Luminous Landscape Forum
Raw & Post Processing, Printing => Digital Image Processing => Topic started by: *Rich on August 28, 2013, 10:49:00 pm
-
My Dell UltraSharp monitor is color calibrated and my calibration software indicates it can show 99% of the Adobe RGB color space. So I have been using that color space in Photoshop editing.
I want to begin processing my RAW images in ProPhoto RGB, but I was wondering how to accurately edit the wider gamut of colors (than Adobe RGB) -- which I would not be able see on my monitor? Do you just guess which colors will not block-up when printed? Do I have to rely solely on printing lots of proofs now, rather than relying on what I see on my monitor? Thanks.
-
As you edit, when moving a slider (say Saturation or Vibrance), as soon as you stop seeing the image update (change), back off! You are affecting colors you can't see (Out of Gamut). That's your sign to move carefully.
-
Thanks Andrew, never knew that.
But what I am concerned about with ProPhoto RGB is that not all of the color space can be represented on my monitor -- or can it? It can only show 99% of the Adobe RGB color space on a good day. How will a particular hue appear on my monitor that is within the ProPhoto color space, like a deep blue or red, but is just outside the Adobe RGB space? Thanks.
-
Color spaces are synthetic mathematical description color map models for a computer to follow in order to allow the user of imaging software tools to manipulate the video preview on a display and utilize all possible colors the device can reproduce in the smoothest manner without artifacts.
The primaries of ProPhotoRGB are not viewable by human eyes because they are written in the form of math maps to tell a computer how far to go in manipulating color managed previews. Thus Andrew's test tells you how far the computer can go with both the data and the video card driven display preview.
In my experience working in ProPhotoRGB has been beneficial even editing on sRGB-ish LCD displays because it allows me to push the saturation to fullest extent of my display's capability.
That's as far as I can understand and explain it and have it make sense and be practical for me and whoever else.
-
Hi Rich,
Ideally the working color space should be as big as you need but no bigger. So how big do you need it to be? It depends on how your images are viewed. 99% of images are viewed on sRGB monitors/prints. Very few commercial printing services print on anything larger than aRGB. Some advanced home inkjet printers may be able to print colors outside of aRGB.
I have a very similar setup to yours and used to use MelissaD65 (a friendlier version of ProPhotoRGB) but cycled back to AdobeRGB in order to keep things tighter. My home printer does not print anything wider than aRGB. Convert to a smaller space at the beginning or at the end, but always convert one must :-)
Cheers,
Jack
-
Anyone editing RAW files in AdobeRGB will throw away data....
You might want to read this:
http://www.afximages.com/articles.php?article=WorkingColorProfile
There is a good reason why all RAW converters use ProPhoto variants internally.
cheers
afx
-
99% of images are viewed on sRGB monitors/prints. Very few commercial printing services print on anything larger than aRGB.
Those stat's come from where? Maybe true for display output. I've never seen any printer's output gamut that isn't larger in some area of color space than sRGB and sRGB is a theoretical color space based on a display. Got piles of output profiles for so called commercial printing services (RGB and CMYK) that don't look anything like sRGB and are larger in many areas of color space than sRGB.
Examples:
(http://digitaldog.net/files/sRGB_vs_SilverPrinters.jpg)
See all those colors falling outside the red (which is sRGB)?
-
Very few commercial printing services print on anything larger than aRGB. Some advanced home inkjet printers may be able to print colors outside of aRGB.
I'm guessing by aRGB he meant Adobe RGB?
-
As you edit, when moving a slider (say Saturation or Vibrance), as soon as you stop seeing the image update (change), back off! You are affecting colors you can't see (Out of Gamut). That's your sign to move carefully.
Just thinking here...
If we work this way, aren't we actually editing in monitor color space?
I guess it's a good argument for wide gamut displays.
-
Ideally the working color space should be as big as you need but no bigger. So how big do you need it to be? It depends on how your images are viewed. 99% of images are viewed on sRGB monitors/prints. Very few commercial printing services print on anything larger than aRGB. Some advanced home inkjet printers may be able to print colors outside of aRGB.
I do fine-art photography and photo-art (photos & digital art combined). I am now producing gallery prints up to 17"X25" on an Epson 3880. Much of my work contains color gradients, shades of blue skies, and deep red sunsets running the entire length of the work. Thus my need to edit 16 bit in the ProPhotoRGB color space to produce accurate color and avoid banding or blocked-up color.
By using a calibrated wide-gamut monitor and editing RAW files in AdobeRGB, I was able to print very close to what I was seeing on my monitor (prior to an output adjustment for substrate and 3500K gallery lighting). That saved a lot of time, ink and paper.
But now with the need to edit in ProPhotoRGB, I was wondering how I can make more accurate edits relative to what the print looks like. How do other fine-art photographers do it? Is editing just guessing from experience, then going through a long drawn-out dance between soft-proofing, re-editing, and then printing lots of proofs? If so, can I streamline this process? Thanks.
-
Those stat's come from where? Maybe true for display output. I've never seen any printer's output gamut that isn't larger in some area of color space than sRGB and sRGB is a theoretical color space based on a display. Got piles of output profiles for so called commercial printing services (RGB and CMYK) that don't look anything like sRGB and are larger in many areas of color space than sRGB.
See all those colors falling outside the red (which is sRGB)?
OP writes about Adobe RGB and ProPhoto RGB.
And you answer with sRGB. Why?
-
OP writes about Adobe RGB and ProPhoto RGB.
And you answer with sRGB. Why?
The OP mentions Adobe RGB (1998) which is a color space. There's no aRGB (but there IS sRGB) which someone else mentioned. Typo?
I'm guessing by aRGB he meant Adobe RGB?
Could be. And even so, there are colors that fall outside that space provided by many output devices in some area's of color space. Do we need to see gamut plots of a modern ink jet next to Adobe RGB (1998)?
-
I do fine-art photography and photo-art (photos & digital art combined). I am now producing gallery prints up to 17"X25" on an Epson 3880.
Hi Rich,
An Epson 3880 inkjet is one of the home/studio printers that can indeed produce colors outside of AdobeRGB (aRGB). So you have two choices: stick with aRGB, ensuring that what you see is faithfully reproduced on paper, but possibly giving up a few tonalities: or move to a larger color space, ensuring that those few extra wide tonalities make it to paper, but possibly resulting in tones on paper that you would not have expected from looking at the monitor. The former will save you ink and paper. The latter is potentially incrementally more 'colorful'.
By using a calibrated wide-gamut monitor and editing RAW files in AdobeRGB, I was able to print very close to what I was seeing on my monitor (prior to an output adjustment for substrate and 3500K gallery lighting). That saved a lot of time, ink and paper. Much of my work contains color gradients, shades of blue skies, and deep red sunsets running the entire length of the work. Thus my need to edit 16 bit in the ProPhotoRGB color space to produce accurate color and avoid banding or blocked-up color.
Imho ProPhoto may indeed produce incrementally more colors, but it will do nothing to help you avoid banding - just the opposite (although you should have no such problems working with 16 bit files coming straight from Raw data). The key is ensuring that your software, printer drivers and printer all work at the highest bit depth possible, ideally all at 16 bits (video drivers, video card, LUTs, monitor panel as well).
But now with the need to edit in ProPhotoRGB...
There is no such set need, it is just a personal choice with the compromises outlined above. I would try ProPhoto/Melissa/BetaRGB and aRGB on a few of your reference prints. If you cannot see a relevant improvement, I would stick with your tried and true workflow.
Cheers,
Jack
-
WHY in raw workflows, you want to use ProPhoto RGB as a working space (IF like Adobe, it's internal color space is using similar sized primaries):
Everything you thought you wanted to know about color gamut
High resolution: http://digitaldog.net/files/ColorGamut.mov
Low Res (YouTube): http://www.youtube.com/watch?v=n0bxSD-Xx-Q
The limitation is our display systems. You can choose to funnel your captured data into something small(er) to see everything but not print it, of not see all the colors and be careful editing and use those colors upon output. The choice is yours.
-
The OP mentions Adobe RGB (1998) which is a color space. There's no aRGB (but there IS sRGB) which someone else mentioned. Typo?
I doubt I'm the only one that regards aRGB as a common acronym for Adobe RGB (1998).
Come on, a degree of pragmatic understanding is more useful here than pure pedantry.
-
I doubt I'm the only one that regards aRGB as a common acronym for Adobe RGB (1998).
As the Chinese proverb says: The first step towards genius is calling things by their proper name.
-
As the Chinese proverb says: The first step towards genius is calling things by their proper name.
So maybe you should type 'The original poster' rather than OP then ?
-
So maybe you should type 'The original poster' rather than OP then ?
It's understandable how that would confuse you.
-
Everything you thought you wanted to know about color gamut
High resolution: http://digitaldog.net/files/ColorGamut.mov
Low Res (YouTube): http://www.youtube.com/watch?v=n0bxSD-Xx-Q
Thanks Andrew. Very helpful, especially the last 30 seconds which addressed my issue.
But if I was to stop moving the adjustment slider when I could no longer see change in the image I'm editing, wouldn't that just be reducing the colors in my image to only the gamut of the monitor, AdobeRGB in my case? It sounds like hard proofing is the only way to test my prints to see if they are indeed capturing the color I remember seeing -- the dark blue gradations in the sky and the deep red-orange hues just after sunset or pre-sunrise.
Looks like I will be stocking up on Epson ink and paper soon. Although it seems like a saw some useful techniques on bracket proofing by John Paul Caponigro that may help. Thanks.
-
But if I was to stop moving the adjustment slider when I could no longer see change in the image I'm editing, wouldn't that just be reducing the colors in my image to only the gamut of the monitor, AdobeRGB in my case?
There could still be colors outside gamut prior to you even moving the sliders. Probably are depending on the image and color space. The idea is that what you see changing as you move the slider should be within display gamut (you see it). But if you move too far, the colors are changing but it's not visible so here's where you want to be kind of careful.
-
It's understandable how that would confuse you.
It doesn't confuse me at all, but it's rather inconsistent to complain about someone using one commonly used acronym, then using acronyms yourself.
You routinely shorten sRGB IEC1966-2.1 to sRGB
-
It doesn't confuse me at all, but it's rather inconsistent to complain about someone using one commonly used acronym, then using acronyms yourself.
You routinely shorten sRGB IEC1966-2.1 to sRGB
I wasn't complaining.
And there are various sRGB profiles!
sRGB Color Space Profile.icm? sRGB IEC1966-2.1? The few sRGB V4 profiles that are available: sRGB_ICC_v4_appearance_beta_displayclass.icc?
http://www.color.org/srgbprofiles.xalter
On this page you will find several different types of sRGB profiles, with information about their intended use.
sRGB v4 Preference
sRGB v4 Appearance (beta)
sRGB v2
-
How about leaving sRGB out of this conversation? The OP's question was about AdobeRGB and ProPhotoRGB.
@andrew: In your video in the last 60 seconds you say something like this: "when moving the slider and you cannot see any change on the screen it means that you have reached the boundary of your screens gamut and you should back up a little."
Now if you have a large gamut screen that has a gamut close to AdobeRGB, then why use ProPhotoRGB? Your edit is in the boundaries of AdobeRGB, isn't it?
-
Now if you have a large gamut screen that has a gamut close to AdobeRGB, then why use ProPhotoRGB? Your edit is in the boundaries of AdobeRGB, isn't it?
As the OP is asking about editing for print, restricting the edits to those that can be displayed on screen could be limiting.
Most printers will exceed aRGB (and monitor gamuts) in some areas, so editing in ProPhoto could make sense if you think you need absolutely all the colour gamut possible.
The difficulty will always be having to edit blind without seeing the final result on screen. Test prints will be the only 100% way to evaluate the final results.
Using OOG indicators in LR or PS will only give an indication in a binary way of where you can't see the actual colour on screen, but using a more specific tool like Gamutvision can help by indicating how far the OOG colours are away from being displayed. Knowing that the colours are only just OOG might be helpful.
-
@andrew: In your video in the last 60 seconds you say something like this: "when moving the slider and you cannot see any change on the screen it means that you have reached the boundary of your screens gamut and you should back up a little."
Now if you have a large gamut screen that has a gamut close to AdobeRGB, then why use ProPhotoRGB? Your edit is in the boundaries of AdobeRGB, isn't it?
Because the source data can be significantly larger than Adobe RGB (1998). Especially for those in raw workflows. For example, if you use Adobe raw processors, the internal color space is using ProPhoto RGB primaries (gamut). Of course the gamut of the scene and device play a role. If you are shooting a field of colorful flowers on a sunny day, very likely that scene and the capture device exceed Adobe RGB (1998). Since there are output devices that also exceed Adobe RGB (1998), you can use that wider gamut data you captured for output. Again, do you limit your data to the lowest common denominator which for many is sRGB and other's Adobe RGB (1998)?
-
Since there are output devices that also exceed Adobe RGB (1998), you can use that wider gamut data you captured for output. Again, do you limit your data to the lowest common denominator which for many is sRGB and other's Adobe RGB (1998)?
Good point. Since the best working color space is as big as you need but no bigger, the question for the OP then becomes:
1) How big is his camera's native color space? No point having anything bigger than this
2) How big is his fine art paper/printer combo color space? No point having anything bigger than this.
The answer to 1) is hard to find without advanced tools. The answer to 2) is probably BetaRGB (http://www.brucelindbloom.com/BetaRGB.html). Both of these spaces are (much, much) smaller than ProPhotoRGB and a bit bigger than AdobeRGB (aRGB).
Cheers,
Jack
-
1) How big is his camera's native color space? No point having anything bigger than this
A D700 - ProPhoto comparison is in the article I already mentioned, also in my signature.
2) How big is his fine art paper/printer combo color space? No point having anything bigger than this.
Wrong.
If your source material is bigger then your final output, then by restricting your pipeline at the beginning to the final output you risk clipping in processing.
Always convert to the bottleneck at the end.
The answer to 1) is hard to find without advanced tools.
I would think this is trivial...
Camera profiles are easy to obtain.
cheers
afx
-
A D700 - ProPhoto comparison is in the article I already mentioned, also in my signature.
Wrong....I would think this is trivial...
Camera profiles are easy to obtain.
Hi afx,
I wonder if the OP was able to find one for his camera. I was not for mine.
If your source material is bigger then your final output, then by restricting your pipeline at the beginning to the final output you risk clipping in processing.
Always convert to the bottleneck at the end.
This is an interesting topic, given three constraints (camera/input, monitor/working, print/output): if your bottleneck is at the output (not always the case), do you convert at the beginning or at the end of the process?
I used to convert at the end, as you seem to suggest. But I often spent a fair amount of time getting things just right (no clipping in the working color space), and then having to get them just right again when I converted to the final color space at the end (clipping!). So now I convert at the beginning: this way I only have to get things just right once :)
Cheers,
Jack
-
I wonder if the OP was able to find one for his camera. I was not for mine.
It used to be that Bibble had them as files, nowadays the majority of the commercial converters don't externalize them anymore ;-(
Still, at least anything mainstream like the D700 is easily grabbed from one of the open source converters.
I used to convert at the end, as you seem to suggest. But I often spent a fair amount of time getting things just right (no clipping in the working color space), and then having to get them just right again when I converted to the final color space at the end (clipping!). So now I convert at the beginning: this way I only have to get things just right once :)
I guess the clipping you see initially is not the working space (if it is ProPhoto), but the limits of your monitor. And the final conversion should not clip, but approximate appropriately, but that depends on the CM engine. And yes, sometimes that approximation sucks and can't be visualized properly in soft proofing, but that is typically the minority case.
cheers
afx
-
1) How big is his camera's native color space? No point having anything bigger than this
2) How big is his fine art paper/printer combo color space? No point having anything bigger than this.
Couple points to address both 1 & 2. Digital cameras don't really have a color gamut (they do have a color mixing function). There is some assumed internal color space a converter has to start with and only the manufacturer would really know this although with the right equipment one could end up with the spectral sensitivities and get very close. Then there's the internal color space of the converter to process our data.
Building an ICC profile doesn't tell us all that much since the target has a gamut. The data feed to the profiler is output referred (processed).
The scene has a gamut. Point the same capture device at a gray card and a sunlit scene of colorful flowers, you end up with different results.
Bottom line using say an Adobe raw converter is that at some point, we're dealing with ProPhoto RGB primaries and thus a gamut. If you work with Adobe RGB (1998), you will clip colors from some (how many?) captures. I think that's illustrated in the video. It's easy to examine this in ACR too. IF your goal is to render all the possible color, you don't want to select Adobe RGB (1998). IF your goal is to see all the colors on-screen, get a wide gamut display and encode in Adobe RGB (1998).
There's a big disconnect in the shape of RGB working spaces and output spaces so keep that in mind. There are vastly wider gamut that can be defined in something like ProPhoto RGB than you could possibly output, true. But we have to live with a disconnect between the simple shapes of RGB working space and the more complex shapes of output color spaces to the point we're trying to fit round pegs in square holes. To do this, you need a much larger square hole. Simple matrix profiles of RGB working spaces when plotted 3 dimensionally illustrate that they reach their maximum saturation at high luminance levels. The opposite is seen with print (output) color spaces. Printers produce color by adding ink or some colorant, working space profiles are based on building more saturation by adding more light due to the differences in subtractive and additive color models. To counter this, you need a really big RGB working space like ProPhoto RGB, again due to the simple size and to fit the round peg in the bigger square hole. Their shapes are simple and predictable. Then there is the issue of very dark colors of intense saturation which do occur in nature and we can capture with many devices. Many of these colors fall outside Adobe RGB (1998) and when you encode into such a space, you clip the colors to the degree that smooth gradations become solid blobs in print, again due to the dissimilar shapes and differences in how the two spaces relate to luminance.
The gamut of the output device you aim for today, and what you might aim for tomorrow is unknown. So again, why encode in a working space that limits this?
-
There could still be colors outside gamut prior to you even moving the sliders. Probably are depending on the image and color space. The idea is that what you see changing as you move the slider should be within display gamut (you see it). But if you move too far, the colors are changing but it's not visible so here's where you want to be kind of careful.
Doesn't LR let you display both print and display out of gamut areas at the same time? At least you would know when you were in the zone between display gamut and printer gamut, though you wouldn't be able to tell what the actual color was. Never mind....
-
I remember thinking I was at the top of the heap when I owned my Sony Artisan monitor. A few years later, my current display is significantly better (by a factor of 10?). It seems that every year that goes by, slivers of improved accuracy and output are built into displays and printers. Knowing that there might be a "someday" when equipment will allow accurate reproduction of ANY color space motivates me to do my current editing in the largest available color space.
-
Thanks for your feedback, Andrew, noted.
@Tim_Smith: I hear you, but there isn't an old image I open that I do not wish to tweak with all the latest and greatest.
Cheers,
Jack
-
Thanks for your feedback, Andrew, noted.
@Tim_Smith: I hear you, but there isn't an old image I open that I do not wish to tweak with all the latest and greatest.
Cheers,
Jack
Same here. You would not believe what varying black point levels between old and new display's does to your perception editing role off shadow detail into black.
Just bought me an LG 27ea63 at Best Buy recently that was factory calibrated with a very expensive high end Color Analyzer (come a long way at least 5 years back finding that at Best Buy). The Colormunki Display didn't have to do much work after the calibration.
I go back to images I edited on the Dell 2209WA which wasn't factory calibrated and where I had to calibrate it with the old original i1Display with i1Match software which made the black levels seem kind of washed out. Now my images viewed on the newer LG seem a bit more contrasty and clammy looking.
Also for some strange reason the Clarity slider editing on the LG behaves noticeably different from what I remember editing on the Dell. It's not as harsh like a High Pass filter effect I used to get with the Dell.
Guess it pays to have a truly linear display. It's just now it doesn't have to cost so much.