Wide Gamut Monitor and photo editing question

[george013]

George, knowing your love of diagrams, I have created a couple to try to explain visually what I cannot seem to get through to you in words. I have created a fictitious slice of part of a 2-D representation of a couple of colour spaces. The boundaries happen to be straight lines and the colours are what they are. What I am trying to demonstrate is why bit depth matters (more) in a wide gamut colour space when it comes to real life editing issues. Ted, this has nothing to do with all the various colour spaces you brought up in your questions, because it really doesn't matter, and no, it is not a perfect analogy to the real world issue I am trying to describe graphically.

In the first diagram I have created two "colour spaces". The first one is a "narrow gamut colour space" and the area shows a limited range of colours and colour depth. Each of the squares in a 6 x 6 (which represents the bit depth) is more or less a contiguous colour and when I look at it visually each square is more or less the same colour.

When I apply the same 6 x 6 grid to a "wide gamut colour space" to the larger square the squares are no longer a relatively pure colour. This is especially noticeable when we see where the green and red colours start to interplay with each other. This is the real world issue I have been trying to explain; a small bit depth is not granular enough to represent the individual colours in a wide gamut colour space and we run into technical issues when manipulating the data that results in artifacts.






The second diagram shows the same wide gamut "colour space" that I used in the first diagram. The square on the left is identical to the one on the right hand side of the first diagram. The square on the right is identical to the one on the left except that rather than a 6 x 6 grid, I have overlaid a 24 x 24 grid. The right hand grid represents what happens when we increase bit depth and the amount of data we have increases. The colour in each individual grid element is much more consistent than what we see in the 6 x 6 grid. Increasing the bit depth gives us more granularity, which in turn lets us define the colours more precisely. This mirrors what happens in the real world when we use a higher bit depth; we get little to no artifacts when we use a higher bit depth when using a wide gamut colour space.

If bit depth does not matter, think about cutting the 8-bits in half and try to describe all the colours using a 4-bit definition. With 4 bits, we get 2 x 2 x 2 x 2 = 16 bits per channel; or with 3 channels; 16 x 3 = 48 colours. So bit depth does matter...

I know what is meant with color space and bit depth. For some reason you don't want to understand what I write. There is no relation between bit depth and color space. They're 2 completely separate items. That's one point I'm shouting out.

The other point is why working in a color space the monitor can't handle. That I still don't understand. In my simple thinking you should work in the color space the monitor has. With color management I try to produce colors on the monitor as close as possible to the real world colors. What I see are analogue colors. My digital image doesn't have colors, it has digital values with which I control the analogue output. If the ouput of my screen is different as what is expected on creation of the digital image, than the pixel values must be corrected. My screen is the demanding object. That's what I think is color management.
So why using a wider gamut and "correcting" the pixel values in a way they can't produce the right colors??? That question hasn't been answered yet.

George

[DanK]

The other point is why working in a color space the monitor can't handle. That I still don't understand.

Manfred explained this. To put in different terms, in a digital environment, we are working with step functions but ideally perceiving them as continuous functions. The smaller the color depth, the larger the steps, and hence the greater the likelihood artifacts--such as posterization, which is a case in which the steps are huge and perceptible.

[george013]

Manfred explained this. To put in different terms, in a digital environment, we are working with step functions but ideally perceiving them as continuous functions. The smaller the color depth, the larger the steps, and hence the greater the likelihood artifacts--such as posterization, which is a case in which the steps are huge and perceptible.

Sorry Dan, not personal meant but I think before the end of the year I've quoted this 100 times.

There is no relation between bit depth and color space. They're 2 completely separate items. That's one point I'm shouting out.

And the second one.

So why using a wider gamut and "correcting" the pixel values in a way they can't produce the right colors??? That question hasn't been answered yet.

George

[DanK]

Sorry Dan, not personal meant but I think before the end of the year I've quoted this 100 times.


And the second one.


George

Read this: http://schewephoto.com/sRGB-VS-PPRGB/. Note that his comparison does not involve bit depth, only the width of the gamut. His sRGB files are 16-bit TIFFs, so they don't involve the reduction to 8-bit that JPEGs entail. It illustrates the principle that using a wider gamut monitor lessens the risk of editing artifacts.

[Manfred M]

I know what is meant with color space and bit depth. For some reason you don't want to understand what I write. There is no relation between bit depth and color space. They're 2 completely separate items. That's one point I'm shouting out.

In that case George, I would like you to demonstrate how we can use 1-bit, 2-bit, 3-bit, 4-bit, etc. representations for any typical colour space we would use in photography. None of these would give what any of us would perceive as giving satisfactory results. A 4-bit colour gives us a total of 64 unique colours and that is simply not enough to create a natural looking image. The question is really determining the minimum bit depth to be able to create an acceptable image.

I think what you are saying is that an 8-bit colour space is more than sufficient to describe a typical photographic colour space like sRGB, AdobeRGB and ProColor RGB. Here I would be ready to agree with you if we left the editing step out, because all of these can look good as is.

How often do we read JPEG files cannot be edited because this process will result in artifacts. This statement is also not quite true as I edit a lot of JPEG files and they look just fine afterwards; BUT if the data is pushed too hard during the edit, we can see artifacts. If these are low quality jpegs, some of the problem is related to what the compression algorithms have done to the data, but part of the problem is that they are using 8-bit data and that does leave a limit of the number of colours that can be created / modified by the editing software. A similar issue exists when using a wide-gamut colour space like ProPhoto. We need enough data to work with in our editing software to ensure that we do not create artifacts. This is the reason that the common recommendation is that ProPhoto should use 16-bit data.

The other point is why working in a color space the monitor can't handle. That I still don't understand. In my simple thinking you should work in the color space the monitor has. With color management I try to produce colors on the monitor as close as possible to the real world colors. What I see are analogue colors. My digital image doesn't have colors, it has digital values with which I control the analogue output. If the ouput of my screen is different as what is expected on creation of the digital image, than the pixel values must be corrected. My screen is the demanding object. That's what I think is color management.
So why using a wider gamut and "correcting" the pixel values in a way they can't produce the right colors??? That question hasn't been answered yet.

A modern camera captures a wider gamut than what even a wide-gamut AdobeRGB compliant screen can capture. While our screens cannot reproduce all of the colours, modern high-end ink jet photo printers can produce more than our screens can show. One reason to work in a wider colour space is that these colours can be printed, so why would we not do so? We may not be able to see them on our screen, but we certainly will see them in our prints.

There is a second argument that is rather subtle and I understand that is why Adobe chose a variant of the ProPhoto colour space as the default colour space in its Lightroom application. If we decide to use a narrower colour space, then the system's Colour Management Module (CMM), recalculates all the colours so they fit within the new colour space. This means any manipulations during the edit process are based on the "less accurate" colours of the narrower gamut. If we work with the image data in a wide gamut colour space, the data manipulation occurs in the wider colour space and gives more accurate colours that are manipulated near the edges of the display's colour space.

Unfortunately it's not easy to explain, but it is consistent with the principle that data should be retained as long as possible and only disposed of at the final step.

[george013]

There is no relation between bit depth and color space. They're 2 completely separate items. That's one point I'm shouting out.

I'll make the hundred before the end of the year.
The color space is the ability to produce colors within a certain range. These colors are analogue.
Bit depth is the result of an A/D conversion, just a mathematical digital number.

I think what you are saying is that an 8-bit colour space is more than sufficient to describe a typical photographic colour space like sRGB, AdobeRGB and ProColor RGB. Here I would be ready to agree with you if we left the editing step out, because all of these can look good as is.

There doesn't exist an 8 bit color space as a single item. It's the color space of your monitor and a 8 bit approach of that monitor. And of course a 16 bit approach when editing is better as a 8 bit approach.

How often do we read JPEG files cannot be edited because this process will result in artifacts. This statement is also not quite true as I edit a lot of JPEG files and they look just fine afterwards; BUT if the data is pushed too hard during the edit, we can see artifacts. If these are low quality jpegs, some of the problem is related to what the compression algorithms have done to the data, but part of the problem is that they are using 8-bit data and that does leave a limit of the number of colours that can be created / modified by the editing software. A similar issue exists when using a wide-gamut colour space like ProPhoto. We need enough data to work with in our editing software to ensure that we do not create artifacts. This is the reason that the common recommendation is that ProPhoto should use 16-bit data.

Despite another thread, an digital image is just a queue of pixels. The JPEG is a compressed disk file of that queue. As a standard it's 8 bit. But it can "contain" any color space. Read your last statement in the above. "This is the reason that the common recommendation is that ProPhoto should use 16-bit data."

A modern camera captures a wider gamut than what even a wide-gamut AdobeRGB compliant screen can capture. While our screens cannot reproduce all of the colours, modern high-end ink jet photo printers can produce more than our screens can show. One reason to work in a wider colour space is that these colours can be printed, so why would we not do so? We may not be able to see them on our screen, but we certainly will see them in our prints.

I did write before there must be something of a color space for the sensor. A kind of definition of how sensitive they're for the R,G and B colors. A combination of the filters and the sensitivity of the sensor. I got all the community rolling over me that the RAW image doesn't have a color space. Glad to hear you also think there is something as an "input gamut", opposed to the "output gamut". I can't argue about the other statements for I just don't know anything of it.

There is a second argument that is rather subtle and I understand that is why Adobe chose a variant of the ProPhoto colour space as the default colour space in its Lightroom application. If we decide to use a narrower colour space, then the system's Colour Management Module (CMM), recalculates all the colours so they fit within the new colour space. This means any manipulations during the edit process are based on the "less accurate" colours of the narrower gamut. If we work with the image data in a wide gamut colour space, the data manipulation occurs in the wider colour space and gives more accurate colours that are manipulated near the edges of the display's colour space.

This was my second question. I'm still not satisfied with the answer. Compare working 16 bit in sRGB and 16 bit in ProPhoto. Due to the wider gamut the steps in ProPhoto are bigger. What I understood of LR and ProPhoto is that it's easier to downgrade in gamut than upgrade. Not sure about it. What I see and what I judge about is in the color space of the monitor. And that must be ok.

George

[DanK]

I think this thread has outlived its usefulness.

[xpatUSA]

I think this thread has outlived its usefulness.

Those less interested would likely agree, Dan, but the bone of contention between the main opponents has not yet been resolved, IMHO.

[Manfred M]

Those less interested would likely agree, Dan, but the bone of contention between the main opponents has not yet been resolved, IMHO.

Nor will it be, I suspect, Ted. It really depends on how one approaches what a colour space is.

If it is purely a mathematical model used to define a set of colours and shades, there will be one answer. If it is a mathematical model designed to be implemented as computer code, that gives a different view.

CIE Lab 1931 is the former and I suspect that Pointer's gamut is as well.

sRGB, AdobeRGB and ROMM RGB (more commonly known as ProPhoto RGB) fall in the latter group. These three models were designed to address specific issues related to digital photography. Once the colour model is implemented in code, the practical considerations become important.

The question at the start of this thread dealt with a very specific, practical question. The answers tried to address those using the appropriate bit depth is part of the answer. To quote the Wikipedia article on ProPhoto: "When working in color spaces with such a large gamut, it is recommended to work in 16-bit color depth to avoid posterization effects. This will occur more frequently in 8-bit modes as the gradient steps are much larger."

[george013]

Nor will it be, I suspect, Ted. It really depends on how one approaches what a colour space is.

If it is purely a mathematical model used to define a set of colours and shades, there will be one answer. If it is a mathematical model designed to be implemented as computer code, that gives a different view.

CIE Lab 1931 is the former and I suspect that Pointer's gamut is as well.

sRGB, AdobeRGB and ROMM RGB (more commonly known as ProPhoto RGB) fall in the latter group. These three models were designed to address specific issues related to digital photography. Once the colour model is implemented in code, the practical considerations become important.

The question at the start of this thread dealt with a very specific, practical question. The answers tried to address those using the appropriate bit depth is part of the answer. To quote the Wikipedia article on ProPhoto: "When working in color spaces with such a large gamut, it is recommended to work in 16-bit color depth to avoid posterization effects. This will occur more frequently in 8-bit modes as the gradient steps are much larger."

Manfred, can you tell me the difference with

There is no relation between bit depth and color space. They're 2 completely separate items. That's one point I'm shouting out.

And with the other arguments I gave.

George

[Manfred M]

Manfred, can you tell me the difference with And with the other arguments I gave.

George

I can't George because your statement that there is no relationship between bit depth and colour space is not correct. Here I am only writing about the colour spaces that are implemented in photo software; implementation matters.

Again, I'm not the only one saying this: "When working in color spaces with such a large gamut, it is recommended to work in 16-bit color depth to avoid posterization effects. This will occur more frequently in 8-bit modes as the gradient steps are much larger." which is a direct quote from the Wikipedia article on ProPhoto:

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

[george013]

I can't George because your statement that there is no relationship between bit depth and colour space is not correct. Here I am only writing about the colour spaces that are implemented in photo software; implementation matters.

Again, I'm not the only one saying this: "When working in color spaces with such a large gamut, it is recommended to work in 16-bit color depth to avoid posterization effects. This will occur more frequently in 8-bit modes as the gradient steps are much larger." which is a direct quote from the Wikipedia article on ProPhoto:

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

Manfred, what does "recommended" mean??? Theoretical you can choice any number x. Editing in 16 bit, a practical number the software uses being a word, is to be preferred for any color space.

And comparing the difference color spaces using an equal bitdepth tells me that the steps for the smaller color space are smaller as from the bigger color space. It all has been said.

Leaves me with the still unanswered question why some people work in a larger gamut/coolr space as their monitor can handle.

Just thinking further, comparing with an equal bit depth the larger color space might face posterization before the smaller color space. Bigger steps. Just a thought.

George

[Manfred M]

Just thinking further, comparing with an equal bit depth the larger color space might face posterization before the smaller color space. Bigger steps. Just a thought.

George

YES! That is exactly the reason a larger bit depth is used. That is what I was trying to show with the diagrams in #17.

For the other questions, I have covered that off in #25.

[george013]

And what about using a smaller color space?

George

[Manfred M]

And what about using a smaller color space?

George

You get less colours, depending of course on what colour data the image contains. Using the CIE Lab 1931 colour space as reference:

1. sRGB - covers ~ 35% of the colours in CIE Lab 1931.

2. Adobe RGB - covers ~ 50% of the colours in CIE Lab 1931

3. ProPhoto - covers ~ 90% of the colours in CIE Lab 1931

Note - these numbers are the ones I remember off the top of my head, I could be off by a few percentage points.

If the colours in your image are contained within the sRGB colour space, you won't see any appreciable differences. If the scene has vibrant colours, wider colour spaces will contain more of these colours and the images will look more vibrant. The problem is trying to demonstrate that on the internet where sRGB is the colour space of choice. It is easy to see in colour prints, when you put images printed in them side by side.

[george013]

You get less colours, depending of course on what colour data the image contains. Using the CIE Lab 1931 colour space as reference:

1. sRGB - covers ~ 35% of the colours in CIE Lab 1931.

2. Adobe RGB - covers ~ 50% of the colours in CIE Lab 1931

3. ProPhoto - covers ~ 90% of the colours in CIE Lab 1931

Note - these numbers are the ones I remember off the top of my head, I could be off by a few percentage points.

If the colours in your image are contained within the sRGB colour space, you won't see any appreciable differences. If the scene has vibrant colours, wider colour spaces will contain more of these colours and the images will look more vibrant. The problem is trying to demonstrate that on the internet where sRGB is the colour space of choice. It is easy to see in colour prints, when you put images printed in them side by side.

That brings us back to the question why working in a larger color space as the monitor has. For just editing.

George

[xpatUSA]

Nor will it be, I suspect, Ted. It really depends on how one approaches what a colour space is.

If it is purely a mathematical model used to define a set of colours and shades, there will be one answer. If it is a mathematical model designed to be implemented as computer code, that gives a different view.

CIE Lab 1931 is the former and I suspect that Pointer's gamut is as well.

sRGB, AdobeRGB and ROMM RGB (more commonly known as ProPhoto RGB) fall in the latter group. These three models were designed to address specific issues related to digital photography. Once the colour model is implemented in code, the practical considerations become important.

Or, chronologically speaking, "ProPhoto RGB (formerly known as Kodak ROMM RGB)"?

The question at the start of this thread dealt with a very specific, practical question. The answers tried to address those using the appropriate bit depth is part of the answer. To quote the Wikipedia article on ProPhoto: "When working in color spaces with such a large gamut, it is recommended to work in 16-bit color depth to avoid posterization effects. This will occur more frequently in 8-bit modes as the gradient steps are much larger."

With all due respect Manfred, the bolded statement is like 'Editing 101' and tells most of us here nothing new, don't you think?

[Manfred M]

With all due respect Manfred, the bolded statement is like 'Editing 101' and tells most of us here nothing new, don't you think?

Actually I bolded it to make the quote stand out from the text, nothing more.

[xpatUSA]

Actually I bolded it to make the quote stand out from the text, nothing more.

Bolded or not, it is still 'editing 101' and tells most of us here nothing new, I suspect.

[xpatUSA]

That brings us back to the question why working in a larger color space [than] the monitor has. For just editing.

George

George, I work in ProPhoto RGB in RawTherapee because that is RT's native working space.

So there's one reason why.

Is that a valid reason in your world, or not?

Others have already told you that, in a larger gamut working space, major editing trips out-of-sRGB-gamut and back can be made without artifacts appearing in the review image.

Is that true or not?

Trying to get George to agree to something [I say] is like pulling teeth, so don't hold y'all's breathe . . .