If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#141
|
|||
|
|||
Color management in Windows
On 2/21/2017 4:36 PM, Eric Stevens wrote:
On Tue, 21 Feb 2017 10:27:17 -0500, Neil wrote: On 2/20/2017 5:19 PM, Eric Stevens wrote: On Mon, 20 Feb 2017 15:47:14 -0500, Neil wrote: On 2/20/2017 2:18 PM, Eric Stevens wrote: On Mon, 20 Feb 2017 10:07:27 -0500, Neil wrote: On 2/19/2017 10:05 PM, Eric Stevens wrote: On Sun, 19 Feb 2017 18:36:36 -0500, Neil wrote: On 2/19/2017 5:26 PM, Eric Stevens wrote: rOn Sun, 19 Feb 2017 09:43:49 -0500, Neil wrote: [...] It may help to narrow things down if you could perform some basic tests. For example, if you have access to a color calibration file (basically a chart that prints scales and a gamut wheel) and reference print, you can see whether the printer is reproducing that chart correctly. If you have a colorimeter, you can measure the printed output and compare that with what you see on the screen, understanding that the CMYK+ gamut will have some differences from the RGB gamut. Some settings seem to be more or less right as far as color is concerned, but that is not my present concern. It's the fact that changes in settings cause changes in color in ways I do not understand. Can you give some specifics? What settings? What kinds of changes do you see? Basically changes in hue and saturation. Different color space settings will result in different color gamuts in a print, which is to be expected. Yep. What is Windows doing in this respect? Nothing new, AFAICT. I hope you're not expecting an OS to make all color spaces look the same in print. That would only happen for a very narrow range of colors and intensities. It's doing something new since Vista. That's where all kinds of bells and whistles were added and I can find nothing which tells me how it works. I'm not seeing any of these issues on my systems, so I can only hope that you find out what is going on with yours. Are you printing with more than one color space? I mean 'color space', not just printer profile. Yes, I do use color spaces dependent on the type of job. My work also requires me to calibrate my proofing printers to match the final output devices, so "standard" printer profiles aren't of much use to me. -- best regards, Neil |
#142
|
|||
|
|||
Color management in Windows
In article , PeterN
wrote: At a more advanced facility (probably not Costco) you might be able to specify wide gamut printing using the colorspace you specify and specialty papers or media you might specify. Those will also have to have matched icc profiles. and they'll tell you exactly what to do with the images if you aren't sure. Fighting temptation. for your images, i know what i'd tell you to do with them. |
#143
|
|||
|
|||
Color management in Windows
In article , Eric Stevens
wrote: That statement encompasses what you have just written about "photoshop's conversion to its own working space" which is set by 'EditColorsettings'. This gives you the choice of ProPhoto plus a zillion others. But I expect you already know tat. no it doesn't, and that suggests that you're *very* confused about what's going on and what needs to happen. Nothing to do with what "needs" to happen What I said is absolutely correct. Whether or not any of the available settings makes sense is another matter entirely. while it might be technically correct (although not exactly), the reality is that it's entirely separate from the two menu commands peter mentioned. But PeterN wasn't mentioning them as menu commands. yes he was, given that he's mentioned using those on multiple occasions. He raised them as things which can be done, as functions. Whether you do it by clicking a button on the window or by letting the machine do the work is quite another matter. nope. photoshop is designed to do the proper thing in normal use. you're trying to micromanage every step without fully understanding what each step does. You are wriggling. Clutching at straws in an attempt to get off a hook of your own making. nope. at the end of the day, you are getting results you do not like, so something is *not* set correctly. the goal is to find out what that is and fix it. Changing the subject away from your own blooper. no blooper whatsoever. ask yourself again who is having problems with prints and who is not. then ask yourself if you want to resolve it or not. you *can* override what photoshop does when necessary, but opening a raw image (or jpeg) and then printing it to a local printer is not one of them. in fact, that's one of the most common use cases there is, which means it should work as expected with little to no fuss. 'Color Space' remember? I was raising questions about the whys and wherefors of the way that the Windows Color Management System handles different color spaces for printing. you're overthinking things. are you planning on writing a windows colour management utility or do you want good prints? at this point, who knows how you have things set, so start with resetting everything and start from known defaults: https://helpx.adobe.com/photoshop/using/preferences.html https://helpx.adobe.com/lightroom/he...s-lightroom.ht ml next, open an image (raw or jpeg), adjust onscreen as desired, then print as you normally would. let *either* the printer *or* photoshop manage colours (not both). evaluate results. I'm way past that. apparently not, given that you're not getting results you want. don't worry about converting/assigning, printer space, srgb/argb or anything else at this time. save that for later. stop overthinking things. photoshop (or lightroom) has you covered. what you'll more than likely find is that the defaults do a damned fine job. What goes on in your little brain? This whole thread exists because it is NOT doing a fine job. only because you're doing something wrong. the goal is to identify why and then remedy it. the best way to do that is start over from scratch, leaving things at their defaults, and work your way up, *not* dive in head first and start tweaking every option. you also didn't say which books you've read, but anything by fraser or schewe is good. if you have any books by marguilis, the best thing to do is to use them to build a fire to save on heating costs. that's probably low this time of year for you, so maybe do that in a couple of months when it gets cold. In no particular order: 3) Color Management - Fraser, Murphy & Bunting 5) The Digital Negative - Jeff Schewe 6) The Digital Print - Jeff Schewe pay attention to those. |
#144
|
|||
|
|||
Color management in Windows
On Fri, 17 Feb 2017 14:06:28 +1300, Eric Stevens
wrote: I have a camera (Nikon D750) with a color space that encompasses AdobeRGB. I use software (Photoshop etc) with an enormous Profile Connection Space (PCS) which can variously swallow or spit out virtually any real world color space. I have a printer (Epson Sure Color P800) which can accept images in the conventional sRB and the wider AdobeRGB. I now have a screen with a gamut which is AdobeRGB for all practical purposes. So - whacko! - I can now photograph, see, edit and print images in more glorious color than ever before. Except that I don't seem to be able to, at least without experiencing color distortions which result in garish colors. At this stage I pointed the finger squarely at Windows which I have many times read is limited to an sRGB color space. But surely this can't be so. This is the 21st century and the days of the kerosine-fired magic lantern are nearly gone. Surely Microsoft have seen the future that is coming? I have been exploring Windows color management and I have found the the that since Windows 2000, Windows has used the Image Color Management (ICM) which as far as I can tell *is* limited to the sRGB color space by default. See http://tinyurl.com/j938m4m or https://msdn.microsoft.com/en-us/lib...(v=vs.85).aspx However, with Windows Vista, MS introduced the Windows Color System (WCS). According to https://en.wikipedia.org/wiki/Windows_Color_System "Windows Color System features a Color Infrastructure and Translation Engine (CITE) at its core" which sounds remarkably like like a working color space and color translation engine supplied by Microsoft. I haven't yet made sense of what little information I have been able to find about this but it sounds as though whether or not CMS is used depends on the supplier of image processing softwa do nothing and your color processing will fall back on the default sRGB. However if the software provider supplies the necessary software and profiles you can use any color space you like. If my understanding is correct it is likely that whether or not I have access to a wider color space than sRGB depens on Adobe providing and utilising the necessary software. The question is, am I right or have I got myself into a knot somewhere? I didn't have a knot but I did have a tangle. I discovered that I had Photoshop set not to Perceptual but Relative Colorimetric. That was not my preferred choice. However ... I decided to make a clean pass and an undoubtedly Adobe RGB test image. My D750 is set for Adobe RGB (and I don't know whether or not this affects raw images) so I made up a test image from four recent photographs. See https://dl.dropboxusercontent.com/u/...st%20Print.jpg I picked the images I did as I wanted to explore the corners of the RGB color triangle. I then made prints on A4 size Epson Premium Glossy paper using the following protocols: (1) Photoshop Manages colors. (2) Printer manages colors, printer set to sRGB (3) " " " " " Adobe RGB (4) " " " " " ICM (5) " " " " " Photo Enhance These were all unintentionally run in relative colorimetric mode. All five prints were different and there were clear winners and losers as far as the finished images were concerned. I spent some time poring over the images with a magnifying glass and in the end I could not really tell whether I was looking at gamut variations or merely differences in saturation. Then I recalled that I had https://dl.dropboxusercontent.com/u/...RGB_colors.png which contains all 24 bit colors. I decided to use this as a test image and I ran all five of the tests listed above as before. By using multiple print passes I was able to get five tests on three sheets of paper. It was at this stage that I discovered that I had been running in relative colorimetric mode rather than perceptual. I muttered a rude word and ran the tests all over again. Anyway, you can find scans or the results in the following URLs. You will have to be patient, they are slow to download but are most edifying when you have them up on different tabs on your browser. First, the perceptual intent. https://dl.dropboxusercontent.com/u/...Percep%201.jpg https://dl.dropboxusercontent.com/u/...Percep%202.jpg https://dl.dropboxusercontent.com/u/...Percep%203.jpg .... and now the relative colorimetric https://dl.dropboxusercontent.com/u/...93/Rel%201.jpg https://dl.dropboxusercontent.com/u/...93/Rel%202.jpg https://dl.dropboxusercontent.com/u/...93/Rel%203.jpg It's clear that perceptual intent is the way to go. It's the only one I will discuss in detail but more or less the same comments generally apply to the relative colorimetric prints. It's also clear is that there are marked differences between the various outputs. When I ran my test image https://dl.dropboxusercontent.com/u/...st%20Print.jpg I found that there is not much to choose between Photoshop managing colors and the printer managing colors when set to sRGB. Using the printer to manage colours when set to Adobe RGB produces a slightly subdued image. The colours are not quite so intense and there are subtle shifts in hue (as one would expect). It is noticeable that the Adobe RGB yellows have less red/orange in them. Setting the printer to ICM produced images which seemed slightly dark and subdued. Setting the printer to Photo Enhance produced an image as bright and cheerful as the original. I noticed that the in the image at the top-right the grain in the floor stood out more clearly than in any of the others and I suspect that Photo Enhance includes a number of image-manipulating tricks in addition to playing around with the color. The prints of the test image were most enlightening. Letting Photoshop manage color produced a print with what was clearly the highest level of saturation. When the printer was managing colors setting the printer to AdobeRGB produced noticeably greener greens than when it was set to sRGB. ICM produced a slightly grayer version of Adobe RGB while Photo Enhance produced a print with high luminosity. When Photoshop managed colors or when the printer was set to either sRGB or Photo Enhance, or even Adobe RGB there was a noticeable touch of yellow at the bottom right-hand corner of most of the cells comprising the test image. When Photoshop managed colors or when the printer was set to sRGB the yellow was associated with a diagonal streak running from the bottom right to the top left of each cell. Clearly the merging of the R, G and B to make intermediate colors is not straight forward. The image printed when the printer is set to ICM is free of all these artifacts and the cells appear to be a straight forward linear blending of the R, G and B. There are all sorts of conclusions which can be drawn from all of this, not the least of which is that if one wishes to print to a larger gamut it is not just a matter of taking an existing image and printing. It is likely that the color and lighting of the whole image will have to be reworked. More later (much later) :-) -- Regards, Eric Stevens |
#145
|
|||
|
|||
Color management in Windows
On 2017-02-25 03:26:00 +0000, Eric Stevens said:
On Fri, 17 Feb 2017 14:06:28 +1300, Eric Stevens wrote: I have a camera (Nikon D750) with a color space that encompasses AdobeRGB. I use software (Photoshop etc) with an enormous Profile Connection Space (PCS) which can variously swallow or spit out virtually any real world color space. I have a printer (Epson Sure Color P800) which can accept images in the conventional sRB and the wider AdobeRGB. I now have a screen with a gamut which is AdobeRGB for all practical purposes. So - whacko! - I can now photograph, see, edit and print images in more glorious color than ever before. Except that I don't seem to be able to, at least without experiencing color distortions which result in garish colors. At this stage I pointed the finger squarely at Windows which I have many times read is limited to an sRGB color space. But surely this can't be so. This is the 21st century and the days of the kerosine-fired magic lantern are nearly gone. Surely Microsoft have seen the future that is coming? I have been exploring Windows color management and I have found the the that since Windows 2000, Windows has used the Image Color Management (ICM) which as far as I can tell *is* limited to the sRGB color space by default. See http://tinyurl.com/j938m4m or https://msdn.microsoft.com/en-us/lib...(v=vs.85).aspx However, with Windows Vista, MS introduced the Windows Color System (WCS). According to https://en.wikipedia.org/wiki/Windows_Color_System "Windows Color System features a Color Infrastructure and Translation Engine (CITE) at its core" which sounds remarkably like like a working color space and color translation engine supplied by Microsoft. I haven't yet made sense of what little information I have been able to find about this but it sounds as though whether or not CMS is used depends on the supplier of image processing softwa do nothing and your color processing will fall back on the default sRGB. However if the software provider supplies the necessary software and profiles you can use any color space you like. If my understanding is correct it is likely that whether or not I have access to a wider color space than sRGB depens on Adobe providing and utilising the necessary software. The question is, am I right or have I got myself into a knot somewhere? I didn't have a knot but I did have a tangle. I discovered that I had Photoshop set not to Perceptual but Relative Colorimetric. That was not my preferred choice. However ... I decided to make a clean pass and an undoubtedly Adobe RGB test image. My D750 is set for Adobe RGB (and I don't know whether or not this affects raw images) so I made up a test image from four recent photographs. See https://dl.dropboxusercontent.com/u/...st%20Print.jpg I picked the images I did as I wanted to explore the corners of the RGB color triangle. I then made prints on A4 size Epson Premium Glossy paper using the following protocols: (1) Photoshop Manages colors. (2) Printer manages colors, printer set to sRGB (3) " " " " " Adobe RGB (4) " " " " " ICM (5) " " " " " Photo Enhance These were all unintentionally run in relative colorimetric mode. All five prints were different and there were clear winners and losers as far as the finished images were concerned. I spent some time poring over the images with a magnifying glass and in the end I could not really tell whether I was looking at gamut variations or merely differences in saturation. Then I recalled that I had https://dl.dropboxusercontent.com/u/...RGB_colors.png which contains all 24 bit colors. I decided to use this as a test image and I ran all five of the tests listed above as before. By using multiple print passes I was able to get five tests on three sheets of paper. It was at this stage that I discovered that I had been running in relative colorimetric mode rather than perceptual. I muttered a rude word and ran the tests all over again. Anyway, you can find scans or the results in the following URLs. You will have to be patient, they are slow to download but are most edifying when you have them up on different tabs on your browser. First, the perceptual intent. https://dl.dropboxusercontent.com/u/...Percep%201.jpg https://dl.dropboxusercontent.com/u/...Percep%202.jpg https://dl.dropboxusercontent.com/u/...Percep%203.jpg ... and now the relative colorimetric https://dl.dropboxusercontent.com/u/...93/Rel%201.jpg https://dl.dropboxusercontent.com/u/...93/Rel%202.jpg https://dl.dropboxusercontent.com/u/...93/Rel%203.jpg It's clear that perceptual intent is the way to go. It's the only one I will discuss in detail but more or less the same comments generally apply to the relative colorimetric prints. It's also clear is that there are marked differences between the various outputs. When I ran my test image https://dl.dropboxusercontent.com/u/...st%20Print.jpg I found that there is not much to choose between Photoshop managing colors and the printer managing colors when set to sRGB. Using the printer to manage colours when set to Adobe RGB produces a slightly subdued image. The colours are not quite so intense and there are subtle shifts in hue (as one would expect). It is noticeable that the Adobe RGB yellows have less red/orange in them. Setting the printer to ICM produced images which seemed slightly dark and subdued. Setting the printer to Photo Enhance produced an image as bright and cheerful as the original. I noticed that the in the image at the top-right the grain in the floor stood out more clearly than in any of the others and I suspect that Photo Enhance includes a number of image-manipulating tricks in addition to playing around with the color. The prints of the test image were most enlightening. Letting Photoshop manage color produced a print with what was clearly the highest level of saturation. When the printer was managing colors setting the printer to AdobeRGB produced noticeably greener greens than when it was set to sRGB. ICM produced a slightly grayer version of Adobe RGB while Photo Enhance produced a print with high luminosity. When Photoshop managed colors or when the printer was set to either sRGB or Photo Enhance, or even Adobe RGB there was a noticeable touch of yellow at the bottom right-hand corner of most of the cells comprising the test image. When Photoshop managed colors or when the printer was set to sRGB the yellow was associated with a diagonal streak running from the bottom right to the top left of each cell. Clearly the merging of the R, G and B to make intermediate colors is not straight forward. The image printed when the printer is set to ICM is free of all these artifacts and the cells appear to be a straight forward linear blending of the R, G and B. There are all sorts of conclusions which can be drawn from all of this, not the least of which is that if one wishes to print to a larger gamut it is not just a matter of taking an existing image and printing. It is likely that the color and lighting of the whole image will have to be reworked. More later (much later) :-) In the light of your discovery, the following might be of interest. The thing I would take away from all of that is, it is going to be your image which dictates which of the intent options is going to work best for any particular print. Typically "Absolute Colorimetric" with many images can deliver so much out of gamut that the resulting print can seem insultingly damaged. I have found that for most of my prints "Perceptual" delivers the most pleasing results, and while "Relative" can deliver nice results with specific images, it mostly renders results with colors alien to my intent. http://www.color-management-guide.com/conversion-mode-perceptual-relative-colorimetric-rendering-intent.html http://www.cambridgeincolour.com/tutorials/color-space-conversion.htm http://dba.med.sc.edu/price/irf/Adobe_tg/manage/renderintent.html -- Regards, Savageduck |
#146
|
|||
|
|||
Color management in Windows
On Fri, 17 Feb 2017 14:06:28 +1300, Eric Stevens
wrote: I have a camera (Nikon D750) with a color space that encompasses AdobeRGB. I use software (Photoshop etc) with an enormous Profile Connection Space (PCS) which can variously swallow or spit out virtually any real world color space. I have a printer (Epson Sure Color P800) which can accept images in the conventional sRB and the wider AdobeRGB. I now have a screen with a gamut which is AdobeRGB for all practical purposes. So - whacko! - I can now photograph, see, edit and print images in more glorious color than ever before. Except that I don't seem to be able to, at least without experiencing color distortions which result in garish colors. At this stage I pointed the finger squarely at Windows which I have many times read is limited to an sRGB color space. But surely this can't be so. This is the 21st century and the days of the kerosine-fired magic lantern are nearly gone. Surely Microsoft have seen the future that is coming? I have been exploring Windows color management and I have found the the that since Windows 2000, Windows has used the Image Color Management (ICM) which as far as I can tell *is* limited to the sRGB color space by default. See http://tinyurl.com/j938m4m or https://msdn.microsoft.com/en-us/lib...(v=vs.85).aspx However, with Windows Vista, MS introduced the Windows Color System (WCS). According to https://en.wikipedia.org/wiki/Windows_Color_System "Windows Color System features a Color Infrastructure and Translation Engine (CITE) at its core" which sounds remarkably like like a working color space and color translation engine supplied by Microsoft. I haven't yet made sense of what little information I have been able to find about this but it sounds as though whether or not CMS is used depends on the supplier of image processing softwa do nothing and your color processing will fall back on the default sRGB. However if the software provider supplies the necessary software and profiles you can use any color space you like. If my understanding is correct it is likely that whether or not I have access to a wider color space than sRGB depens on Adobe providing and utilising the necessary software. The question is, am I right or have I got myself into a knot somewhere? I've continued to pursue this matter and while I haven't got to the bottom of the way that Windows handles color I may have got closer to the root of my problem. It seems that lower priced wide gamut monitors us GB-LED technology and this, unless properly calibrated gives rise to garish greens and reds. Calibrating properly seems to entail the use of an i1 which means that my Spyder5 won't do. See the AnandaTech review of the Dell U3014 http://www.anandtech.com/show/6890/d...014-lcd-review for an explanation of how it all works. "Of course you also have the assumption that more colors = better, right? Well, unfortunately that isn’t the case. If you only have a larger gamut and not software that understands how to use that gamut, what you get are colors that are further outside of the gamut than they should be. Greens are too green, reds are too red, and everything looks like a badly calibrated OLED smartphone. Anyone buying something like the U3014 is going to need to have accurate colors in any colorspace, and the Dell offers an sRGB mode as well." No doubt there will be more to follow. -- Regards, Eric Stevens |
#147
|
|||
|
|||
Color management in Windows
On Thu, 30 Mar 2017 13:15:35 +1300, Eric Stevens
wrote: I've continued to pursue this matter and while I haven't got to the bottom of the way that Windows handles color I may have got closer to the root of my problem. It seems that lower priced wide gamut monitors us GB-LED technology and this, unless properly calibrated gives rise to garish greens and reds. Calibrating properly seems to entail the use of an i1 which means that my Spyder5 won't do. See the AnandaTech review of the Dell U3014 http://www.anandtech.com/show/6890/d...014-lcd-review for an explanation of how it all works. "Of course you also have the assumption that more colors = better, right? Well, unfortunately that isn’t the case. If you only have a larger gamut and not software that understands how to use that gamut, what you get are colors that are further outside of the gamut than they should be. Greens are too green, reds are too red, and everything looks like a badly calibrated OLED smartphone. Anyone buying something like the U3014 is going to need to have accurate colors in any colorspace, and the Dell offers an sRGB mode as well." No doubt there will be more to follow. And there is. See https://pcmonitors.info/articles/the...ed-backlights/ -- -- Regards, Eric Stevens |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
color management use-cases | Dale[_4_] | In The Darkroom | 2 | February 1st 14 08:13 AM |
Color Management-Color Spyder | ____ | Digital SLR Cameras | 16 | October 12th 08 08:13 AM |
Color Management Process | Gordo | Digital Photography | 24 | January 5th 06 12:35 PM |
So confused about color management Help! | paul | Digital Photography | 14 | January 30th 05 05:16 PM |
Color Management | Gary Eickmeier | Digital Photography | 64 | November 30th 04 12:00 PM |