New article on color management

Started Apr 26, 2013 | Discussions
Detail Man
Forum ProPosts: 14,950
Like?
Re: New article on color management
In reply to gollywop, Apr 27, 2013

gollywop wrote:

Detail Man wrote:

gollywop wrote:

For better or for worse, I have just published a new article: Color Management - a Walkthrough. It can be found at

http://www.dpreview.com/articles/7270088913/color-management-a-walkthrough

So I reiterate. You describe a situation:

You've just taken a shot of a fire engine, and, behind the scenes, it is captured in the raw data. But you want an OOC jpeg to show Junior. ...

Who the hell is Junior ?

By the way, do you like it better now?

Haven't really poked through it looking for changes. Good that you added Bruce Lindbloom's site. Have seen a place or two where it seems that you may have re-worked the language a bit.

The nitty-gritty of when color-management really matters most is (perhaps) where it fails to exist. Thus, the matter of many image-viewers and browsers still not recognizing and making the necessary adjustments pertaining to the display monitor's color-profile becomes a case by case miasma to attempt to discover information about and to subsequently decode.

Every time that I have endeavored to research color management on the interent (particularly with respect to internet browsers), I have become nauseated with the plethora of incomplete and sometimes conflicting (complicated by particular version-dependent) "information".

The mentality where it comes to in-depth and relevant information surrounding browsers is largely one where "the new" is all that is perceived to ever matter. Add that to the fact that all too little in the way of web-pages are identified by their date of creation (or last editing), and a real live mess that approaches useless (for the lone internet researcher) can rapidly evolve ...

Such a situation hardly encourages the in-depth understanding of such relatively esoteric matters, and a positive feeback-loop evolves from the fact that many have become accustomed to assuming that other people (surely must) have thought all these things through. Add proprietary application marketing hype, free browser applications, and there often exists all too little coherence to be found.

The limited attention-span, focus only upon "the new", the dearth of dated web-page documents, and the persistence of obsolete information on the internet does not say much for the internet's potential as a (necessarily) viable information source for any but the most simple and visceral of subjects.

Case in point: gear-head forums. Thus, your studious and thoughtful contributions are appreciated !

DM ...

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,388
Like?
Re: Since I have you here...
In reply to gollywop, Apr 27, 2013

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

For better or for worse, I have just published a new article: Color Management - a Walkthrough. It can be found at

http://www.dpreview.com/articles/7270088913/color-management-a-walkthrough

This is a subject I need a great deal of education in -- I'm glad to see such a comprehensive article on the subject!

Turnabout's fair play.

many thanks, GB

...let me ask you a question.  If someone worked entirely in sRGB to produce a print, under what circumstances might working in a different color space result in a "significantly better" print?

Am I to assume here that the shot was taken as a jpeg in sRGB, or are we shooting raw?

RAW, of course.  What's jpg, by the way? 

If it starts out life as an 8-bit sRGB jpeg, there isn't much sense in going anywhere else.

If you've shot raw and your printer has a gamut larger than sRGB, then you'll do better, when processing images that have saturated red, oranges, and/or blues, using a wider space, at least Adobe RGB.  This would often be the case for your red roses or dayglo orange graffiti or cobalt blue underwear.

If you've got a wide-gamut monitor that is capable of a broader gamut than sRGB (otherwise forget it), you can see just what is likely to happen by opening your Adobe RGB-processed file in PS, duplicating it and soft proofing the duplicate using the sRGB profile.

So, unless your monitor can work in a colorspace other than sRGB, you're better off just staying in sRGB?  That is, can't the RAW converter and/or editor work in aRGB and approximate the results on an sRGB monitor, yet still retain the advantages of working in a larger color space?  Or are you saying that unless your monitor can display the larger color space, any conversion or editing you do in sRGB will be counterproductive?

Toggling between the two will give you an idea of what is possible. Or, even more simply, use the Convert (to sRGB) dialog and toggle the Preview checkbox.  If you see no important changes, then using the wider gamut hasn't bought you much, if anything.

I'm not sure what you're saying here.  How do you toggle between colorspaces if your monitor can only handle sRGB?  Or are you saying toggle between color spaces on a monitor that can display both color spaces?

Clearly, if your monitor's gamut is roughly sRGB, you'll not be able to see any effect even if it exists.  This is even more true for a gamut like ProPhoto RGB. No monitor is going to let you see a lot of what might be happening by using this space.  Similarly, if your printer's gamut is close to sRGB, there is not likely to be any advantage in using a wider working space.  The conversion will be done by the printer's profile and you'd be better off controlling it yourself prior to printing.

Sounds to me like what you're saying is you should only work in the colorspaces that are common to all the devices in the imaging chain.  Is this correct?

I've got a number of sunset shots with saturated oranges that I've processed with both sRGB and Adobe RGB, and the differences are quite noticeable from my Canon PixmaPro 9000 II (particularly when using Red River papers).  The sRGB results are dull and disappointing.

So, I take it that both your monitor and printer can work in aRGB?

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Does Melissa "Gloss Over" Image Noise ?
In reply to Jeff Schewe, Apr 27, 2013

Jeff Schewe wrote:

Detail Man wrote:

It's also not clear in what sense the RGB color readouts at low levels are "erroneous."

Because they (according to Adobe's Jeff Schewe) report tone-level values within Melissa RGB's linearized low-level region that do not accurately reflect ProPhoto RGB or Adobe RGB values ...

Just to be clear, Melissa RGB is ONLY used for the Histogram display and the color readouts...not you image preview display. The image preview goes from raw>ProPhoto RGB Linear>Display profile. The sRGB tone curve has no impact on your image preview display.

My apologies, Jeff. I see now that I missed the significance of your statement (below):

... the ONLY place where Melissa RGB is used is for the Develop histogram and the color readouts.

So, the only impact of the sRGB tone curve would be color readouts in 0-100% (which don't relate to 0-255) and the way the levels are displayed on the histogram.

So, the gamma-correction function that is applied to the Adobe CR/LR preview display-screen corresponds to whatever particular color-space that the OS recognized "system display profile" is ?

If that is the case, then it would seem that a system having an sRGB color-space "system display profile" would not be able to accurately reflect the gamma-correction function that results in cases when the output images are rendered in ProPhoto RGB or Adobe RGB 1998 color-spaces ?

It seems that it would be useful for the preview-display to have the operational capacity to reflect the low-level gamma-correction functionality of those other (and differing) color-spaces - in that user's rely upon judgments surrounding image-noise levels in deciding upon the settings of various processor/editor controls.

.

From quoted text attributed by Sean McCormack to Adobe's Jeff Schewe:

Lightroom (and Camera Raw) uses a "working space" (meaning the processing
color space) of Pro Photo chomaticities (colors) and a linear gamma (1.0 gamma).

Martin is correct to call LR's working space as linear, but (and this is
where it gets confusing), "Melissa RGB" (the color space coined by Mark
Hamburg) isn't the processing space of LR/CR but the display space in Develop.

Melissa RGB does indeed have the Pro Photo chomaticities but the gamma is
actually that of sRGB and the ONLY place where Melissa RGB is used is for
the Develop histogram and the color readouts.

So, Lightroom's internal processing space is ProPhoto RGB in a linear gamma
(1.0), but Melissa RGB is ProPhoto RGB with an sRGB tone curve. Also note
that the sRGB tone curve is not a simple gamma curve but is a tweaked curve
based on gamma 2.2 but with the toe adjusted.
Jeff Schewe

- Sean Mccormack, 11 May 2008, 1:55 am: http://www.lightroomforums.net/showthread.php?1758-LR-ProPhotoRGB-Gamma-1-8-and-monitor-calibration&s=5ff4347babf64107e62cd96b95f60288

Reply   Reply with quote   Complain
Jeff Schewe
Regular MemberPosts: 369
Like?
Re: Does Melissa "Gloss Over" Image Noise ?
In reply to Detail Man, Apr 27, 2013

Detail Man wrote:

So, the only impact of the sRGB tone curve would be color readouts in 0-100% (which don't relate to 0-255) and the way the levels are displayed on the histogram.

So, the gamma-correction function that is applied to the Adobe CR/LR preview display-screen corresponds to whatever particular color-space that the OS recognized "system display profile" is ?

If that is the case, then it would seem that a system having an sRGB color-space "system display profile" would not be able to accurately reflect the gamma-correction function that results in cases when the output images are rendered in ProPhoto RGB or Adobe RGB 1998 color-spaces ?

Don't assume the display profile is sRGB...my NEC displays are 98% of Adobe RGB...also don't get too wrapped up in the whole display/output profile gamma. They are two different things. The only place where the display gamma would have any impact is if an application isn't color managed. There the differences in the display gamma and output gamma would have an impact since sRGB/Adobe RGB are both nominally 2.2 (sRGB has a toe tweak) and ProPhoto RGB which is gamma 1.8.

But, as long as viewing apps are color managed, the final appearance on-screen should be accurately transformed from the image color space/gamma to the display space/gamma.

-- hide signature --

Regards,
Jeff Schewe

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Does Melissa "Gloss Over" Image Noise ?
In reply to Jeff Schewe, Apr 27, 2013

Jeff Schewe wrote:

Detail Man wrote:

So, the only impact of the sRGB tone curve would be color readouts in 0-100% (which don't relate to 0-255) and the way the levels are displayed on the histogram.

So, the gamma-correction function that is applied to the Adobe CR/LR preview display-screen corresponds to whatever particular color-space that the OS recognized "system display profile" is ?

If that is the case, then it would seem that a system having an sRGB color-space "system display profile" would not be able to accurately reflect the gamma-correction function that results in cases when the output images are rendered in ProPhoto RGB or Adobe RGB 1998 color-spaces ?

Don't assume the display profile is sRGB...my NEC displays are 98% of Adobe RGB...

Right.

... also don't get too wrapped up in the whole display/output profile gamma. They are two different things.

Could you explain that further ? It seems like you have stated that the gamma-correction funtion that is applied to the CR/LR (preview) display is solely determined by the particular (OS recognized) color-space that the (utilized) system output profile exists as a set of "tweaks" to. Right ?

The only place where the display gamma would have any impact is if an application isn't color managed. There the differences in the display gamma and output gamma would have an impact since sRGB/Adobe RGB are both nominally 2.2 (sRGB has a toe tweak) and ProPhoto RGB which is gamma 1.8.

But, as long as viewing apps are color managed, the final appearance on-screen should be accurately transformed from the image color space/gamma to the display space/gamma.

I recognize that the common (and most likely case) is that a user would be processing an image-file for display on the very same system that they are performing the image processing on.

My thought, however, is that such is not always and invariably the case - and that it would be useful if CR/LR preview display was capable of using the particular gamma-correction function directly associated with whatever color-space(s) they may be choosing to render in and save to.

Otherwise, the previewing of low-level image-noise is locked into whatever particular color-space that the display associated with the system being used to do the processing happens to be.

That would be fine if one has a wider gamut display and might choose to (also) alternately restrict the (OS recognized system) color-space to a narrower gamut color-space for rendering in and saving to.

However, if one is performing the processing/editing on a system having a narrow gamut (OS recognized) system color-space (only), but would like to (also) render in and save to wider gamut color-spaces for display on other systems, then they would seem to be (unecessarily) limited with respect to the ability to accurately preview low-level image-noise as it would appear when viewed in color-spaces having wider gamuts.

Thus, the linearized low-level region of an sRGB (only) system would not serve to accurately reflect the appearance of image-noise when rendered in and saved to ProPhoto RGB (which has a lower threshold of low-level linearization), or Adobe RGB 1998 (which has no low-level linearization at all).

For such (albeit not necessarily the most common) scenarios, having the ability via user controls to (in the preview display) accurately simulate the low-level gamma-correction functionality of wider gamut color-spaces would seem to be a useful option (existing as an alternative to the default settings).

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,950
Like?
Re: Does Melissa "Gloss Over" Image Noise ?
In reply to Detail Man, Apr 27, 2013

Jeff Schewe,

From browsing through some of your recent DPReview posts I came across this clarification:

Jeff Schewe wrote:

Just to be clear, I'm not an Adobe engineer...I don't work for Adobe (although I sometimes work with them) and I don't do code...

Now that I'm aware of the specifics of your association, I'll word my future writings accordingly.

Regards, DM

.

Detail Man wrote:

From quoted text attributed by Sean McCormack to Adobe's Jeff Schewe:

Lightroom (and Camera Raw) uses a "working space" (meaning the processing
color space) of Pro Photo chomaticities (colors) and a linear gamma (1.0 gamma).

Martin is correct to call LR's working space as linear, but (and this is
where it gets confusing), "Melissa RGB" (the color space coined by Mark
Hamburg) isn't the processing space of LR/CR but the display space in Develop.

Melissa RGB does indeed have the Pro Photo chomaticities but the gamma is
actually that of sRGB and the ONLY place where Melissa RGB is used is for
the Develop histogram and the color readouts.

So, Lightroom's internal processing space is ProPhoto RGB in a linear gamma
(1.0), but Melissa RGB is ProPhoto RGB with an sRGB tone curve. Also note
that the sRGB tone curve is not a simple gamma curve but is a tweaked curve
based on gamma 2.2 but with the toe adjusted.
Jeff Schewe

- Sean Mccormack, 11 May 2008, 1:55 am: http://www.lightroomforums.net/showthread.php?1758-LR-ProPhotoRGB-Gamma-1-8-and-monitor-calibration&s=5ff4347babf64107e62cd96b95f60288

Reply   Reply with quote   Complain
Jeff Schewe
Regular MemberPosts: 369
Like?
Re: Does Melissa "Gloss Over" Image Noise ?
In reply to Detail Man, Apr 27, 2013

Detail Man wrote:

... also don't get too wrapped up in the whole display/output profile gamma. They are two different things.

Could you explain that further ? It seems like you have stated that the gamma-correction funtion that is applied to the CR/LR (preview) display is solely determined by the particular (OS recognized) color-space that the (utilized) system output profile exists as a set of "tweaks" to. Right ?

Well, you have to separate ACR & LR here...in ACR, you must choose a Workflow Options color space and the preview (and histogram & readouts) will be in that chosen color space.

In LR, it'a  bit different. While the ACR/LR working space is the same (ProPhoto RGB linear gamma) LR doesn't alter the image preview based on any external settings unless you turn on the Soft Proof option.

As a result, ACR will show the main image preview in the final output space set in the workflow option, LR will only shouw the image preview based on the PP RGG Linear>display profile only (again with soft proofing turned off).

The only time this really makes any difference is when viewing an image outside of a color managed workflow–otherwise, it's a rabbit hole not worth going down.

Otherwise, I really don't think it's worth worrying about the gamma differences...color management takes care of gamma corrections.

-- hide signature --

Regards,
Jeff Schewe

Reply   Reply with quote   Complain
Jeff Schewe
Regular MemberPosts: 369
Like?
Re: Does Melissa "Gloss Over" Image Noise ?
In reply to Detail Man, Apr 27, 2013

Detail Man wrote:

Jeff Schewe wrote:

Just to be clear, I'm not an Adobe engineer...I don't work for Adobe (although I sometimes work with them) and I don't do code...

Now that I'm aware of the specifics of your association, I'll word my future writings accordingly.

Regards, DM

Yeah, I'm closely associated with Adobe, but I don't speak for Adobe...it's complicated: I work with the engineers, but don't work for Adobe, the company. As I said, it's complicated (and sometimes makes Adobe a bit uncomfortable).

:~)

-- hide signature --

Regards,
Jeff Schewe

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,439
Like?
Re: Since I have you here...
In reply to Great Bustard, Apr 27, 2013

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

For better or for worse, I have just published a new article: Color Management - a Walkthrough. It can be found at

http://www.dpreview.com/articles/7270088913/color-management-a-walkthrough

This is a subject I need a great deal of education in -- I'm glad to see such a comprehensive article on the subject!

Turnabout's fair play.

many thanks, GB

...let me ask you a question.  If someone worked entirely in sRGB to produce a print, under what circumstances might working in a different color space result in a "significantly better" print?

Am I to assume here that the shot was taken as a jpeg in sRGB, or are we shooting raw?

RAW, of course.  What's jpg, by the way? 

If it starts out life as an 8-bit sRGB jpeg, there isn't much sense in going anywhere else.

If you've shot raw and your printer has a gamut larger than sRGB, then you'll do better, when processing images that have saturated red, oranges, and/or blues, using a wider space, at least Adobe RGB.  This would often be the case for your red roses or dayglo orange graffiti or cobalt blue underwear.

If you've got a wide-gamut monitor that is capable of a broader gamut than sRGB (otherwise forget it), you can see just what is likely to happen by opening your Adobe RGB-processed file in PS, duplicating it and soft proofing the duplicate using the sRGB profile.

So, unless your monitor can work in a colorspace other than sRGB, you're better off just staying in sRGB?  That is, can't the RAW converter and/or editor work in aRGB and approximate the results on an sRGB monitor, yet still retain the advantages of working in a larger color space?  Or are you saying that unless your monitor can display the larger color space, any conversion or editing you do in sRGB will be counterproductive?

No, not really - at least if you're processing for print and your printer can outstrip sRGB to some degree.  But then your limited monitor will have you flying somewhat blind in the processing and will only be able to see the effects after the fact in the print.

Toggling between the two will give you an idea of what is possible. Or, even more simply, use the Convert (to sRGB) dialog and toggle the Preview checkbox.  If you see no important changes, then using the wider gamut hasn't bought you much, if anything.

I'm not sure what you're saying here.  How do you toggle between colorspaces if your monitor can only handle sRGB?  Or are you saying toggle between color spaces on a monitor that can display both color spaces?

Yes, the latter.

Clearly, if your monitor's gamut is roughly sRGB, you'll not be able to see any effect even if it exists.  This is even more true for a gamut like ProPhoto RGB. No monitor is going to let you see a lot of what might be happening by using this space.  Similarly, if your printer's gamut is close to sRGB, there is not likely to be any advantage in using a wider working space.  The conversion will be done by the printer's profile and you'd be better off controlling it yourself prior to printing.

Sounds to me like what you're saying is you should only work in the colorspaces that are common to all the devices in the imaging chain.  Is this correct?

As indicted above, if the printer has a gamut larger than the monitor, you might try, somewhat blindly, to take advantage of it.  Some computer monitors have a really poor gamut, much less than a good inkjet printer. That combination would clearly put you at a real disadvantage, and I suspect it's not all that uncommon.

And, of course, you're always flying blind to some degree in using ProPhoto RGB (or LAB) because even the best of monitors can't really deal with the entirety of those spaces.  That being said, I've certainly had a stretch where I used ProPhoto RGB almost all the time and printed out of that working space with great success.

And, by the way, it gets messier than that because monitor/printer gamuts rarely just encompass one another.  The monitor may be broader than the printer in some color dimensions and narrower in others, and vice versa.

I've got a number of sunset shots with saturated oranges that I've processed with both sRGB and Adobe RGB, and the differences are quite noticeable from my Canon PixmaPro 9000 II (particularly when using Red River papers).  The sRGB results are dull and disappointing.

So, I take it that both your monitor and printer can work in aRGB?

Yes, but I the monitor is better in the reds than the printer.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,388
Like?
Re: Since I have you here...
In reply to gollywop, Apr 27, 2013

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

For better or for worse, I have just published a new article: Color Management - a Walkthrough. It can be found at

http://www.dpreview.com/articles/7270088913/color-management-a-walkthrough

This is a subject I need a great deal of education in -- I'm glad to see such a comprehensive article on the subject!

Turnabout's fair play.

many thanks, GB

...let me ask you a question.  If someone worked entirely in sRGB to produce a print, under what circumstances might working in a different color space result in a "significantly better" print?

Am I to assume here that the shot was taken as a jpeg in sRGB, or are we shooting raw?

RAW, of course.  What's jpg, by the way? 

If it starts out life as an 8-bit sRGB jpeg, there isn't much sense in going anywhere else.

If you've shot raw and your printer has a gamut larger than sRGB, then you'll do better, when processing images that have saturated red, oranges, and/or blues, using a wider space, at least Adobe RGB.  This would often be the case for your red roses or dayglo orange graffiti or cobalt blue underwear.

If you've got a wide-gamut monitor that is capable of a broader gamut than sRGB (otherwise forget it), you can see just what is likely to happen by opening your Adobe RGB-processed file in PS, duplicating it and soft proofing the duplicate using the sRGB profile.

So, unless your monitor can work in a colorspace other than sRGB, you're better off just staying in sRGB?  That is, can't the RAW converter and/or editor work in aRGB and approximate the results on an sRGB monitor, yet still retain the advantages of working in a larger color space?  Or are you saying that unless your monitor can display the larger color space, any conversion or editing you do in sRGB will be counterproductive?

No, not really - at least if you're processing for print and your printer can outstrip sRGB to some degree.  But then your limited monitor will have you flying somewhat blind in the processing and will only be able to see the effects after the fact in the print.

Toggling between the two will give you an idea of what is possible. Or, even more simply, use the Convert (to sRGB) dialog and toggle the Preview checkbox.  If you see no important changes, then using the wider gamut hasn't bought you much, if anything.

I'm not sure what you're saying here.  How do you toggle between colorspaces if your monitor can only handle sRGB?  Or are you saying toggle between color spaces on a monitor that can display both color spaces?

Yes, the latter.

Clearly, if your monitor's gamut is roughly sRGB, you'll not be able to see any effect even if it exists.  This is even more true for a gamut like ProPhoto RGB. No monitor is going to let you see a lot of what might be happening by using this space.  Similarly, if your printer's gamut is close to sRGB, there is not likely to be any advantage in using a wider working space.  The conversion will be done by the printer's profile and you'd be better off controlling it yourself prior to printing.

Sounds to me like what you're saying is you should only work in the colorspaces that are common to all the devices in the imaging chain.  Is this correct?

As I indicted above, if the printer has a gamut larger than the monitor, you might try, somewhat blindly, to take advantage of it.  Some computer monitors have a really poor gamut, much less than a good inkjet printer. That combination would clearly put you at a real disadvantage, and I suspect it's not all that uncommon.

I've got a number of sunset shots with saturated oranges that I've processed with both sRGB and Adobe RGB, and the differences are quite noticeable from my Canon PixmaPro 9000 II (particularly when using Red River papers).  The sRGB results are dull and disappointing.

So, I take it that both your monitor and printer can work in aRGB?

Yes.

OK -- good stuff!  I have two followup questions:

  1. Are there monitors and printers that do PPRGB?  If so, what do they cost?
  2. What role does the paper the photo is printed on play in all this?
Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,439
Like?
Re: New article on color management
In reply to Detail Man, Apr 27, 2013

Detail Man wrote:

gollywop wrote:

Detail Man wrote:

gollywop wrote:

For better or for worse, I have just published a new article: Color Management - a Walkthrough. It can be found at

http://www.dpreview.com/articles/7270088913/color-management-a-walkthrough

So I reiterate. You describe a situation:

You've just taken a shot of a fire engine, and, behind the scenes, it is captured in the raw data. But you want an OOC jpeg to show Junior. ...

Who the hell is Junior ?

By the way, do you like it better now?

Haven't really poked through it looking for changes. Good that you added Bruce Lindbloom's site. Have seen a place or two where it seems that you may have re-worked the language a bit.

The nitty-gritty of when color-management really matters most is (perhaps) where it fails to exist. Thus, the matter of many image-viewers and browsers still not recognizing and making the necessary adjustments pertaining to the display monitor's color-profile becomes a case by case miasma to attempt to discover information about and to subsequently decode.

Every time that I have endeavored to research color management on the interent (particularly with respect to internet browsers), I have become nauseated with the plethora of incomplete and sometimes conflicting (complicated by particular version-dependent) "information".

The mentality where it comes to in-depth and relevant information surrounding browsers is largely one where "the new" is all that is perceived to ever matter. Add that to the fact that all too little in the way of web-pages are identified by their date of creation (or last editing), and a real live mess that approaches useless (for the lone internet researcher) can rapidly evolve ...

Yes, I fail completely to understand why the "browser situation" hasn't been cleared up.  The issue has been known for a long time, and FireFox has shown that it can be solved.  Since a great deal of photographic display is now taking place on the web (and it is growing all the time), the issue is one of real significance.

As to the nausea, yes indeed.  It really is a head-banging, stomach-turning situation.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,439
Like?
Re: Since I have you here...
In reply to Great Bustard, Apr 27, 2013

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

Great Bustard wrote:

gollywop wrote:

For better or for worse, I have just published a new article: Color Management - a Walkthrough. It can be found at

http://www.dpreview.com/articles/7270088913/color-management-a-walkthrough

This is a subject I need a great deal of education in -- I'm glad to see such a comprehensive article on the subject!

Turnabout's fair play.

many thanks, GB

...let me ask you a question.  If someone worked entirely in sRGB to produce a print, under what circumstances might working in a different color space result in a "significantly better" print?

Am I to assume here that the shot was taken as a jpeg in sRGB, or are we shooting raw?

RAW, of course.  What's jpg, by the way? 

If it starts out life as an 8-bit sRGB jpeg, there isn't much sense in going anywhere else.

If you've shot raw and your printer has a gamut larger than sRGB, then you'll do better, when processing images that have saturated red, oranges, and/or blues, using a wider space, at least Adobe RGB.  This would often be the case for your red roses or dayglo orange graffiti or cobalt blue underwear.

If you've got a wide-gamut monitor that is capable of a broader gamut than sRGB (otherwise forget it), you can see just what is likely to happen by opening your Adobe RGB-processed file in PS, duplicating it and soft proofing the duplicate using the sRGB profile.

So, unless your monitor can work in a colorspace other than sRGB, you're better off just staying in sRGB?  That is, can't the RAW converter and/or editor work in aRGB and approximate the results on an sRGB monitor, yet still retain the advantages of working in a larger color space?  Or are you saying that unless your monitor can display the larger color space, any conversion or editing you do in sRGB will be counterproductive?

No, not really - at least if you're processing for print and your printer can outstrip sRGB to some degree.  But then your limited monitor will have you flying somewhat blind in the processing and will only be able to see the effects after the fact in the print.

Toggling between the two will give you an idea of what is possible. Or, even more simply, use the Convert (to sRGB) dialog and toggle the Preview checkbox.  If you see no important changes, then using the wider gamut hasn't bought you much, if anything.

I'm not sure what you're saying here.  How do you toggle between colorspaces if your monitor can only handle sRGB?  Or are you saying toggle between color spaces on a monitor that can display both color spaces?

Yes, the latter.

Clearly, if your monitor's gamut is roughly sRGB, you'll not be able to see any effect even if it exists.  This is even more true for a gamut like ProPhoto RGB. No monitor is going to let you see a lot of what might be happening by using this space.  Similarly, if your printer's gamut is close to sRGB, there is not likely to be any advantage in using a wider working space.  The conversion will be done by the printer's profile and you'd be better off controlling it yourself prior to printing.

Sounds to me like what you're saying is you should only work in the colorspaces that are common to all the devices in the imaging chain.  Is this correct?

As I indicted above, if the printer has a gamut larger than the monitor, you might try, somewhat blindly, to take advantage of it.  Some computer monitors have a really poor gamut, much less than a good inkjet printer. That combination would clearly put you at a real disadvantage, and I suspect it's not all that uncommon.

I've got a number of sunset shots with saturated oranges that I've processed with both sRGB and Adobe RGB, and the differences are quite noticeable from my Canon PixmaPro 9000 II (particularly when using Red River papers).  The sRGB results are dull and disappointing.

So, I take it that both your monitor and printer can work in aRGB?

Yes.

OK -- good stuff!  I have two followup questions:

  1. Are there monitors and printers that do PPRGB?  If so, what do they cost? 
  2. What role does the paper the photo is printed on play in all this?

I think (but am not 100% positive) the most ambitious gamut for monitors is NTSC, which is a bit wider than Adobe RGB. That gamut is, I think, supposed to be the best that a CRT can produce.  I know there are supposed to be monitors that are NTSC, but they are in the thousands.  I've never really looked that direction.

And, if you're going to go in that direction, you want more than just the monitor.  You've got to control the room ambient lighting to a fare-the-well and do the monitor calibration taking it into account.

sRGB is about 72% of NTSC and the decent wide-gamut LCDs available today get upwards of 95-98%, and pretty well encompass Adobe RGB.

NTSC is only about 54% of LAB, whereas ProPhoto RGB is 91%.

As to papers, wow!  They can differ in so many ways that one doesn't know where to start.  Decent glossy papers (I love Red River) have some of the best reflectance and can provide the best reflective DR.  But different glossy papers also differ in how well they take inks, etc.  Many people forgo DR for artistic effect and love to use matte and other textured papers.  And, then, of course, there are major differences in pigment vs. dye inks.

That's why it's so important that, when printing, you use a printer profile that is matched not only for the printer but the paper.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Jack Hogan
Senior MemberPosts: 3,626Gear list
Like?
Camera-->sRGB versus Camera-->ProPhoto-->sRGB
In reply to Great Bustard, Apr 27, 2013

Great Bustard wrote:

...let me ask you a question.  If someone worked entirely in sRGB to produce a print, under what circumstances might working in a different color space result in a "significantly better" print?

For example, let's say I'm taking a pic of a bunch of red roses, which, obviously, has lots of hues of heavily saturated reds.  Might working in a colorspace other than sRGB produce a "significantly better" photo, assuming, of course, that the printer worked in that color space?

Coincidentally I asked a similar question on a LuLa thread today before seeing this one.  Here's the gist of it:

Since in 99% of cases people are either printing or viewing in sRGB, it would be interesting to know whether it makes a practical difference in day to day use to work in sRGB from the very start versus using ProPhoto/other large color space to work in and convert to sRGB at the end of PP.  Squeeze at the beginning or at the end, but always squeeze we (non-pros) must.  So if your red flowers clipped when going straight to sRGB, they probably still will when ending up in sRGB after ProPhoto: except that in the latter you'd only realize that you are clipping the reds at the end of your session.

My feeling is that most of the color differences/shifts are introduced during conversion into the final color space (sRGB), and that the more (and the more extreme) adjustments one makes in a larger color space, the more the chances that final colors will be wild guesstimates in sRGB.

On the other hand when there are minimal adjustments, there is no reason why Camera Space --> XYZ --> ProPhoto --> sRGB at the very start should not result in very similar values to Camera Space --> XYZ --> sRGB, assuming proper conversion parameters. So:

. Many/extreme adjustments = potentially wild guesstimates going from large to small
. Few/no adjustments = potentially no difference to going straight from camera to small

And the fewer the color space conversions, the less the noise (unchecked gamma curves are not invertible at the origin).

I'd be interested in a discussion based on examples from a couple of 'difficult' Raw files, as opposed to opinions.  Anybody tried this who could provide the Raw materiall?

Jack

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,439
Like?
Re: Camera-->sRGB versus Camera-->ProPhoto-->sRGB
In reply to Jack Hogan, Apr 27, 2013

Jack Hogan wrote:

Great Bustard wrote:

...let me ask you a question.  If someone worked entirely in sRGB to produce a print, under what circumstances might working in a different color space result in a "significantly better" print?

For example, let's say I'm taking a pic of a bunch of red roses, which, obviously, has lots of hues of heavily saturated reds.  Might working in a colorspace other than sRGB produce a "significantly better" photo, assuming, of course, that the printer worked in that color space?

Coincidentally I asked a similar question on a LuLa thread today before seeing this one.  Here's the gist of it:

Since in 99% of cases people are either printing or viewing in sRGB, it would be interesting to know whether it makes a practical difference in day to day use to work in sRGB from the very start versus using ProPhoto/other large color space to work in and convert to sRGB at the end of PP.  Squeeze at the beginning or at the end, but always squeeze we (non-pros) must.  So if your red flowers clipped when going straight to sRGB, they probably still will when ending up in sRGB after ProPhoto: except that in the latter you'd only realize that you are clipping the reds at the end of your session.

My feeling is that most of the color differences/shifts are introduced during conversion into the final color space (sRGB), and that the more (and the more extreme) adjustments one makes in a larger color space, the more the chances that final colors will be wild guesstimates in sRGB.

On the other hand when there are minimal adjustments, there is no reason why Camera Space --> XYZ --> ProPhoto --> sRGB at the very start should not result in very similar values to Camera Space --> XYZ --> sRGB, assuming proper conversion parameters. So:

. Many/extreme adjustments = potentially wild guesstimates going from large to small
. Few/no adjustments = potentially no difference to going straight from camera to small

And the fewer the color space conversions, the less the noise (unchecked gamma curves are not invertible at the origin).

I'd be interested in a discussion based on examples from a couple of 'difficult' Raw files, as opposed to opinions.  Anybody tried this who could provide the Raw materiall?

Jack

Hello Jack. I basically agree with your two bulleted summary statements.  I'm wondering, however, where the XYZ came from unless you're talking about OOC jpegs.

I have a couple of potentially good candidate raw files for experimentation, but I have no way to make them available. They all, according to RawDigger are right on the edge, and they all fit into ProPhoto RGB but clip in sRGB. If you have means for making them available, I could e-mail them to you.  They are dngs (I didn't keep the originals).

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Jack Hogan
Senior MemberPosts: 3,626Gear list
Like?
Re: Camera-->sRGB versus Camera-->ProPhoto-->sRGB
In reply to gollywop, Apr 27, 2013

gollywop wrote:  I'm wondering, however, where the XYZ came from unless you're talking about OOC jpegs.

Hi gollywop.  As I understand it, all conversions (including from camera space, if not short circuited by vendors' proprietary Raw converters) need to go through Profile Connection Space (XYZ D50) .  Come to think of it, the two color space flows should probably look like this:

1) Camera Space --> PCS --> ProPhotoRGB --> PCS --> sRGB
2) Camera Space --> PCS --> sRGB

If my understanding is incorrect I'd be interested to know.

If you have means for making them available, I could e-mail them to you.  They are dngs (I didn't keep the originals).

Excellent, fire away to my mail and I will share them (google Drive, linked to the Gmail account - you could probably do it too).  If you also had a NEF, that'd be great.

Jack

Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,439
Like?
Re: Camera-->sRGB versus Camera-->ProPhoto-->sRGB
In reply to Jack Hogan, Apr 27, 2013

Jack Hogan wrote:

gollywop wrote:  I'm wondering, however, where the XYZ came from unless you're talking about OOC jpegs.

Hi gollywop.  As I understand it, all conversions (including from camera space, if not short circuited by vendors' proprietary Raw converters) need to go through Profile Connection Space (XYZ D50) .  Come to think of it, the two color space flows should probably look like this:

1) Camera Space --> PCS --> ProPhotoRGB --> PCS --> sRGB
2) Camera Space --> PCS --> sRGB

If my understanding is incorrect I'd be interested to know.

It's my understanding that the camera's jpeg process begins with a trip to XYZ.  I wouldn't take bets, but I don't believe ACR does and I'm of an even stronger belief that RPP doesn't.  But, I too, wouldn't mind if someone chimed on this point.

How does dcraw treat a raw file?  That would give us a good hint.

If you have means for making them available, I could e-mail them to you.  They are dngs (I didn't keep the originals).

Excellent, fire away to my mail and I will share them (google Drive, linked to the Gmail account - you could probably do it too).  If you also had a NEF, that'd be great.

I fear I no longer have the NEFs.  I think some may have started life as CR2s and some ORFs.  But I'll ship them along to you.  I'll look into google Drive, but just at the moment I don't have time (and won't for the coming month of May) to ride herd on such things.  I'll be away from the internet most of the time.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Jack Hogan
Senior MemberPosts: 3,626Gear list
Like?
Re: Camera-->sRGB versus Camera-->ProPhoto-->sRGB -- Files
In reply to gollywop, Apr 27, 2013

Three files by gollywop that allegedly fit within ProPhotoRGB but not sRGB.

File 1

File 2

File 3

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 22,388
Like?
Re: Since I have you here...
In reply to gollywop, Apr 27, 2013

gollywop wrote:

Great Bustard wrote:

OK -- good stuff!  I have two followup questions:

  1. Are there monitors and printers that do PPRGB?  If so, what do they cost? 
  2. What role does the paper the photo is printed on play in all this?

I think (but am not 100% positive) the most ambitious gamut for monitors is NTSC, which is a bit wider than Adobe RGB. That gamut is, I think, supposed to be the best that a CRT can produce.  I know there are supposed to be monitors that are NTSC, but they are in the thousands.  I've never really looked that direction.

And, if you're going to go in that direction, you want more than just the monitor.  You've got to control the room ambient lighting to a fare-the-well and do the monitor calibration taking it into account.

sRGB is about 72% of NTSC and the decent wide-gamut LCDs available today get upwards of 95-98%, and pretty well encompass Adobe RGB.

NTSC is only about 54% of LAB, whereas ProPhoto RGB is 91%.

As to papers, wow!  They can differ in so many ways that one doesn't know where to start.  Decent glossy papers (I love Red River) have some of the best reflectance and can provide the best reflective DR.  But different glossy papers also differ in how well they take inks, etc.  Many people forgo DR for artistic effect and love to use matte and other textured papers.  And, then, of course, there are major differences in pigment vs. dye inks.

That's why it's so important that, when printing, you use a printer profile that is matched not only for the printer but the paper.

Many thanks for another great post!

Reply   Reply with quote   Complain
xpatUSA
Senior MemberPosts: 2,648Gear list
Like?
Re: Camera-->sRGB versus Camera-->ProPhoto-->sRGB
In reply to gollywop, Apr 28, 2013

gollywop wrote:

How does dcraw treat a raw file?  That would give us a good hint.

For what it's worth, dcraw offers the following -o (output) options:

0 = raw [sic]

1 = sRGB

2 = Adobe RGB

3 = Adobe Wide

4 = ProPhoto (but D65, not D50).

5 = XYZ [sic]

<file> = apply output ICC from file

or

<file> = apply camera ICC profile from file or "embed" [sic]

-- hide signature --

Regards,
Ted http://kronometric.org
SD9, SD10, EF-500, GH1.

 xpatUSA's gear list:xpatUSA's gear list
Sigma SD9 Sigma SD10 Panasonic Lumix DMC-GH1
Reply   Reply with quote   Complain
gollywop
Veteran MemberPosts: 5,439
Like?
Re: Camera-->sRGB versus Camera-->ProPhoto-->sRGB
In reply to xpatUSA, Apr 28, 2013

xpatUSA wrote:

gollywop wrote:

How does dcraw treat a raw file?  That would give us a good hint.

For what it's worth, dcraw offers the following -o (output) options:

0 = raw [sic]

1 = sRGB

2 = Adobe RGB

3 = Adobe Wide

4 = ProPhoto (but D65, not D50).

5 = XYZ [sic]

<file> = apply output ICC from file

or

<file> = apply camera ICC profile from file or "embed" [sic]

Thanks, xpatUSA.  Yes, those are the choices for an output space.  So it would appear that XYZ is a possible output space, but it is not employed as a matter of course during processing, which is what I'd figure.  It's like RPP allows LAB as a choice for an output space, but it doesn't figure in any of the basic rendering operations.

-- hide signature --

gollywop

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads