Does “Color Science” even matter?

[JimKasson]

Seems like its a simple tool that can be used before every photoshoot for a given lighting condition to get accurate colors in post. Would this tool be accurate enough to match the colors on my Sony and Fuji?

Depends on how picky you are. There will be measurable differences.

Is there another tool that would match the colors more accurately?

You can match a small number of patches closely. But differences in sensor spectral sensitivities may keep you from matching all the patches you want to match. For example, what do you do if one camera sees two different colored patches as the same, but another sees them as different? Force both to map the two different patches to the same color?

Metameric failure - the last word

This is the eighth in a series of posts on color reproduction. The series starts here. I said in the last post that I’d get to 3D LUT based camera profiles in this one. I lied. Sorry, but the story of this series on color reproduction, which started out as an attempt to explain how

blog.kasson.com

 

[Dan Cook]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

 

[paratom]

to me it matters a lot how color comes out of camera and raw converter.

for example from x1d/x2d or my Leicas it comes much to my liking without having to spend much time in post processing.

Canon looks fine to me, with Nikon I sometimes struggled with some kind of yellowish tint.

How were you doing white balance?

loan a x2d and shoot some portraits and you know what I mean.

I usually use either auto or daylight. Since I allways shoot raw + jpg its easy to adjust wb later in case it doesnt look right. I am not talking about wb differences, more about overall color, given that wb is ok.

How can you have an overall yellowish tint if the WB is right?

Its not yellowish for all colors, but just for certain ones. If you manual white balance a Nikon and. a Canon and a Leica image, white will look white in all three images, but face color will look still different often. Canon will have a little more magenta, nikon a little more yellow, and of course it also depends on the raw converter and profiles.

 

[paratom]

Seems like its a simple tool that can be used before every photoshoot for a given lighting condition to get accurate colors in post. Would this tool be accurate enough to match the colors on my Sony and Fuji?

I have used it in the past. it doesnt do wonders.

 

[paratom]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

I second the high quality of x1d/x2d color output.

 

[BackToNature1]

Yes, why spend tons of time in PP just to adjust the color science if one does not have to.

 

[spilla]

Found this interesting video on color science.

https://fstoppers.com/education/wha...hould-it-actually-matter-photographers-298843

I didnt know Sony colors were accurate and that Canon colors were pleasing but inaccurate. I agree with the conclusion that one should choose a camera based on features and the workflow and color science shouldnt be a deciding factor. I would add the quality of glass as a deciding factor as well. After all the lens is the “eye” of the camera and an imperfect eye cant create a perfect vision.

What do you folks think about color science?

I’m for color science, if for no other reason than it got me paid well for 6 years.

Yes, but in todays age of digital cameras and RAW processors would you have chosen color science as a top priority over other aspects of choosing a system?

I believe you and Jim have a different concept of "color science." You did not answer my question about what you mean by "color science," but I assume you mean the initial look in the post-processor (which depends on the post-processor used, profiles, in-camera WB, the camera's firmware, etc.).

Jim probably means science when he talks about "color science."

If we're going to get technical, I consider color science a branch of psychology, and what most people call color science, I'd like people to call color engineering. In that regard, I'm with Jim King of Adobe, who passed out buttons with "I'm a color engineer" written on them at a SPIE conference. I still have mine around somewhere.

But I've lost that battle, and except when I'm being pedantic, I will use color science for both.

 

[ggbutcher]

Found this interesting video on color science.

https://fstoppers.com/education/wha...hould-it-actually-matter-photographers-298843

I didnt know Sony colors were accurate and that Canon colors were pleasing but inaccurate. I agree with the conclusion that one should choose a camera based on features and the workflow and color science shouldnt be a deciding factor. I would add the quality of glass as a deciding factor as well. After all the lens is the “eye” of the camera and an imperfect eye cant create a perfect vision.

What do you folks think about color science?

I’m for color science, if for no other reason than it got me paid well for 6 years.

Yes, but in todays age of digital cameras and RAW processors would you have chosen color science as a top priority over other aspects of choosing a system?

I believe you and Jim have a different concept of "color science." You did not answer my question about what you mean by "color science," but I assume you mean the initial look in the post-processor (which depends on the post-processor used, profiles, in-camera WB, the camera's firmware, etc.).

Jim probably means science when he talks about "color science."

If we're going to get technical, I consider color science a branch of psychology, and what most people call color science, I'd like people to call color engineering. In that regard, I'm with Jim King of Adobe, who passed out buttons with "I'm a color engineer" written on them at a SPIE conference. I still have mine around somewhere.

But I've lost that battle, and except when I'm being pedantic, I will use color science for both.

It seems to me that a conceptual model would be really helpful here, showing the relationship between all the factors contributing to color science/engineering. Jim is there something like this out there?

Indeed there is:

https://blog.kasson.com/the-last-word/the-color-reproduction-problem/

Just in case Jim's too modest...

I'm not a Licensed Color Scientist, more of a "color mechanic". I cut my teeth in that regard reading Jim's posts, and the articles on Elle Stone's site, ninedegrees.com

There is a need for more "mechanical" literature. Some time back I tried to write one, you can peruse it here:

https://discuss.pixls.us/t/article-color-management-in-raw-processing/11521

HTH

 

[A127943]

What do you folks think about color science?

I think if the check clears, who cares.

I understand why really smart people fret, worry, and work extremely hard on this stuff, but at the end of the day, all but the absolute rarest of clients have ever given a crap about the color science. It's stuff for nerds, but not stuff that matters to what matters in the end.

I have no preference for whose color science I work with. I only think about it when processing RAW photos and then it's simply a matter of aesthetics, which defeats the point of the science anyways.

 

[JimKasson]

Found this interesting video on color science.

https://fstoppers.com/education/wha...hould-it-actually-matter-photographers-298843

I didnt know Sony colors were accurate and that Canon colors were pleasing but inaccurate. I agree with the conclusion that one should choose a camera based on features and the workflow and color science shouldnt be a deciding factor. I would add the quality of glass as a deciding factor as well. After all the lens is the “eye” of the camera and an imperfect eye cant create a perfect vision.

What do you folks think about color science?

I’m for color science, if for no other reason than it got me paid well for 6 years.

Yes, but in todays age of digital cameras and RAW processors would you have chosen color science as a top priority over other aspects of choosing a system?

I believe you and Jim have a different concept of "color science." You did not answer my question about what you mean by "color science," but I assume you mean the initial look in the post-processor (which depends on the post-processor used, profiles, in-camera WB, the camera's firmware, etc.).

Jim probably means science when he talks about "color science."

If we're going to get technical, I consider color science a branch of psychology, and what most people call color science, I'd like people to call color engineering. In that regard, I'm with Jim King of Adobe, who passed out buttons with "I'm a color engineer" written on them at a SPIE conference. I still have mine around somewhere.

But I've lost that battle, and except when I'm being pedantic, I will use color science for both.

It seems to me that a conceptual model would be really helpful here, showing the relationship between all the factors contributing to color science/engineering. Jim is there something like this out there?

You could start here:

The color reproduction problem - the last word

This is the first in a series of 33 posts attempting to deal with the question, “Do some cameras have better color than others?” The best way to navigate to all the other parts is to scroll down to the bottom of this page and you'll see permalinks to all the other posts in this

blog.kasson.com

If you want to dig deeper, I recommend starting with Hunt’s Measuring Color, now In its fourth or fifth edition.

 

[JimKasson]

to me it matters a lot how color comes out of camera and raw converter.

for example from x1d/x2d or my Leicas it comes much to my liking without having to spend much time in post processing.

Canon looks fine to me, with Nikon I sometimes struggled with some kind of yellowish tint.

How were you doing white balance?

loan a x2d and shoot some portraits and you know what I mean.

I usually use either auto or daylight. Since I allways shoot raw + jpg its easy to adjust wb later in case it doesnt look right. I am not talking about wb differences, more about overall color, given that wb is ok.

How can you have an overall yellowish tint if the WB is right?

Its not yellowish for all colors, but just for certain ones. If you manual white balance a Nikon and. a Canon and a Leica image, white will look white in all three images, but face color will look still different often. Canon will have a little more magenta, nikon a little more yellow, and of course it also depends on the raw converter and profiles.

It is common in commercial color profiles to bias Caucasian skin towards red or yellow or both. Magenta, not so much. Green, almost never.

--
https://blog.kasson.com

 

[Mads Bjerke]

Matching the "look" from one camera to another is much more complicated than just using a colour target.

Color targets will only match the colour at a certain luminosity and doesn't account for the variations in colour response as the tones get darker or brighter.

Contrast has a huge impact on the perceived colour and matching that between camera profiles can be very complicated.

The best tool I have at my disposal for adjusting camera profiles is 3dLUT Creator.
It features a tool set that can get very granular with with the colour and contrast at various luminosity levels. It can also auto-match colour patches and generate very accurate curve adjustments to balance the colour.
That said, it is still not truly simple to match two different camera profiles even with this software.

What we perceive as the colour science is the "look" that the RAW convertor applies to the RAW file as it is developed via the camera profile.
A camera profile is also known as a Lookup Table (LUT).

The RAW software will apply the LUT and perhaps also apply its own algorithm to further adjust the "look".

For this reason there are very big differences between the starting point from one software to another.

Generally the camera manufactures own software will have some extra secret sauce as they have first hand knowledge to the way they designed their own profiles.

I have been trying to match the Canon DPP "look" in Lightroom and it is nearly impossible to completely recreate. The very subtle differences in colour and contrast curves are too complex to match.

At the end of the day we have to choose a balance between the "look" the software delivers and the overall convenience of the other tools it offers.
I cannot work in Canon DPP as it is incredibly slow and the general tools are not very sophisticated. It does, however, deliver the better starting point.

One option is to export a TIFF from your OEM software and continue working on that TIFF in C1, LR, PS, DxO etc as preferred.
That is fine if you are only working on a few images, but not great if you need to process lots of images in a short timeframe.

I agree with those who say it is more complex than simply adjust WB, curves and HSL to match the look.

 

[DavidMillier]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

But if we are talking about raw files, you are still talking about the difference in the raw software, not the camera/sensor. It's possible there are some subtle differences in the CFA on different sensors (that special Phase One camera for example where they concentrated on tweaking the CFA dyes) but mostly the colour is determined by the raw software. As some people have said, you don't have to use the canned camera profiles, you can make your own custom profiles and tweak the colours to be anything you want.

 

[Canadianguy]

Found this interesting video on color science.

https://fstoppers.com/education/wha...hould-it-actually-matter-photographers-298843

I didnt know Sony colors were accurate and that Canon colors were pleasing but inaccurate. I agree with the conclusion that one should choose a camera based on features and the workflow and color science shouldnt be a deciding factor. I would add the quality of glass as a deciding factor as well. After all the lens is the “eye” of the camera and an imperfect eye cant create a perfect vision.

What do you folks think about color science?

Who wants their photographs to be identified by the manufacturer of their camera?

Do you identify Mark Seliger's Oscar photos by the brand of camera or the brand of lights he used? Ok - the manufacturers would like that but you identify those photos as Mark Seliger's - or at least I hope so.

I want the photos I take to be identified by the viewer to be linked to me and not by the brand of camera I used.

Oh - look at those colours in the photo - it must be a Hasselblad or a Fujifilm MF camera - if a viewer said that about my photos - I have failed as a photographer.

Colour is a tool you have in your tool bag - if you don't want to play with it and just leave it to the camera manufacturers - save yourself time & money and shoot with your cell phone.

 

[yogi4fitness]

Found this interesting video on color science.

https://fstoppers.com/education/wha...hould-it-actually-matter-photographers-298843

I didnt know Sony colors were accurate and that Canon colors were pleasing but inaccurate. I agree with the conclusion that one should choose a camera based on features and the workflow and color science shouldnt be a deciding factor. I would add the quality of glass as a deciding factor as well. After all the lens is the “eye” of the camera and an imperfect eye cant create a perfect vision.

What do you folks think about color science?

Who wants their photographs to be identified by the manufacturer of their camera?

Do you identify Mark Seliger's Oscar photos by the brand of camera or the brand of lights he used? Ok - the manufacturers would like that but you identify those photos as Mark Seliger's - or at least I hope so.

I want the photos I take to be identified by the viewer to be linked to me and not by the brand of camera I used.

Oh - look at those colours in the photo - it must be a Hasselblad or a Fujifilm MF camera - if a viewer said that about my photos - I have failed as a photographer.

Very good point. Never thought this way. I agree if people can identify my photos by the camera used I am not a good photographer

Colour is a tool you have in your tool bag - if you don't want to play with it and just leave it to the camera manufacturers - save yourself time & money and shoot with your cell phone.

 

[paratom]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

But if we are talking about raw files, you are still talking about the difference in the raw software, not the camera/sensor. It's possible there are some subtle differences in the CFA on different sensors (that special Phase One camera for example where they concentrated on tweaking the CFA dyes) but mostly the colour is determined by the raw software. As some people have said, you don't have to use the canned camera profiles, you can make your own custom profiles and tweak the colours to be anything you want.

I think its wishfull thinking that you can tweak a raw file from any camera to any (color) look.

For example if you use a camera with 16-bit color depth instead of 14, you get 64 times more color tones.

if you feed raw files from different brands cameras to lightroom, the output will still look somewhat different. You can spend hours to try to make them look similar. Sometimes one will be successfull, sometimes not.

 

[JimKasson]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

But if we are talking about raw files, you are still talking about the difference in the raw software, not the camera/sensor. It's possible there are some subtle differences in the CFA on different sensors (that special Phase One camera for example where they concentrated on tweaking the CFA dyes) but mostly the colour is determined by the raw software. As some people have said, you don't have to use the canned camera profiles, you can make your own custom profiles and tweak the colours to be anything you want.

I think its wishfull thinking that you can tweak a raw file from any camera to any (color) look.

For example if you use a camera with 16-bit color depth instead of 14, you get 64 times more color tones.

I have found negligible difference between the colors in GFX 100x 16 bit and 14 bit files, except with photographically ridiculous pushes, and that can be calibrated out.

if you feed raw files from different brands cameras to lightroom, the output will still look somewhat different. You can spend hours to try to make them look similar. Sometimes one will be successfull, sometimes not.

 

[ggbutcher]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

But if we are talking about raw files, you are still talking about the difference in the raw software, not the camera/sensor. It's possible there are some subtle differences in the CFA on different sensors (that special Phase One camera for example where they concentrated on tweaking the CFA dyes) but mostly the colour is determined by the raw software. As some people have said, you don't have to use the canned camera profiles, you can make your own custom profiles and tweak the colours to be anything you want.

I think its wishfull thinking that you can tweak a raw file from any camera to any (color) look.

For example if you use a camera with 16-bit color depth instead of 14, you get 64 times more color tones.

if you feed raw files from different brands cameras to lightroom, the output will still look somewhat different. You can spend hours to try to make them look similar. Sometimes one will be successfull, sometimes not.

Davinci Resolv and 3DLUTCreator both have LUT-based tools to do this with ColorChecker target shots. Yeah, tweaking color knobs in LR or any software would be tedious and fiddly...

 

[Erik Kaffehr]

I bought a pair Sony A9 for all their technical capability when I made a career change into full time photography.

I fought it's raw colour output for years creating preset after preset, until I found a third party colour profile which helped, but not long after..

..I bought a pair of Sony A1's and except for a colour inconsistency between 70-200 & 24-70, I am happy.

But the first time I loaded up a Hasselblad X1D raw portrait, I was amazed by the colour presented to me, so much so it made it difficult to know where to start processing because I felt like I can only make it worse. I've since had X1D II, 907X and the X2D - just love the raw.

And now, even if I'm going to change colours considerably, I still prefer having a great base to work from.

But if we are talking about raw files, you are still talking about the difference in the raw software, not the camera/sensor. It's possible there are some subtle differences in the CFA on different sensors (that special Phase One camera for example where they concentrated on tweaking the CFA dyes) but mostly the colour is determined by the raw software. As some people have said, you don't have to use the canned camera profiles, you can make your own custom profiles and tweak the colours to be anything you want.

I think its wishfull thinking that you can tweak a raw file from any camera to any (color) look.

No, I don't think it is so, in general.

For example if you use a camera with 16-bit color depth instead of 14, you get 64 times more color tones.

Well, that is pure nonsense. Cameras cannot differentiate 64 000 colors per channel, the real number may be closer to 250 colors per channels.

if you feed raw files from different brands cameras to lightroom, the output will still look somewhat different. You can spend hours to try to make them look similar. Sometimes one will be successfull, sometimes not.

That may depend on the color profiles used. The examples below compare Pentax 645Z (left) and Sony A7rIV (right) to the GFX 50S, based on DPReviews studio test scene, using profiles generated by Lumariver Profile Designer. The colors are virtually identical between the three.


No visible differences comparing GFX 50S/Pentax 645Z or GFX50S/Sony A7rIV


The numbers tell the same story.

Comparing same cameras, Hasselad X1D added, with reference data indicates significant errors:


From left: Pentax 645Z, Fujifilm GFX 50S, Hasselblad X1D, Sony A7rIV

So, all sensors have some reproduction errors, but they all seem to have similar reproduction errors.

Different Sony sensors - a simple comparison of color

Hi, Geek warning. This is geek stuff! A recent discussion was about who makes the sensor for the GFX 100. The answer is well known. The next question is how much sensors made by Sony differ? Here, I think that DPReview test charts can provide some good answers. In this comparison I used my...

www.dpreview.com

Best regards

Erik

--
Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles

 

[Erik Kaffehr]

Found this interesting video on color science.

https://fstoppers.com/education/wha...hould-it-actually-matter-photographers-298843

I didnt know Sony colors were accurate and that Canon colors were pleasing but inaccurate. I agree with the conclusion that one should choose a camera based on features and the workflow and color science shouldnt be a deciding factor. I would add the quality of glass as a deciding factor as well. After all the lens is the “eye” of the camera and an imperfect eye cant create a perfect vision.

What do you folks think about color science?

That video covers a lot of ground and is pretty good introduction to the issues involved.

We need to differentiate raw images, those don't have color. Color is applied in 'raw development' with the color response of the sensor presumed to be known. Raw development deduces color from raw data and applies a vendor or profile dependent interpretation to that deduced color.

Cameras can do raw conversion internally into JPEG files, that conversion is camera specific.

Raw conversion applies a lot of correction using tone curves and LUTs to the deduced color.

Best regards

Erik