,

Look here: Leica L-Mount

[phongph]

16 hours ago, Stuart Richardson said:

I happened to be in the same spot with both cameras at the same time. Same lens, same exposure settings, same Adobe Standard profile, both set to Daylight.

It just reminded me of Freud's statement, "the narcissism of small differences". The general idea being that the closer things are to each other, the more hypersensitive we become to small differences. The colors are indeed slightly different, and I have noticed that the SL2S tends to be a bit more saturated out of the box, but I think that is probably because my standard SL2 profile is Cobalt Neutral, and the SL2S is still Adobe Standard for me (did not feel like buying another set of profiles!). The biggest difference I see in this photo is that the SL2S is a bit more magenta when both are set to daylight. It is most visible in the sky. But who are we kidding? They are so close! Adding 3 or 4 points of magenta and they are almost indistinguishable. It's actually surprising to me how close they are given totally different sensor manufacturers. Leica did a really good job at keeping things consistent. I don't usually shoot them directly side by side, but I thought it was funny how, especially when the SL2S came out, so many were talking about how much better its color was than the SL2. I did not really see a big difference at the time, and I don't see one now either. No offense to any who find it in their own pictures, but I honestly think it is mostly related to selection bias. Each camera is great and has its own strengths. Obviously the SL2 has much higher resolution, while the SL2S is better at high ISO. 

 

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!

Dear Stuart Richardson!

If possible, please take the portrait from the SL2 vs SL2s with the same condition to see the difference of color, render of two pictures, eps. on the skintone.

Have a good day!

Many thanks!