Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: BrianToth on April 17, 2018, 10:30:53 pm
-
I'm sure this is probably a Photoshop Color Management 101 question, but since my last basic question (http://forum.luminous-landscape.com/index.php?topic=124273.0) resulted in me discovering that my Canon glossy paper profiles were broken, I figured it can't hurt to make sure I properly understand a few things regarding Absolute Colorimetric rendering intents.
Quick context for the questions: My goal has been to duplicate the appearance of some old photos as close to the original as possible. I think I made the mistake of asking a too-detailed question about color casts in my scans (http://forum.luminous-landscape.com/index.php?topic=124117.0) instead of breaking it down into core components first and checking my understanding of each phase. My follow-up question about the broken rendering intents of my Canon paper profiles helped me realize that printing my scans using a functioning Absolute rendering intent improved my results by emulating the paper white of the source photos onto the paper that I have at my disposal.
As a test, I'm currently working with a simulated ColorChecker image from BabelColor (http://server.briantoth.com/luminous/ColorChecker_Lab_from_X-Rite_16bit_AfterNov2014.tif), since its Lab values match the reference file of the physical ColorChecker Mini that I have and I can tell what they're supposed to be.
In Photoshop under Edit > Color Settings
Does this just control how the current image profile is translated into my display profile and what the Info panel numbers represent? The default is Relative, but when doing reproduction-style work, is it best to leave this in Absolute Colorimetric to make it easier to check that profiles are working properly?
Using Edit > Convert to Profile vs Soft Proofing
If I take this ColorChecker image and convert from Lab to sRGB using Absolute Colorimetric for example, most of the patches stay the same since they're almost all within the sRGB gamut, the cyan shifts a bit. What I'm confused by is that if I convert from Lab to most of my paper profiles using Absolute Colorimetric, is that the patches will obviously shift. For example: If I choose a semi-gloss paper, the L of the white patch goes down, but the RGB values go up and the image gets brighter on the screen as a result. It seems counter-inuitive. Does this have something to do with how it preserves the white point of the source relative to the destination profile? (I can include photos if my description is gibberish.)
If I take the ColorChecker and proof it using Absolute Intent, unlike Converting the Lab values all stay the same but the image on the screen will clearly change depending on my settings for simulating paper color, etc. Is the proof just trying to emulate what the actual Lab values will look like when printed vs. the Convert reflecting the actual changes that must take place to preserve the appearance? I just want to be clear on the distinction going forward.
Finally… evaluating the print
If I print the ColorChecker image using Absolute Colorimetric, I think I read in a few posts that it's possible to sample the output using a spectrophotometer and compare to the original. Is that correct? If so, what should the numbers be a match to – the reference Lab values, or the values after converting to the paper profile using an Absolute intent? Or can I really only just view it by eye to evaluate?
The sample I printed yesterday on my Pro-10 (from Windows with a working paper profile) using Absolute Colorimetric looks nice and vibrant with a slightly off-white white patch and the greys appear pretty neutral, though my glossy print vs the physical ColorChecker obviously reflect light differently.
For the heck of it, I did use spotread from ArgyllCMS to measure my printed patches…
Printed target using Absolute Colorimetric:
Canon glossy paper white: 94.055, -0.611, -2.047
ColorChecker white: 92.887, -1.425, 3.816
ColorChecker grey (under the yellow): 47.370, -0.039, -0.405
ColorChecker black: 16.089, 0.091, -1.236
Target Reference Lab values:
White: 95.19, -1.030, 2.930
Grey: 50.76, -0.116, 0.141
Black: 20.64, 0.070, -0.460
Target converted to the glossy paper profile using Absolute Colorimetric:
White: 93.31, -1.147, 2.273
Grey: 51.04, 0.164, -0.444
Black: 20.45, -0.077, -1.827
Thanks for any feedback!
*** 5/12/2018 - Edited the title
-
Let’s start with the color settings and move on:
Photoshop CC's Color Settings & the Convert to Profile and Assign Profile command.
This new video covers everything you thought you wanted to know about the Photoshop Color Setting dialog. It also discusses the Convert to Profile Command and the Assign Profile Command. Photoshop CC 2017 is used in this video and it updates the video on this subject I Published on June 28, 2012.
High Rez: http://digitaldog.net/files/PhotoshopColorSettings.mp4
Low Rez (YouTube): https://www.youtube.com/watch?v=6JaHOGDK5OI
-
I'm sure this is probably a Photoshop Color Management 101 question, but since my last basic question (http://forum.luminous-landscape.com/index.php?topic=124273.0) resulted in me discovering that my Canon glossy paper profiles were broken, I figured it can't hurt to make sure I properly understand a few things regarding Absolute Colorimetric rendering intents.
Quick context for the questions: My goal has been to duplicate the appearance of some old photos as close to the original as possible. I think I made the mistake of asking a too-detailed question about color casts in my scans (http://forum.luminous-landscape.com/index.php?topic=124117.0) instead of breaking it down into core components first and checking my understanding of each phase. My follow-up question about the broken rendering intents of my Canon paper profiles helped me realize that printing my scans using a functioning Absolute rendering intent improved my results by emulating the paper white of the source photos onto the paper that I have at my disposal.
This is exactly what Abs. Col. Intent was designed for.
As a test, I'm currently working with a simulated ColorChecker image from BabelColor (http://server.briantoth.com/luminous/ColorChecker_Lab_from_X-Rite_16bit_AfterNov2014.tif), since its Lab values match the reference file of the physical ColorChecker Mini that I have and I can tell what they're supposed to be.
In Photoshop under Edit > Color Settings
Does this just control how the current image profile is translated into my display profile and what the Info panel numbers represent? The default is Relative, but when doing reproduction-style work, is it best to leave this in Absolute Colorimetric to make it easier to check that profiles are working properly?
This should just be set on Relative Colorimetric. It doesn't affect printing or soft proofing.
Using Edit > Convert to Profile vs Soft Proofing
If I take this ColorChecker image and convert from Lab to sRGB using Absolute Colorimetric for example, most of the patches stay the same since they're almost all within the sRGB gamut, the cyan shifts a bit. What I'm confused by is that if I convert from Lab to most of my paper profiles using Absolute Colorimetric, is that the patches will obviously shift. For example: If I choose a semi-gloss paper, the L of the white patch goes down, but the RGB values go up and the image gets brighter on the screen as a result. It seems counter-inuitive. Does this have something to do with how it preserves the white point of the source relative to the destination profile? (I can include photos if my description is gibberish.)
Don't convert to paper profiles. There are very few reasons to do so unless needed to send a converted image file to an outside printer. Also, the way Photoshop displays images converted to a printer profile is messed up, To see what the printer prints, use the soft proof dialog.
If I take the ColorChecker and proof it using Absolute Intent, unlike Converting the Lab values all stay the same but the image on the screen will clearly change depending on my settings for simulating paper color, etc. Is the proof just trying to emulate what the actual Lab values will look like when printed vs. the Convert reflecting the actual changes that must take place to preserve the appearance? I just want to be clear on the distinction going forward.
If you soft proof using Abs. Col., select the show paper color. Otherwise the image will brighten overall. This is because without the paper color checked, it uses Rel. Col. in the reversal process and will boost brightness. The lower the paper's white point, the more the boost.
Finally… evaluating the print
If I print the ColorChecker image using Absolute Colorimetric, I think I read in a few posts that it's possible to sample the output using a spectrophotometer and compare to the original. Is that correct? If so, what should the numbers be a match to – the reference Lab values, or the values after converting to the paper profile using an Absolute intent? Or can I really only just view it by eye to evaluate?
After converting to a paper profile, in order to correctly view the colors printed, make a copy and convert it back to Lab us Abs. Col. This avoids any interaction with the color settings menu which can result in incorrect values when displaying an image in the printer profile colorspace. The values shown after converting represent the profile's A2B tables estimate of the color printed. If they are off more than 1 dE then it's a safe bet that color is out of the printer's gamut.
A spectro is the best way to test it.
The sample I printed yesterday on my Pro-10 (from Windows with a working paper profile) using Absolute Colorimetric looks nice and vibrant with a slightly off-white white patch and the greys appear pretty neutral, though my glossy print vs the physical ColorChecker obviously reflect light differently.
For the heck of it, I did use spotread from ArgyllCMS to measure my printed patches…
Printed target using Absolute Colorimetric:
Canon glossy paper white: 94.055, -0.611, -2.047
ColorChecker white: 92.887, -1.425, 3.816
ColorChecker grey (under the yellow): 47.370, -0.039, -0.405
ColorChecker black: 16.089, 0.091, -1.236
These are very low L* numbers for a CC. You might want to have your spectro calibrated. The a* and b* (neutrality) are within reason.
Target Reference Lab values:
White: 95.19, -1.030, 2.930
Grey: 50.76, -0.116, 0.141
Black: 20.64, 0.070, -0.460
Target converted to the glossy paper profile using Absolute Colorimetric:
White: 93.31, -1.147, 2.273
Grey: 51.04, 0.164, -0.444
Black: 20.45, -0.077, -1.827
These are consistent aside from the anomaly of the instrument's low L* reading of CC white patches.
-
If you don't know what or why you'd use Absolute Colorimetric rendering intent, don't use it ;D .
-
A spectro is the best way to test it.These are very low L* numbers for a CC. You might want to have your spectro calibrated. The a* and b* (neutrality) are within reason.These are consistent aside from the anomaly of the instrument's low L* reading of CC white patches.
Thanks Doug. Do you suppose it's possible my readings of my prints could be affected by brighteners in the Canon Glossy II paper I'm using? I have a brand new i1Studio (formerly ColorMunki) spectrophotometer, which is obviously an "entry level" piece of equipment, but I would hope it to be within certain specs and not too wildly off. I did recently read though that the i1Studio has a UV filter and might be useless with papers that use OBAs. I don't have another spectrophotometer to test against, and all I have to try to sample with "known" reference values are my targets. If I attempt a spot-check on my physical ColorChecker mini with the matte patches…
Argyll Spotread vs. X-Rite Reference File
White: 96.269, -0.134, 1.139 vs. 95.19, -1.030, 2.930 = CIE 2000: 2.14
Grey: 50.718, -0.296, -0.331 vs. 50.76, -0.116, 0.141 = CIE 2000: 0.54
Black: 20.586, 0.343, -0.320 vs. 20.64, 0.070, -0.460 = CIE 2000: 0.43
Though this isn't exactly a perfect test either as there's a margin of error between X-Rite's target manufacturing, and their reading, and my readings even if everything were working. I would really need an individually measured target and reference I would think. But the results are closer than my readings from the paper.
-
Forgot to do the obvious test, and just measure my reading of the paper vs the profile's whitepoint:
My Argyll spotread on Canon Glossy II paper: 94.055, -0.611, -2.047
ICC profile's white point: 94.19024, -0.5772792, -2.4742979
100% Lab in Photoshop with relative conversion to paper profile: 94.19, -0.562, -2.491
So, not sure what to make of things now, looks like my readings aren't too bad?
-
Thanks Doug. Do you suppose it's possible my readings of my prints could be affected by brighteners in the Canon Glossy II paper I'm using? I have a brand new i1Studio (formerly ColorMunki) spectrophotometer, which is obviously an "entry level" piece of equipment, but I would hope it to be within certain specs and not too wildly off. I did recently read though that the i1Studio has a UV filter and might be useless with papers that use OBAs. I don't have another spectrophotometer to test against, and all I have to try to sample with "known" reference values are my targets. If I attempt a spot-check on my physical ColorChecker mini with the matte patches…
Argyll Spotread vs. X-Rite Reference File
White: 96.269, -0.134, 1.139 vs. 95.19, -1.030, 2.930 = CIE 2000: 2.14
Grey: 50.718, -0.296, -0.331 vs. 50.76, -0.116, 0.141 = CIE 2000: 0.54
Black: 20.586, 0.343, -0.320 vs. 20.64, 0.070, -0.460 = CIE 2000: 0.43
Though this isn't exactly a perfect test either as there's a margin of error between X-Rite's target manufacturing, and their reading, and my readings even if everything were working. I would really need an individually measured target and reference I would think. But the results are closer than my readings from the paper.
The i1studio doesn't use a uV cut filter, rather it uses a white led that has virtually no uV in it. It's the same technology that the ColorMunki and iSiS uses for generating M2 profiles.
IMO, M2 profiles are better for most uses than M0 or M1 profiles these days for indoor use because most lighting no longer uses incandescent lamps and have lower or no uV. M1 is really best for displaying prints outdoors and M0 for natural lighting though a window.
Because of this shift in lighting tech., indoor lighting will, in most cases, be a better match for M2 profiles regardless of whether a paper has OBAs or not. And for papers with no OBAs there is no difference between M0, M1, or M2 profiles. They should produce identical results.
I think your spectro is working correctly. The lower b* reading on the CC white patch is not uncommon. Most CCs have a b* between 2 and 3 but I've seen numbers around 1 as well. The other neutral patches are quite reasonable and are also more consistent between different CCs. So that's a good sign things are working spectro wise.
The problem may be in your profile or printer driver settings. You should expect lower dE numbers from the CC image print using Abs. Col. How did you make or acquire the profile you are using?
-
If you don't know what or why you'd use Absolute Colorimetric rendering intent, don't use it ;D .
I general yes, avoid it. But there are some appropriate places it is needed.
This explains where Abs. can be of use when reproducing scanned images:
https://books.google.com/books?id=OxlBqY67rl0C&pg=PA290&lpg=PA290&dq=scanners+%22absolute+colorimetric%22+relative&source=bl&ots=IyXihwQSog&sig=mcJZcCFx9P1uCCjSfyvK_Sv2qC8&hl=en&sa=X&ved=2ahUKEwi68N65yO_aAhUM4IMKHZpDD_U4FBDoATAAegQIABAn#v=onepage&q=scanners%20%22absolute%20colorimetric%22%20relative&f=false
-
Forgot to do the obvious test, and just measure my reading of the paper vs the profile's whitepoint:
My Argyll spotread on Canon Glossy II paper: 94.055, -0.611, -2.047
ICC profile's white point: 94.19024, -0.5772792, -2.4742979
100% Lab in Photoshop with relative conversion to paper profile: 94.19, -0.562, -2.491
So, not sure what to make of things now, looks like my readings aren't too bad?
Perfectly good numbers. Very typical of the small variations seen on glossy paper. Also, the slightly negative b* readings are also typical of most papers. It's from a colorant manufacturers add to make the paper look less warm. BTW, printing L*=100 with Colorimetric intent (or anything other than Abs.) should result in nothing printing at all so those numbers are all good.
-
The problem may be in your profile or printer driver settings. You should expect lower dE numbers from the CC image print using Abs. Col. How did you make or acquire the profile you are using?
My methodology has been:
- Open the attached BabelColor Lab ColorChecker TIFF in Photoshop CC 2018.
- Print with Photoshop managed colors in Absolute, choosing the Canon-supplied Photo Paper Plus Glossy II profile.
- Under the Canon Pro-10 printer settings in Windows I made sure to go to Manual Color Adjustment > Matching and set it to None. I then choose the correct paper type, and set the quality to high, leaving all other settings alone. (On the Mac when I print I can just choose Photoshop managed color and it'll disable the printer driver settings automatically, but the Mac version of Photoshop isn't working with those particular profiles anyway.)
- Print and wait until the ink is dry [enough].
- Measure using Argyll's spotread tool and the i1Studio, doing multiple readings to eliminate outliers.
I'm currently doing some similar tests with some different papers that I have profiles for and that work on the Mac too and I'll see what I get for results.
Perfectly good numbers. Very typical of the small variations seen on glossy paper. Also, the slightly negative b* readings are also typical of most papers. It's from a colorant manufacturers add to make the paper look less warm. BTW, printing L*=100 with Colorimetric intent (or anything other than Abs.) should result in nothing printing at all so those numbers are all good.
That particular non-absolute conversion test was just a Photoshop sanity check, but I didn't actually try printing it. Maybe I should have. (Glad to know I'm not crazy though, just a bit ignorant when it comes to Photoshop and color science. This project has been a crash course in both, does help a bit having a computer science background.)
-
Canon Pro Platinum Glossy
Used Canon supplied ICC profile for Canon Pro-10. All measurements were done against a background of black construction paper. (Didn't notice a notable difference against a white background though.)
ICC profile white point: 96.2402999, -0.6663628, 0.9857583
Measured paper white: 95.328430 -0.796546 1.600733 CIE2000: 0.81 CIE1979: 1.11
ColorChecker Lab values after converting to paper profile using absolute colorimetric in Photoshop:
White: 95.11 -1.022 2.930
Grey: 51.23 0.172 -0.155
Black: 20.43 -0.491 -1.983
Printed using absolute 16-bit from Photoshop CC 2018 on Mac:
White: 93.909180 -1.209533 3.318551 CIE2000: 0.83 CIE1976: 1.28
Grey: 46.562440 -0.001105 -0.874291 CIE2000: 4.71 CIE1976: 4.73
Black: 15.956659 0.485438 -1.726592 CIE2000: 3.37 CIE1976: 4.59
Printed using absolute 8-bit from Photoshop CC 2018 on Windows 10:
White: 93.836831 -1.229150 3.539094 CIE2000: 0.97 CIE1976: 1.43
Grey: 47.579424 0.216308 -0.602152 CIE2000: 3.76 CIE1976: 3.68
Black: 15.992631 0.530592 -1.728224 CIE2000: 3.38 CIE1976: 4.56
Printed using absolute 8-bit from Photoshop CC 2018 on Windows 10 and measured a week (or 2) later:
White: 93.785744 -1.421952 3.711370 CIE2000: 1.16 CIE1976: 1.59
Grey: 48.070289 0.230349 -0.587270 CIE2000: 3.24 CIE1976: 3.21
Black: 16.604610 0.528541 -1.519456 CIE2000: 3.04 CIE1976: 3.9
-
Finally, Red River Pecos River Gloss
Used ICC profile from Red River for Canon Pro-10, following their instructions for Canon paper/quality driver settings. All measurements were done against a background of black construction paper. (Didn't notice a notable difference against a white background though.)
ICC profile white point: 96.3303456, 0.3228647, -0.6502543
Measured paper white: 95.170087 0.094665 -1.445915 CIE2000: 1.09 CIE1976: 1.43
ColorChecker Lab values after converting to paper profile using absolute colorimetric in Photoshop:
White: 94.69 -0.937 2.727
Grey: 50.76 -0.093 0.164
Black: 20.69 0.156 -0.499
Printed using absolute 16-bit from Photoshop CC 2018 on Mac (measured later that day):
White: 92.771324 -1.153834 1.552349 CIE2000: 1.62 CIE1976: 2.26
Grey: 48.815054 -0.191202 -1.126463 CIE2000: 2.32 CIE1976: 2.34
Black: 18.736083 0.498224 -0.592850 CIE2000: 1.44 CIE1976: 1.99
Printed using absolute 8-bit from Photoshop CS6 on Mac (measured later that day):
White: 92.857677 -0.944758 1.192230 CIE2000: 1.8 CIE1976: 2.39
Grey: 47.703535 -0.106346 -1.433525 CIE2000: 3.42 CIE1976: 3.45
Black: 18.039598 0.571037 -0.802760 CIE2000: 1.94 CIE1976: 2.7
Printed using absolute 8-bit from Photoshop CS6 on Mac (measured next day):
White: 92.696839 -1.203567 1.527344 CIE2000: 1.69 CIE1976: 2.34
Grey: 48.613465 -0.123885 -1.223072 CIE2000: 2.54 CIE1976: 2.56
Black: 18.690034 0.484400 -0.521634 CIE2000: 1.46 CIE1976: 2.03
Forgot to print from Windows, probably wouldn't deviate too much… wanted to compare different versions of Photoshop.
-
I general yes, avoid it. But there are some appropriate places it is needed.
This explains where Abs. can be of use when reproducing scanned images:
https://books.google.com/books?id=OxlBqY67rl0C&pg=PA290&lpg=PA290&dq=scanners+%22absolute+colorimetric%22+relative&source=bl&ots=IyXihwQSog&sig=mcJZcCFx9P1uCCjSfyvK_Sv2qC8&hl=en&sa=X&ved=2ahUKEwi68N65yO_aAhUM4IMKHZpDD_U4FBDoATAAegQIABAn#v=onepage&q=scanners%20%22absolute%20colorimetric%22%20relative&f=false
Thank you for the link to that section of that book. I found a couple other good books on the subject, but everyone words things a bit differently and I thought that one you shared gave a good example of how the two work.
I was initially skeptical of using Absolute at all because most intro discussions to the topic of rendering intents outright say "don't use it" or they ignore it altogether. Lightroom doesn't even give an option to use it, which is probably reasonable when printing digital photos that aren't trying to reproduce existing artwork. (Clearly in most cases you'd probably want the white of the paper to match the white of what's being printed.) And if I print using any other apps, the printer drivers don't give any options at all for intents, so I'm not sure what method the driver uses internally if you set it to use a profile. (But that doesn't matter right now.) Photoshop just says "suitable for proofing to simulate the output of a particular device" which to me wasn't clear that it would be useful for preserving the actual colors of my scanned images.
So, I feel I'm understanding the subject. But when my input and my output results don't match what my understanding is, I start to second guess all the basics. Which is why I started this thread. :)
-
Finally, Red River Pecos River Gloss
Used ICC profile from Red River for Canon Pro-10, following their instructions for Canon paper/quality driver settings. All measurements were done against a background of black construction paper. (Didn't notice a notable difference against a white background though.)
ICC profile white point: 96.3303456, 0.3228647, -0.6502543
Measured paper white: 95.170087 0.094665 -1.445915 CIE2000: 1.09 CIE1976: 1.43
ColorChecker Lab values after converting to paper profile using absolute colorimetric in Photoshop:
White: 94.69 -0.937 2.727
Grey: 50.76 -0.093 0.164
Black: 20.69 0.156 -0.499
Printed using absolute 16-bit from Photoshop CC 2018 on Mac (measured later that day):
White: 92.771324 -1.153834 1.552349 CIE2000: 1.62 CIE1976: 2.26
Grey: 48.815054 -0.191202 -1.126463 CIE2000: 2.32 CIE1976: 2.34
Black: 18.736083 0.498224 -0.592850 CIE2000: 1.44 CIE1976: 1.99
Printed using absolute 8-bit from Photoshop CS6 on Mac (measured later that day):
White: 92.857677 -0.944758 1.192230 CIE2000: 1.8 CIE1976: 2.39
Grey: 47.703535 -0.106346 -1.433525 CIE2000: 3.42 CIE1976: 3.45
Black: 18.039598 0.571037 -0.802760 CIE2000: 1.94 CIE1976: 2.7
Printed using absolute 8-bit from Photoshop CS6 on Mac (measured next day):
White: 92.696839 -1.203567 1.527344 CIE2000: 1.69 CIE1976: 2.34
Grey: 48.613465 -0.123885 -1.223072 CIE2000: 2.54 CIE1976: 2.56
Black: 18.690034 0.484400 -0.521634 CIE2000: 1.46 CIE1976: 2.03
Forgot to print from Windows, probably wouldn't deviate too much… wanted to compare different versions of Photoshop.
These numbers just indicate that the profiles you are using don't quite match your printer's response. Consumer printers tend to vary a bit, perhaps as much as 3 dE or more from each other. The higher end printers are individually tuned to match a reference response but this is less practical for the high volume consumer or semi-pro printers. And printers age and change over time. Normally, people don't notice these differences printing regular photos but it becomes more apparent when people are trying to match colors. And that, of course, is why Abs. Col. exists. This is one of the reasons people buy printer profiling hardware.
You should be profiling the papers you use, not using profiles from the paper vendors or printer OEMs. You will then get much better results.
BTW, when you create a profile you should be able to use the same one for Windows as well as iOS and it shouldn't make any difference what computer you make them on. Use your XRite software, it's quite good.
-
Finally, Red River Pecos River Gloss
Used ICC profile from Red River for Canon Pro-10, following their instructions for Canon paper/quality driver settings. All measurements were done against a background of black construction paper. (Didn't notice a notable difference against a white background though.)
ICC profile white point: 96.3303456, 0.3228647, -0.6502543
Measured paper white: 95.170087 0.094665 -1.445915 CIE2000: 1.09 CIE1976: 1.43
ColorChecker Lab values after converting to paper profile using absolute colorimetric in Photoshop:
White: 94.69 -0.937 2.727
Grey: 50.76 -0.093 0.164
Black: 20.69 0.156 -0.499
Exactly what process are you doing by converting Abs. Col. in Photoshop and what are you expecting from that process? If you are going through the transform of a working RGB space to a printer RGB space then back to a working space the transform will produce small changes based on rounding and interpolation errors in the ICC profiles for all colors that are in gamut (printable)*. This tells you nothing about how accurate the profile is, only that it is well formed. The numbers will typically change very little.
The only way to test a profile's accuracy is print known colors and measure them. If you print using Abs. Col. and a custom profile then measure them you should have average dEs under 1.0.
Also, ICC profiles are based on measurements using a white backing. This differs from some parts of graphic arts where a black backing is used. for best match when measuring print colors, use a white backing.
* If you, in Photoshop, take an image in a working space and convert it to the printer space via "Convert Profile" you can select whatever ICC intent you wish including BPC. Once you have that and wish to know what colors the printer profile expects will be printed, you must convert the image back to Lab or a standard RGB working space that is at least as large as the printer's gamut. I recommend either Lab or ProPhoto RGB. When converting back, select Abs. Col. This resulting image will have the color values that it expects to be printed. How close that is depends on how good your profile is and generally custom profiles produce the best match.
BTW, this a a good way to see what gets printed when you start with colors outside a printer's gamut. It will tell you quite closely what the actual, printable color is expected to be.
-
Hey Doug, thanks for taking the time and having the patience to provide all the details you have. I really appreciate it. I did a little experiment based on what you shared with me, just to make sure I'm understanding completely:
Given I print an image with known color values (in this case the "synthetic" ColorChecker from BabelColor) using a particular rendering intent…
If I want to reliably evaluate the accuracy of the print, I can:
- Take the image I printed and convert it to the same printer paper profile using the same rendering intent settings.
- Convert the image to Lab using absolute colorimetric.
- Measure the print with my spectrophotometer and compare the Lab values to the Lab values that I see in Photoshop's info panel. (In the attached image those would be the values highlighted in green or blue.)
Assuming I'm on the same page as you :) then at least I know that my comparisons above were done correctly. I have my Photoshop color settings currently set to absolute colorimetric and the image I've been working with is natively in the Lab color space and I've only been testing absolute printing, so hopefully I haven't muddied anything up.
-
These numbers just indicate that the profiles you are using don't quite match your printer's response.
…
You should be profiling the papers you use, not using profiles from the paper vendors or printer OEMs. You will then get much better results.
I had expected a certain degree of error with OEM profiles, given manufacturing tolerances, etc., but never quite expected it to be that off. Using the OEM profiles probably wouldn't really be a concern for regular photo printing for an amateur like me, the Pro-10 is already much better than most photo printers I've used over the years. But I did as you suggested yesterday and waited until today to measure my results…
The OEM Canon Glossy II profile gave me an average ∆E (1976) of 3.681.
My custom i1Studio profile brought that down to an average ∆E of 1.312! It's possible I could get that closer with more patience. – For example I only waited about 30 minutes for dry time for my profiling, perhaps I could've let that settle down a bit more. It'd be nice if i1Studio gave me some data to look at. If I use Argyll I can double-check my readings before making the profile… I may try that down the road.
An interesting observation: The printed ColorChecker using the OEM Canon profile had much more saturated colors, almost like it was setup to give the output more punch. The colors are toned down using the custom profile.
-
Assuming I'm on the same page as you :) then at least I know that my comparisons above were done correctly. I have my Photoshop color settings currently set to absolute colorimetric and the image I've been working with is natively in the Lab color space and I've only been testing absolute printing, so hopefully I haven't muddied anything up.
Your Photoshop methodology is dead on. Those numbers are exactly what I would expect. Even the difference when using sRGB tiffs compared to Lab. With sRGB you are hitting up against the sRGB gamut edge and that accounts for the different L* drops for RI. as well as the saturation decrease. Nice work.
if you use ProPhoto RGB space you won't see the differences you get with sRGB.
-
I had expected a certain degree of error with OEM profiles, given manufacturing tolerances, etc., but never quite expected it to be that off. Using the OEM profiles probably wouldn't really be a concern for regular photo printing for an amateur like me, the Pro-10 is already much better than most photo printers I've used over the years. But I did as you suggested yesterday and waited until today to measure my results…
The OEM Canon Glossy II profile gave me an average ∆E (1976) of 3.681.
My custom i1Studio profile brought that down to an average ∆E of 1.312! It's possible I could get that closer with more patience. – For example I only waited about 30 minutes for dry time for my profiling, perhaps I could've let that settle down a bit more. It'd be nice if i1Studio gave me some data to look at. If I use Argyll I can double-check my readings before making the profile… I may try that down the road.
An interesting observation: The printed ColorChecker using the OEM Canon profile had much more saturated colors, almost like it was setup to give the output more punch. The colors are toned down using the custom profile.
The high dE numbers for the canned profiles is pretty typical. So is the higher saturation, My Canon also produces higher saturation using canned profiles. My suspicion is that Canon is trying to create more "Pop." It's even worse on my Canon if you let the printer manage color.
As for an average dE76 of 1.3 for your i1Studio profiled print, that's not bad really. Also, you will likely find that if you use dE2000 the average is quite a bit lower. Likely around .8. i1Studio has, I believe, an option for a second pass where they print colors based on the first set that improve the profile. This should improve things further.
Also, don't worry too much about dry time. Most of the changes occur in the first 10 minutes of drying. After 30 minutes dE errors from print drying become a fraction of the total and likely accounts for no more than 1/3 of the dEs. After 2 hours around 1/5th of the dEs compared to a two day dry time. Of course it's best to let prints dry 24 hrs to make the best profiles but don't expect significant differences.
I posted an extensive series of measurements on drying time color shifts some months back.
BTW, if you want to get a sense of how gamut mapping works at and beyond the printer's gamut boundary, try this:
In Photoshop create a gradient from Lab (50,-120,0) to (50,0,0). Then convert it using Abs. Col. to Printer space and back again to Lab. Now run the cursor down the gradient and watch the small variations of b* around 0 until you get to the place where it can no longer print the requested colors. What it actually prints will be reasonably accurately reported and you will see shifts in the Lab values as it maps the OOG colors to the nearest point on the 3D gamut space. It can be interesting to do the same thing but converting to Perceptual, then back to Lab using Abs. You will see the different ways Perc. maps colors.
-
My custom i1Studio profile brought that down to an average ∆E of 1.312! It's possible I could get that closer with more patience. – For example I only waited about 30 minutes for dry time for my profiling, perhaps I could've let that settle down a bit more. It'd be nice if i1Studio gave me some data to look at. If I use Argyll I can double-check my readings before making the profile… I may try that down the road.
Just as a reference point, I use Argyll to do all my printer profiling (Epson 3880) and my deltaE is usually right around 1.0. Profiles are quite smooth when visualized.
-
If you don't know what or why you'd use Absolute Colorimetric rendering intent, don't use it ;D .
Absolute Colorimetric rendering intent is my preferred setting for printing scenes such as the one attached. By the way, in deference to Mark Segal, I am retracting statements I made last October in re letting the printer (Epson P800) manage color. ... Photoshop is superior 99% of the time. Oddly, the first few images that I'd printed on the P800 had large patches of saturated blues and oranges (still barely within aRGB gamut). My unchecked enthusiasm resulted in a firestorm.
Seriously though, absolute colorimetric rendering intent, time and time again, is the better choice for the type of nighttime images I like to print. I typically make two master files: one for print, and the other for screen sRGB (with some additional tweaking). I use an Eizo for soft proofing and for inspecting sRGB conversions. Lastly, I check to see how the sRGB files appear on a current, but basic, Microsoft Surface Pro. The comparison is not dead-on, but it's reasonable.
The print of the shack photo is virtually identical to how it looks on both a Surface 3 and on a Surface Pro.
-
Absolute Colorimetric rendering intent is my preferred setting for printing scenes such as the one attached.
The difference between it and RelCol only affects white. So what do you see differently doing that? The absolute colorimetric rendering intent reproduces the exact color that existed in the source—absolutely. If the source was light color on the dingy yellow-white of newsprint, the resulting color on your brilliant coated ink jet paper will be dingy yellow. This intent is really designed for making one device simulate the appearance of another device for use in proofing
-
It allows for better detail in the shadows while still keeping the blacks nice and dark. Of course, I make a curve adjustment for the shadows and another one for the highlights. And depending on the media, I may add an additional curve for HSL.
I soft proof on a current entry-level Eizo Coloredge monitor. The white point is set at 4700K, which matches my Solux bulbs. I've found that 4700K is a good general purpose temperature. My monitor is placed in a controlled environment where the ambient light is dim and constant. The luminance of the monitor is 88 CDM.
Whether or not it's relevant, my workstation runs on Windows 7. I've not used an Apple since the late 1990s. My decision to migrate over to PC had more to do with practical business needs. ...not saying one platform is better than the other. I think both have strengths and weaknesses...
-
It's certainly possible that I could achieve comparable output using relative colorimetric intent. But from my experience using absolute colorimetric intent has greatly reduced the amount of time required in post. It's also streamlines the process of going from aRGB to sRGB for screen display.
I hope my description is clear.
-
It's certainly possible that I could achieve comparable output using relative colorimetric intent. But from my experience using absolute colorimetric intent has greatly reduced the amount of time required in post. It's also streamlines the process of going from aRGB to sRGB for screen display.
I hope my description is clear.
I see absolutely no difference converting from Adobe RGB (1998) to sRGB using Absolute or RelCol nor would I expect to see any difference. Nor do the numbers change between the two options. Can you explain what you mean?
-
It allows for better detail in the shadows while still keeping the blacks nice and dark.
Because there's no option for Black Point Compensation with Absolute. With RelCol you have the option to use it or not.
-
I see absolutely no difference converting from Adobe RGB (1998) to sRGB using Absolute or RelCol nor would I expect to see any difference. Nor do the numbers change between the two options. Can you explain what you mean?
I suppose I could take screenshots to demonstrate. As mentioned earlier, it's more about being a shortcut workaround in post than anything else.
-
I suppose I could take screenshots to demonstrate. As mentioned earlier, it's more about being a shortcut workaround in post than anything else.
Yes, you should!
Let me also help you with this example:
- Open image in Adobe RGB (1998)
- Duplicate it, name it Absolute
- Convert one from Adobe RGB (1998) to sRGB using RelCol
- Convert the other from Adobe RGB (1998) to sRGB using Absolute Colorimetric
- Use Apply Image Command, subtract one from the other. Results below. THEY ARE IDENTICAL!
If you get something different, please supply the source document in Adobe RGB (1998). If you get the same results as I do, you fall into the camp of users who shouldn’t be using Absolute Colorimetric as I posted here!
-
Yes, you should!
Let me also help you with this example:
- Open image in Adobe RGB (1998)
- Duplicate it, name it Absolute
- Convert one from Adobe RGB (1998) to sRGB using RelCol
- Convert the other from Adobe RGB (1998) to sRGB using Absolute Colorimetric
- Use Apply Image Command, subtract one from the other. Results below. THEY ARE IDENTICAL!
If you get something different, please supply the source document in Adobe RGB (1998). If you get the same results as I do, you fall into the camp of users who shouldn’t be using Absolute Colorimetric as I posted here!
Well, of course you are correct re converting Adobe RGB <> sRGB. There is a possibility David was thinking of something he did in the past like converting ProPhoto to either Adobe RGB or sRGB using the Microsoft ICM instead of Adobe's ACE. Big difference between those two.
But generally, stay away from Abs. Col. without good reason to use it. For instance if you print a gradient going from RGB 220,220,220 to 255,255,255 you will get clipping and likely a color shift as well using Abs. Col. that you will not get with any of the other intents.
-
Well, of course you are correct re converting Adobe RGB <> sRGB. There is a possibility David was thinking of something he did in the past like converting ProPhoto to either Adobe RGB or sRGB using the Microsoft ICM instead of Adobe's ACE. Big difference between those two.
I have no idea what David is thinking, only what he wrote. I await his screen captures.
As for MS CMM vs. ACE, if there is a big difference, it’s a bug.
-
I have no idea what David is thinking, only what he wrote. I await his screen captures.
As for MS CMM vs. ACE, if there is a big difference, it’s a bug.
It's a difference of interpretation of how media white points of D65 profiles (sRGB, Adobe RGB) are supposed to be mapped to/from D50 ones like ProPhoto. The ICC clarified it more than 15 years ago. Adobe follows the clarification. Microsoft doesn't. It is also a divergence between how CIE and the ICC defines "Absolute Colorimetric." There is a tech. note on it over at color.org.
-
It's a difference of interpretation of how media white points of D65 profiles (sRGB, Adobe RGB) are supposed to be mapped to/from D50 ones like ProPhoto.
Yes. But again, in Photoshop, I see zero differences in how Absolute vs. RelCol convert from ProPhoto RGB let alone Adobe RGB (1998) with ACE. When I toggle to Apple CMM, the differences noted are absolutely awful appearing so I can't fathom why anyone would do so. Can you? And when the WP's are the same, as we see with Adobe RGB (1998)/sRGB, there's no difference with Apple's CMM. So until we hear back from David, I'm again at a loss to what or why he's doing what he reports he's doing. Maybe aRGB as he calls it, isn't Adobe RGB (1998); I wish people would use the proper names for these color spaces but that's another issue.
-
Yes, you should!
Let me also help you with this example:
- Open image in Adobe RGB (1998)
- Duplicate it, name it Absolute
- Convert one from Adobe RGB (1998) to sRGB using RelCol
- Convert the other from Adobe RGB (1998) to sRGB using Absolute Colorimetric
- Use Apply Image Command, subtract one from the other. Results below. THEY ARE IDENTICAL!
If you get something different, please supply the source document in Adobe RGB (1998). If you get the same results as I do, you fall into the camp of users who shouldn’t be using Absolute Colorimetric as I posted here!
Here are three screen shots: Absolute, Relative with Black Point Compensation, and Relative (soft proofs for Epson Ultra Premium Luster Photo Paper). Photos were taken with an Olympus Pen F in 8-shot mode for color fidelity and to control moire. The three RAW files were minimally processed as one batch with all adjustment in sync.
-
Here are three screen shots: Absolute, Relative with Black Point Compensation, and Relative (soft proofs for Epson Ultra Premium Luster Photo Paper).
A. I see no difference.
B. None are conversions you stated below (Adobe RGB (1998) or what you call aRGB to sRGB.
So exactly what are you referring to?
-
I see significant differences in the shadow detail of the trees in the sky and the luminance of the light projecting onto the street. The differences are clearly there, but being that you're looking at photos of a screen, the differences will not hit you in the face as much as what is visible to the naked eye when looking directly at the monitor.
-
A. I see no difference.
B. None are conversions you stated below (Adobe RGB (1998) or what you call aRGB to sRGB.
So exactly what are you referring to?
The pictures on the monitor are aRGB. I photographed the screens in aRGB, and then downsized and saved as jpgs (8-bit, sRGB).
-
I see significant differences in the shadow detail of the trees in the sky and the luminance of the light projecting onto the street. The differences are clearly there, but being that you're looking at photos of a screen, the differences will not hit you in the face as much as what is visible to the naked eye when looking directly at the monitor.
I can't see significant differences in your screen captures which BTW are untagged here.
Next and again, NEITHER is a conversion from the RGB working space you specifically referred to! Here's your exact text:
It's also streamlines the process of going from aRGB to sRGB for screen display.
Next and again, the differences between RelCol and Absolute, as you were told, is one allows Black Point Compensation while the other does not. Yet you have it ON in one screen capture but not the other. They have nothing to do with the quote above you made about aRGB (presumably Adobe RGB (1998)) and sRGB.
I'll ask again, what exactly are you referring to?
-
Should be tagged. I exported as sRGB. I will post them again if you want.
-
Should be tagged. I exported as sRGB. I will post them again if you want.
I assigned sRGB. But that doesn’t explain what you wrote.
Is there going to be a difference going from any RGB working space to a print space using Absolute vs. RelCol? Sure. No one is disputing that. Is there a difference converting Adobe RGB (1998) to sRGB using either? No, there is NO difference. Begging the question again, what does this mean:
It's also streamlines the process of going from aRGB to sRGB for screen display.
Was this question unclear to you?
I see absolutely no difference converting from Adobe RGB (1998) to sRGB using Absolute or RelCol nor would I expect to see any difference. Nor do the numbers change between the two options. Can you explain what you mean?
-
I can't see significant differences in your screen captures which BTW are untagged here.
Next and again, NEITHER is a conversion from the RGB working space you specifically referred to! Here's your exact text: Next and again, the differences between RelCol and Absolute, as you were told, is one allows Black Point Compensation while the other does not. Yet you have it ON in one screen capture but not the other. They have nothing to do with the quote above you made about aRGB (presumably Adobe RGB (1998)) and sRGB.
I'll ask again, what exactly are you referring to?
The file is an aRGB file. The monitor is set in soft proof to show Absolute, Relative, and Relative with black point compensation. Forget about sRGB--I only brought that up because Chrome as of yet is not color managed while Firefox and Safari are. I use chrome.
-
Forget about sRGB--I only brought that up because Chrome as of yet is not color managed while Firefox and Safari are. I use chrome.
Which has absolutely nothing to do with what you wrote and I had to keep asking you to explain! Again, you really seem the correct candidate I discussed below in terms of not using Absolute Colorimetric intent....
Chrome is color managed!
https://www.color-management-guide.com/web-browser-color-management.html
The reason there's so much ignorance on the subject of color management, is that those who have it are so eager to regularly share it! - The Digital Dog
-
Yes. But again, in Photoshop, I see zero differences in how Absolute vs. RelCol convert from ProPhoto RGB let alone Adobe RGB (1998) with ACE. When I toggle to Apple CMM, the differences noted are absolutely awful appearing so I can't fathom why anyone would do so. Can you? A
Sure. I can explain what's going on with Microsoft and Apples (now I know Apple also isn't abiding by the ICC's clarification) engines.
When you have an image in an RGB D50 media space, like ProPhoto, the convert to Adobe RGB or sRGB the Absolute Intent in Apple and Microsoft's engines attempts to create the same "colors" that are would occur if the displays were switched from D50 to D65. To do that they strongly yellow tint the image. It's highly problematic since most people set their monitors independently of the working space's "media whitepoint" and usually at the same CCT regardless of working space. The ICC recognizes this. Apple and Microsoft apparently do not.
The only possible use of something like that is if you have an image you want to paste into a printable image, possibly inserted into a "monitor" screen and want it to reflect the difference between the D50 of a print and the D65 of a typical monitor. In that case you would convert the image from sRGB to ProPhoto using Abs. Col. with either the Microsoft or Apple engine then paste it into the "monitor's" face.
That extreme corner case aside, it's still problematic. Any full white (RGB 255,255,255) will be luminance clipped because either the reds or blues would need to go above 255 it get the same luminance of "white" post conversion. You can see this by converting from sRGB to ProPhoto and back again to sRGB using Abs. Col and one of these screwy engines. The white is no longer 255,255,255. In the case of ProPhoto->sRGB->ProPhoto the white is changed to 245,253,253. It's not even a neutral color because of the conversion clipping at 255. Pretty obvious why the ICC indicated color engines should not do this.
-
Sure. I can explain what's going on with Microsoft and Apples (now I know Apple also isn't abiding by the ICC's clarification) engines.
When you have an image in an RGB D50 media space, like ProPhoto, the convert to Adobe RGB or sRGB the Absolute Intent in Apple and Microsoft's engines attempts to create the same "colors" that are would occur if the displays were switched from D50 to D65. To do that they strongly yellow tint the image. It's highly problematic since most people set their monitors independently of the working space's "media whitepoint" and usually at the same CCT regardless of working space. The ICC recognizes this. Apple and Microsoft apparently do not.
The only possible use of something like that is if you have an image you want to paste into a printable image, possibly inserted into a "monitor" screen and want it to reflect the difference between the D50 of a print and the D65 of a typical monitor. In that case you would convert the image from sRGB to ProPhoto using Abs. Col. with either the Microsoft or Apple engine then paste it into the "monitor's" face.
That extreme corner case aside, it's still problematic. Any full white (RGB 255,255,255) will be luminance clipped because either the reds or blues would need to go above 255 it get the same luminance of "white" post conversion. You can see this by converting from sRGB to ProPhoto and back again to sRGB using Abs. Col and one of these screwy engines. The white is no longer 255,255,255. In the case of ProPhoto->sRGB->ProPhoto the white is changed to 245,253,253. It's not even a neutral color because of the conversion clipping at 255. Pretty obvious why the ICC indicated color engines should not do this.
Agreed on all points above. And the results are also ugly. But it appears, this has nothing to do with what BobDavid was writing about. His text about conversions from RGB working spaces was just confused. And incorrect :(
-
Agreed on all points above. And the results are also ugly. But it appears, this has nothing to do with what David was writing about. His text about conversions from RGB working spaces was just confused. And incorrect :(
Yes. Conversions of sRGB <> Adobe RGB do not change whites in any of the color engines. But it's possible David had run across the Apple or Microsoft weirdness converting between spaces that had differing media white points and has confused this with Adobe RGB / sRGB. It might also show up if for some reason one of the conversion spaces was a monitor's profile space instead of a standard RGB working space and he was using the Apple/Microsoft engine. Sometimes it's hard to remember exactly what one did 5 years ago and it could just be something fuzzy in his memory.
-
Re absolute: So you do not see more shadow detail in the sky area, meaning the trees. The color of the auto is also slightly punchier. Perhaps we're missing each other's points.
Anyway, for whatever reason, printing with the rendering intent set to Absolute is working fine. Soft proofing in Absolute and then toggling soft proof to aRGB matches closely, much better than with Relative with or without black point compensation.
In the end, it's what a print shows or hides that matters. Outside of studio work, I typically photograph scenes I'm familiar with. I've been observing that shack for six months. It often takes me months to figure out how to photograph a scene. The print looks as I'd imagined it would.
It took about an hour to process the file for print. I made an 8.5" X 11" test and then a final 24" X 36" exhibition print.
-
Here are three screen shots: Absolute, Relative with Black Point Compensation, and Relative (soft proofs for Epson Ultra Premium Luster Photo Paper). Photos were taken with an Olympus Pen F in 8-shot mode for color fidelity and to control moire. The three RAW files were minimally processed as one batch with all adjustment in sync.
I see the differences between Rel. Col. and Abs. and they are exactly what I would expect. The Rel. Col. soft proof is slightly bluer and less luminous. This is to be expected as Abs. Col. prints will be brighter because Rel Col drops the print density and shifts tint so that L*=100 matches the paper white. Another side effect is that when the black point is reached, Abs Col typically undergoes larger a* and b* changes than Rel. Col. this can create the impression of more texture in deep shadows, or it can produce less desirable effects.
The basic problem with Abs. is that white gradients will clip when you go from smooth RGB 220 to 255. Try it. You don't have those in this image but this is a big reason Abs. Col. is discouraged for the normal photo printer.
-
Yes. Conversions of sRGB <> Adobe RGB do not change whites in any of the color engines. But it's possible David had run across the Apple or Microsoft weirdness converting between spaces that had differing media white points and has confused this with Adobe RGB / sRGB. It might also show up if for some reason one of the conversion spaces was a monitor's profile space instead of a standard RGB working space and he was using the Apple/Microsoft engine. Sometimes it's hard to remember exactly what one did 5 years ago and it could just be something fuzzy in his memory.
I can’t read his mind nor can I read his memories. Bottom line is the text that he wrote about the two RGB working spaces and an absolute colorimetric conversion is either wrong or confused or both. ::)
-
There is discord re whether Chrome is color managed or not.
Chrome Okay (https://www.color-management-guide.com/web-browser-color-management.html)
Chrome not Okay (http://cameratico.com/guides/web-browser-color-management-guide/)
-
There is discord re whether Chrome is color managed or not.
Chrome Okay (https://www.color-management-guide.com/web-browser-color-management.html)
Chrome not Okay (http://cameratico.com/guides/web-browser-color-management-guide/)
Either way IF you use a non color managed product, you get what you deserve.The larger point, is that what you wrote about the two working spaces and an absolute called metric conversion is incorrect and has nothing to do with a color managed or non color managed web browser! Nice try at a digression, not going to fly here. A professional photographer and technician knows better (http://forum.luminous-landscape.com/index.php?topic=124581.msg1042064#msg1042064) ;D
-
I can’t read his mind nor can I read his memories. Bottom line is the text that he wrote about the two RGB working spaces and an absolute colorimetric conversion is either wrong or confused or both. ::)
No, it's simply wrong. He may also be confused or just mistaken.
Edit: It appears he has edited his posts. He states that he may have been unclear, and was not claiming a difference in converting Adobe RGB <> sRGB using Abs. v. Rel Col.
In any case he seems to be focusing on the soft proofing and Abs. v Rel. I've pointed out the risks and issues of using Abs. Col. but if he tweaks it soft proofing as he indicated he did then he may well like the results and that's fine. There are cases where Abs. Col. can create a nice effect in deep shades as a result of the a* and b* shift that occurs at the black point. I've seen it maybe once or twice but it's rare. The negatives of having high key clipping are the major reason I don't use Abs. for photos. I only use Abs. for repro work.
-
I can’t read his mind nor can I read his memories. Bottom line is the text that he wrote about the two RGB working spaces and an absolute colorimetric conversion is either wrong or confused or both. ::)
...never wrote anything about converting aRGB to Absolute or sRGB or anything else. I may have stumbled while trying to explain that soft proofing an image and then photographing that soft proof from the screen and then posting that photo of the photo on on the web to be viewed on another screen is problematic.
What I did write is that when soft proofing paper, Absolute intent rather than Relative matches how the original file looks on the screen as opposed to a print.
-
This is a good facsimile of what the print looks like. It's the same picture I posted hours ago. So back to the beginning, if for whatever reason one is able to get a better print via Absolute intent rather than Relative intent, so be it.
-
...never wrote anything about converting aRGB to Absolute or sRGB or anything else. I may have stumbled while trying to explain that soft proofing an image and then photographing that soft proof from the screen and then posting that photo of the photo on on the web to be viewed on another screen is problematic.
You did very much so stumble:
It's also streamlines the process of going from aRGB to sRGB for screen display.
-
This is a good facsimile of what the print looks like. It's the same picture I posted hours ago. So back to the beginning, if for whatever reason one is able to get a better print via Absolute intent rather than Relative intent, so be it.
As long as you soft proof it pick whatever intent produces the most appealing results. It's your eyes. Just be particularly aware of the high key clipping issue in Abs. Col. due to the luminance increase between Rel. Col and Abs. Col.
-
As long as you soft proof it pick whatever intent produces the most appealing results. It's your eyes. Just be particularly aware of the high key clipping issue in Abs. Col. due to the luminance increase between Rel. Col and Abs. Col.
Exactly! Bravo, well spoken! ... Yes it's often necessary to add a curve layer to maintain detail in the highlights.
-
Yes it's often necessary to add a curve layer to maintain detail in the highlights.
How do you accomplish that when, as Doug suggests, the highlights are clipped:
Just be particularly aware of the high key clipping issue in Abs. Col. due to the luminance increase between Rel. Col and Abs. Col.
Curves layers can't bring back clipped detail or data, only make them less than RGB 255 (Lstar 100).
-
You did very much so stumble:
Tenacity, extraordinary self-confidence, and infallibility are great traits. ... So, is Chrome a color managed browser? There's a good amount of dissension on that issue. And somehow things got muddy along the way when discussing how an aRGB image when displayed on an sRGB device often doesn't look the same.
Quite a mashup of red herrings and garden paths today...
-
Here's what happens with a neutral gradient from RGB 200 to 255:
As you can see, the AbsCol tone curve is more luminous until it reaches the clipping point, about 240 with this paper then the RelCol exceeds it by about 1 L*. There are reasons why AbsCol is discouraged for normal photo work and this is a major one.
-
Tenacity, extraordinary self-confidence, and infallibility are great traits. ... So, is Chrome a color managed browser?
You really don't know, do you. Or how to check. Would you like me to teach you how to confirm if it's color managed or not? Ask.
I don't have an answer yet for you because I don't use Chrome and I don't run Windows. But I know how one can easily check if it is or isn't. Since you admit you use it, then tell us it's not color managed, then ask if it is, all I can do is wonder how much you don't know about this subject....
There's a good amount of dissension on that issue.
It would take a knowledgeable technical savvy person about 1 minute to figure it out.
And somehow things got muddy along the way when discussing how an aRGB image when displayed on an sRGB device often doesn't look the same.
Muddy like this: It's also streamlines the process of going from aRGB to sRGB for screen display.
Quite a mashup of red herrings and garden paths today...
Who joined this discussion today? « Reply #21 on: Today at 02:35:48 PM »
-
How do you accomplish that when, as Doug suggests, the highlights are clipped:Curves layers can't bring back clipped detail or data, only make them less than RGB 255 (Lstar 100).
If there is data in the highlights, use a curve to ensure the data doesn't get crushed. If there isn't any data to begin with (255, 255, 255), paint it in if so desired 30 years of retouching and a high degree of competency in drawing helps, a lot.
Stills photography is about rendering three dimension onto a two dimensional plain. As for fine art and commercial photography, if one is able to "draw" or "paint" credible detail into a scene using Photoshop, GREAT! Not being a photo journalist or a forensics photographer, what difference does it make?
I'd love to see some of your work. How about putting some up?
-
I've had a lot of fun bantering with you today, Rodney. I'm fortunate to have the time. That's one of the several benefits of being retired. Sorry I pulled you away from work. I know there's an opportunity cost to being distracted with trivialities.
-
If there is data in the highlights, use a curve to ensure the data doesn't get crushed.
It's clipped! There isn't data there. See again what Doug stated:
Here's what happens with a neutral gradient from RGB 200 to 255:
Now what you could and should do is examine if the raw file is clipped and if not, re-render it with this alteration of 200 to 255 in mind. That is, if you captured raw correctly (I'll not assume everyone here can do so).
I'd love to see some of your work. How about putting some up?
Some on my web page. Including shots taken in the Amazon when Michael, Chris, Jay Maisel and I taught a two week workshop for LuLa. And a workshop I taught with Greg Gorman (ever heard of him?). If you want to see something like a national ad for Microsoft, in the early 90's, when very few pro photographers were shooting AND doing their own Photoshop work, here you go. Got anything to show other than lampposts?
-
No, it's simply wrong. He may also be confused or just mistaken.
Edit: It appears he has edited his posts. He states that he may have been unclear, and was not claiming a difference in converting Adobe RGB <> sRGB using Abs. v. Rel Col.
In any case he seems to be focusing on the soft proofing and Abs. v Rel. I've pointed out the risks and issues of using Abs. Col. but if he tweaks it soft proofing as he indicated he did then he may well like the results and that's fine. There are cases where Abs. Col. can create a nice effect in deep shades as a result of the a* and b* shift that occurs at the black point. I've seen it maybe once or twice but it's rare. The negatives of having high key clipping are the major reason I don't use Abs. for photos. I only use Abs. for repro work.
Absolute rendering intent is great for the style of my nighttime photos. I do not use Absolute for general work. Doug hints that I do this for the aesthetic. And that is the point: it's an aesthetic that meshes with my intentions as an artist.
-
When you have an image in an RGB D50 media space, like ProPhoto, the convert to Adobe RGB or sRGB the Absolute Intent in Apple and Microsoft's engines attempts to create the same "colors" that are would occur if the displays were switched from D50 to D65.
Which is exactly what should happen - that's what "Absolute" means!
If you don't want to render the source D50 as D50 on your D65 display, don't use absolute intent.
[ Basically the ICC have got it wrong. There is a fundamental difference between print and display -
a prints white point is primarily controlled by the illuminant. A display doesn't work that way -
the "illuminant" is part of the display and can't be changed independently.
Absolute intent should correspond to measured colorimetric values - it should be the bedrock
from which all other color interpretations of the device can be derived, and while a spectrometer print
measurement XYZ can be with made any virtual illuminant (ICC standard being D50), the display
white point is a property of the display, not the measurement instrument.
So a CMM with Absolute intent that doesn't return instrument measurement values,
is broken in my view.
If you want some other effect in your soft proofing, such as rendering a relative paper white tint
preview, then do it by setting your color management rendering intent appropriately,
i.e. Source profile intent = Absolute, Destination profile (i.e. display) intent = Relative.
Redefining display profile Absolute intent to be the same as Relative colorimetric
wrecks fundamental functionality for no good reason (I'm looking at you, ICC V4!).
]
-
"It's clipped! There isn't data there. See again what Doug stated:
Here's what happens with a neutral gradient from RGB 200 to 255:
Now what you could and should do is examine if the raw file is clipped and if not, re-render it with this alteration of 200 to 255 in mind. That is, if you captured raw correctly (I'll not assume everyone here can do so)."
The only clipped area is the light from the LED bulb in the lamp, as it should be. Have you ever looked into a bright light in the dead of night? The Raw file is fine--sure, the light is clipped.
As an aside, thank you for posting a picture. Though the CRT monitor screens look off and it's hard to read the alpha numeric symbols composited (with halation artifacts) onto the screens. I'm surprised you didn't do some retouch work on his left eye. The dim catch light on his left eye bumps into the specular highlight on the rim of his glasses. Although I could, I wont get into the aesthetics. Do you have anything more current? This looks like it was taken fifteen to twenty years ago.
About your credentials: Now I know you know I know.
-
Which is exactly what should happen - that's what "Absolute" means!
If you don't want to render the source D50 as D50 on your D65 display, don't use absolute intent.
[ Basically the ICC have got it wrong. There is a fundamental difference between print and display -
a prints white point is primarily controlled by the illuminant. A display doesn't work that way -
the "illuminant" is part of the display and can't be changed independently.
Absolute intent should correspond to measured colorimetric values - it should be the bedrock
from which all other color interpretations of the device can be derived, and while a spectrometer print
measurement XYZ can be with made any virtual illuminant (ICC standard being D50), the display
white point is a property of the display, not the measurement instrument.
So a CMM with Absolute intent that doesn't return instrument measurement values,
is broken in my view.
If you want some other effect in your soft proofing, such as rendering a relative paper white tint
preview, then do it by setting your color management rendering intent appropriately,
i.e. Source profile intent = Absolute, Destination profile (i.e. display) intent = Relative.
Redefining display profile Absolute intent to be the same as Relative colorimetric
wrecks fundamental functionality for no good reason (I'm looking at you, ICC V4!).
]
Yes? ... This discussion, I thought, was about setting the rendering intent to "Absolute" for making a print. Soft proofing is a simulation tool. And when I print nighttime scenes on resin paper with the intent set to absolute rendering the results are better and actually more predictable.
Not disputing your commentary, it's just not germane to the particular application I brought up.
-
Which is exactly what should happen - that's what "Absolute" means!
...
[ Basically the ICC have got it wrong.
...
I don't disagree but the ICC is the authority on the meaning of the words it uses for it's standards and they chose a different meaning than the CIE which conforms with the usual meaning in color science. It's not the only place they differ. Another notable difference is the ICC specified that reflectance spectral data should be taken on paper on a white backing whereas the CIE specifies black backing.
However, there are interesting side effects to using the CIE (and Microsoft/Apple CME's) definition. ProPhoto no longer would contain all of sRGB for one. Joofa had a post about that some time back. But using the ICC definition sRGB is contained within ProPhoto.
The ICC gets to play Humpty Dumpty and Absolute Colorimetric means exactly what they say. No more, no less. At least within their domain.
-
The only clipped area is the light from the LED bulb in the lamp, as it should be. Have you ever looked into a bright light in the dead of night? The Raw file is fine--sure, the light is clipped.
Got a raw histogram to show that fact?
As an aside, thank you for posting a picture. Though the CRT monitor screens look off and it's hard to read the alpha numeric symbols composited (with halation artifacts) onto the screens. I'm surprised you didn't do some retouch work on his left eye. The dim catch light on his left eye bumps into the specular highlight on the rim of his glasses. Although I could, I wont get into the aesthetics.
I'll try to dig up the art director from the LA office of BBDO and send her your comments.
What ad agency were you an AD at?
Do you have anything more current? This looks like it was taken fifteen to twenty years ago.
It IS well over 20 years since that was shot. Do you ever read what people post before you bang on that keyboard of yours? Because I clearly stated that, but will use formatting to draw your attention to exactly what I stated:
If you want to see something like a national ad for Microsoft, in the early 90's, when very few pro photographers were shooting AND doing their own Photoshop work, here you go
You want something more recent, can you read and understand this:
Some on my web page. Including shots taken in the Amazon when Michael, Chris, Jay Maisel and I taught a two week workshop for LuLa. And a workshop I taught with Greg Gorman (ever heard of him?)
About your credentials: Now I know you know I know.
Tell us about your credentials; you're retired, from what? It isn't professional photography, that was made clear in our last discussion when you falsely and incorrectly stated I was 'just a technician' (http://forum.luminous-landscape.com/index.php?topic=124581.msg1042074#msg1042074) and when I asked about your professional photography client list, you had nothing to share (but I did). So you did exactly what to feed your family before retiring to be 'an artist' shooting lampposts. Got anything else photographic to share?
-
Yes? ... This discussion, I thought, was about setting the rendering intent to "Absolute" for making a print.
Now what gave you that idea? Please refer to post #1 where the OP specifically states:
Using Edit > Convert to Profile vs Soft Proofing
IN BOLD!
You yourself brought up soft proofing in your 2nd post.
Not disputing your commentary, it's just not germane to the particular application I brought up.
And you're not the OP.
-
There is a fundamental difference between print and display -
a prints white point is primarily controlled by the illuminant. A display doesn't work that way -
the "illuminant" is part of the display and can't be changed independently.
And I suspect how we perceive those two differing media (print vs. display) plays a role, no?
-
And I suspect how we perceive those two differing media (print vs. display) plays a role, no?
Graeme's Argyll uses Bradford adaptation for profiles by default. Bradford provides better Colorimetric printer profiles at the cost of slightly increased Abs. Col. variance using Photoshop since the latter uses the ICC's "wrong von Kries" formula. This produces only minor differences in both RelCol or AbsCol from ICC conformant printer profiles unless the paper white is pretty far out. Argyll provides an option to conform with ICC profiles:
I think it's a bigger factor in displays. However, it's also a different issue than the interpretation of AbsCol by ICC v CIE and the resulting differences in ACE and Microsoft/Apple engines.
See this for details:
http://www.argyllcms.com/doc/Environment.html#ARGYLL_CREATE_WRONG_VON_KRIES_OUTPUT_CLASS_REL_WP
-
I had expected a certain degree of error with OEM profiles, given manufacturing tolerances, etc., but never quite expected it to be that off. Using the OEM profiles probably wouldn't really be a concern for regular photo printing for an amateur like me, the Pro-10 is already much better than most photo printers I've used over the years. But I did as you suggested yesterday and waited until today to measure my results…
ore punch. The colors are toned down using the custom profile.
Drying time makes only a small difference in your dE measurements so don't worry about drying overnight.
Drying time v dE00s for the 5, in gamut CC neutral patches. The white patch is out of my paper's gamut.
Drying hrs: 0.6, Neutral patches dE00 Ave: 0.45 Max: 0.59
Drying hrs: 3.9, Neutral patches dE00 Ave: 0.45 Max: 0.60
Drying hrs: 8.1, Neutral patches dE00 Ave: 0.43 Max: 0.59
Drying hrs: 23.3, Neutral patches dE00 Ave: 0.40 Max: 0.55
Drying hrs: 43.6, Neutral patches dE00 Ave: 0.40 Max: 0.57
Drying hrs: 68.1, Neutral patches dE00 Ave: 0.39 Max: 0.55
Drying time v dE00s for the 18 CC colored patches
Drying hrs: 0.6, Colored patches dE00 Ave: 0.54 Max: 0.83
Drying hrs: 3.9, Colored patches dE00 Ave: 0.53 Max: 0.83
Drying hrs: 8.1, Colored patches dE00 Ave: 0.51 Max: 0.81
Drying hrs: 23.3, Colored patches dE00 Ave: 0.48 Max: 0.79
Drying hrs: 43.6, Colored patches dE00 Ave: 0.47 Max: 0.79
Drying hrs: 68.1, Colored patches dE00 Ave: 0.46 Max: 0.79
-
The ICC gets to play Humpty Dumpty and Absolute Colorimetric means exactly what they say. No more, no less. At least within their domain.
Well - no they don't. They can't magically make all those multi-millions of sRGB and AdobeRGB profiles out there conform to a new definition of how to make display profiles, nor can they magically changed how existing CMMs and applications work, nor can they dictate how CMM's implement their APIs.
So for instance, using ArgyllCMS Absolute still means Absolute.
-
Andrew, you asked me to post some of my pictures (whether for you to pounce on or to admire, I don't know).
Reminiscing about the old days, it's too bad you had to struggle with crude tools to make pictures for Microsoft. During the early 90s, I used Silicon Graphics machines and Maya (FKA Alias) software along with a bevy of custom applications while having unlimited access to a studio equipped with a high end video camera rigged for motion control (X,Y,Z,R, pitch, yaw), Mole Richardson lights/modifiers, and any kind of grip equipment imaginable. I often combined CGI with digital video captured on the motion-control rig. I worked on projects for WGBH (Boston), ESPN, The Cartoon Network, The Sci-Fi Channel, Ocean Spray, Heidelberg, Polaroid, CVS, and more.
In the 1980s, I produced, designed, and photographed motion graphics and FX sequences onto 35mm cine film. Shooting with a pin registered camera, much like a 1920s Mitchell but rigged for motion control, was technically challenging.
Prior to retirement, I was the sole proprietor of a studio specializing in restoration, retouching, and reproduction of fine art and antique documents for NGO archives, museums, academic institutions, foundations, and a few Fortune 500 companies.
I studied dogs for years, learned how to photograph them, and earned a reputation as a dog portrait photographer known for clean minimalist portraits that captured the true character of pet dogs. I enjoyed that work the most. I guess you like dogs too as your moniker suggests.
You asked me to post something other than a "lamp post." Here are a few. The picture of the grassy field was taken last week with a Pen F and a cheap $59 reconditioned kit lens. The "Field" required hardly any work in post. I printed it on premium luster paper, relative intent (I observed no difference when toggling BPC on and off). The shot is more or less straight out of the camera. ... "Haircut" was taken with a Pen F and the miraculous 12-100 f/4 Olympus pro lens. "Phillips 66" was taken, handheld, with a FF mirrorless camera. Can you tell how much time, if any, was spent tweaking the nighttime photos? Does it matter? The prints sparkle (both printed onto premium luster paper, rendering intent set to absolute).
Incidentally, I always shoot in RAW aRGB. The prints are more nuanced than what you are able to view on the screen. I made double sure the files were tagged for sRGB. If they are not tagged, I don't know where to attribute the fault.
-
Andrew, you asked me to post some of my pictures (whether for you to pounce on or to admire, I don't know).
Great photographs are made, not taken. Thanks for posting pictures you took.
Reminiscing about the old days, it's too bad you had to struggle with crude tools to make pictures for Microsoft
I purchased the first version of Photoshop two months after it was released in 1990. Yes, for those of us (few) Pro photographers working on desktop equipment in those days (Mac IIci 13" color displays, 8mb of RAM), producing commercial work was a bit of a struggle. Thanks to those efforts, and feedback to Adobe (I've been an alpha/beta tester since version 2.5), we see where Photoshop has progressed since then, and where your beloved SGI's are today (collecting dust).
I studied dogs for years, learned how to photograph them, and earned a reputation as a dog portrait photographer known for clean minimalist portraits that captured the true character of pet dogs. I enjoyed that work the most. I guess you like dogs too as your moniker suggests.
I'm a dog trainer too... ;)
Now it would be interesting to see some of these photo's that presumably where commercial (you got paid to produce them?) rather than what you provided. True character of a pet? Let other's be the judge.
The picture of the grassy field was taken last week with a Pen F and a cheap $59 reconditioned kit lens. The "Field" required hardly any work in post. I printed it on premium luster paper, relative intent (I observed no difference when toggling BPC on and off). The shot is more or less straight out of the camera. ... "Haircut" was taken with a Pen F and the miraculous 12-100 f/4 Olympus pro lens. "Phillips 66" was taken, handheld, with a FF mirrorless camera.
You seem to believe I (and others) care about the equipment used to produce these 'photo's; I'm not. That data if I can be so kind doesn't aid in making them anything but what they appear to me to be, basically snapshots.
Can you tell how much time, if any, was spent tweaking the nighttime photos? Does it matter?
It doesn't matter a lick. Which is why I didn't state anything like this with the one example of a national ad campaign I shot for Microsoft. BTW, it paid very well!
Incidentally, I always shoot in RAW aRGB.
No, you don't! Not if we are to accept that confusing and mostly incorrect statement as provided. It's similar to another confusing if not incorrect statement you made about sRGB/Adobe RGB (1998) and the use of the Colorimetric Rendering of which this topic is about. Raw is raw. It has no defined color space. Certainly not a rendered RGB working space like Adobe RGB (1998) unless you're talking about some other color space; you seem unable to name the RGB working space correctly so I'm not totally sure, based on your confusing text, what you may mean by aRGB. History here shows it's not safe to take your writings totally seriously so perhaps you're speaking of some other color space. But no, raw is not anything like Adobe RGB (1998)!
Now, IF you shoot raw+JPEG (which is a possibility), you could produce an Adobe RGB (1998) JPEG. Otherwise no, you're not shooting raw (or an acronym RAW) in aRGB/Adobe RGB (1998).
You'd do well to cease typing text like this and reading how this stuff actually works. Here's a white paper from the ICC I co- authored discussing the difference between rendering and encoding and the role of color spaces after rendering:
http://www.color.org/ICC_white_paper_20_Digital_photography_color_management_basics.pdf
-
Graeme's Argyll uses Bradford adaptation for profiles by default.
Fine but that isn't what I'm asking him about. Rather, how we humans perceive the color appearance of those two media's differently and that effect on that appearance.
-
Good morning, Mr. Rodney. I'm glad to see you won't let sleeping dogs lie.
-
Per your request Mr. Rodney, a small collection of my dog photos is here:Bob Dogs (https://bobrosinsky.com/portfolio/dogs/). My website is still in development, and I have postponed tweaking it and filling it up with more portfolios. So, I offer preemptive apologies. Peace and Love, Bob
-
Good morning, Mr. Rodney. I'm glad to see you won't let sleeping dogs lie.
Peer review is a bitch. What do we owe people who post technical text that's wrong?
-
Per your request Mr. Rodney, a small collection of my dog photos is here:Bob Dogs (https://bobrosinsky.com/portfolio/dogs/). My website is still in development, and I have postponed tweaking it and filling it up with more portfolios. So, I offer preemptive apologies. Peace and Love, Bob
Now that's more like it! VERY nice. You can shoot, no question. Now if we can simply get you to enter a technical topic and post technically correct comments, we're on the same page.
BTW, it is far, far more difficult to produce the very lovely images of dogs on your site, then learn and post technically correct text about color and imaging. So I have hope you can eventually do both.
-
All of the dog photos were commissions. I loved taking pictures of ordinary dogs for ordinary people. I loved not having to deal with ad agencies, art directors, account executives, etc. ...Being that you are curious about monetary remuneration, a session and a single photo printed onto 17" X 22" commanded $1,200. ... Due to severe autoimmune health problems, I could no longer photograph dogs. Although I usually took around 30 photos per session, the amount of time lying prone along with moving a light on wheels with one hand while framing the picture in the camera with the other became too much of a strain. Throughout the time I photographed dogs, I did a couple shoots for national campaigns. Although the pay was good, the stress of dealing with ADs, account execs, and superfluous people shuffling around in my studio was not conducive for taking pictures of dogs.
I've also taken studio photographs of birds, hedgehogs, and tarantulas. I only photographed snakes once and that was for Hill Holiday in Boston. Once was enough. I don't like bats, snakes, and rats.
-
I looked over all the images. The dog work is the best by far IMHO; quite stunning. Horizons; some very interesting and lovely. I will refrain from commenting on the other's.
I am probably equally guilty as many of the photographers who's work I view: too many photographs. Less is more. I suppose that's why in the old days, I'd get a rep's who among other tasks, aided in being an good photo editor for promotion of my work.
It's too bad we got off on the wrong foot about photography, I know why (http://forum.luminous-landscape.com/index.php?topic=124581.msg1042128#msg1042128) The technical stuff about your problems posting about color, outlined and with suggested outside peer reviewed articles, stands. ;)
-
Well - no they don't. They can't magically make all those multi-millions of sRGB and AdobeRGB profiles out there conform to a new definition of how to make display profiles, nor can they magically changed how existing CMMs and applications work, nor can they dictate how CMM's implement their APIs.
So for instance, using ArgyllCMS Absolute still means Absolute.
lol. That's why I made the allusion to Humpty Dumpty in Lewis Carroll's exchange.
The ICC has addressed their unique interpretation of "Absolute" this way:
The "absolute colorimetric" rendering intent is renamed "ICC absolute colorimetric", to avoid confusion with CIE absolute colorimetry.
Appears that Adobe's CME adopted the revisionist definition even on long existing profiles. I would actually prefer to use Microsoft's but it turns out to have significant errors converting sRGB in the standard intents.
-
Now that's more like it! VERY nice. You can shoot, no question. Now if we can simply get you to enter a technical topic and post technically correct comments, we're on the same page.
BTW, it is far, far more difficult to produce the very lovely images of dogs on your site, then learn and post technically correct text about color and imaging. So I have hope you can eventually do both.
Thank you for the compliment. ... I have technical chops. Admittedly, I'm not a color scientist, so my argot is sometimes off the mark. But I can assure you, I know how to read a historgram, edit photos non-destructively, and produce exhibition quality prints. I understand and use color managed workflow--I don't waste material. I am in tune with what I see on the screen and what the printer spits out. ... I have the tools and knowledge for generating ICC profiles, although in this day and age, I settle for canned profiles. I cannot create a better ICC profile for Epson papers for an Epson P800. When my studio was in full swing, I often made custom profiles, and on occasion hired "experts in the field" to create them. I am sure you have the skill to create better ICC profiles than the canned ones. ... After working with cine film, color timing, reading control strips, etc., I burned out on minutia. I self-learned digital photography. ... I enjoy straight ahead jazz. Some of the best musicians were not able to read charts. Perhaps they had dyslexia, lacked interest, or stubbornly refused. ... Apparently, a guy such as myself lacks the qualifications to participate on this forum. I'll save for the color scientists and the newbies who request help with technical issues. ... I would like to be on good terms with you. So, moving forward, I will practice restraint and not ruffle feathers on the Colour Management forum.
EM-5 II (http://theonlinephotographer.typepad.com/the_online_photographer/2015/08/the-e-m5-mark-ii-dissenting-opinion-shell-station.html)
Resuscitate Tintype (http://topdogimaging.net/blog/restoring-a-photograph-from-the-1870s)
Esoteric Review (https://petapixel.com/2016/06/11/review-cambo-actar-24mm-f3-5-mirrorless-camera/)
-
Now that's more like it! VERY nice. You can shoot, no question. Now if we can simply get you to enter a technical topic and post technically correct comments, we're on the same page.
BTW, it is far, far more difficult to produce the very lovely images of dogs on your site, then learn and post technically correct text about color and imaging. So I have hope you can eventually do both.
This is very true. You are both far more skilled photographers than I am. I come to it from the engineering side and a stint in a company that made various imager products with a focus on quality control issues, mechanical registration, color "accuracy" and such. My knowledge of printing tech., which was necessary for making certain custom targets with accurate colorimetry required deep diving into the physics as well as the more geeky aspects of ICC profiles.
-
Apparently, a guy such as myself lacks the qualifications to participate on this forum. I'll save for the color scientists and the newbies who request help with technical issues.[/url]
Not at all. First, you have good control over color management and understand soft proofing. While your use of Abs. Col. is unorthodox, you have combined it with good soft proofing practices and Photoshop adjustments to produce what you want and have avoided the gotchas associated with high key clipping because you soft proof effectively. Ultimately, color management is about enabling photographers to produce appealing art.
But to summarize why AbsCol compared to RelCol is rarely used in printing photos it's these two factors:
1. Clipping when L* exceeds the paper white.
2. Neutral colors are printed neutral and not shifted to the paper's actual white point. This creates pretty awful effects on prints with a white point that is more than 1 or 2 dE tint (a* and b* only). But it's only awful when the print has a white border. Borderless prints are printed more consistently with AbsCol as the printer takes out any tint the paper has. However, a significant tint reduces the printable gamut further beyond the effect in #1.
Soft proofing and adjusting as required is a way of dealing with the above without going into the tech. weeds since WYSIWYG. Still, while soft proofing isn't perfect for sundry reasons it is a huge leap above not soft proofing.
-
This is very true. You are both far more skilled photographers than I am. I come to it from the engineering side and a stint in a company that made various imager products with a focus on quality control issues, mechanical registration, color "accuracy" and such. My knowledge of printing tech., which was necessary for making certain custom targets with accurate colorimetry required deep diving into the physics as well as the more geeky aspects of ICC profiles.
Thanks, Doug. I'm not an engineer, nor am I a scientist. It's people like you who've made it possible to work in the digital domain. I also take my hat off to those who've advanced analog stills and motion photography--the same for lens designers, those who design image sensors, etc.. Funny thing, I did my graduate work at MIT, was awarded a two-year research fellowship, and was the first recipient of the Kepes Prize. I have a keen interest in visual perception, and dipped my toes in neural science. I became friendly with students, physicists, electrical engineers, computer scientists, etc. I was fortunate to be present during the time the Media Lab was set up.
My special interests centered on depth perception (I'm stereo blind as well as have difficulties with recognizing faces), afterimages, and designing experiments to test human color perception.
I'm currently a member of a makers space. Many of the folks there are scientists and engineers. Lately, I've been tinkering with OLED displays and building a robot that will read bitmap files and use RGB LEDs to draw for time exposures. I do not have a background in EE. Fortunately, I've got access to those who do. Having been a lifelong tinkerer has afforded me opportunities to acquire skills without necessarily understanding the underlying technology.
-
So, the project I've been struggling with the last 4 months has been to "simply" reproduce a bunch of old photos (40s-90s) as accurately as possible, e.g scan them, print them, and hold them up side-by-side and have them look pretty close. This has been a project that's been on my mom's to-do list for a while but I had been putting it off because my past experience with a non-color-managed workflow of scanning and printing had been extremely frustrating and disappointing. It's been suddenly made a priority because of family circumstances I don't need to get into, but the hope was to have some of this working quite a while ago.
Last year I tried doing some research so that I could "do it right" - so I bought an Epson v800, i1Studio, a Canon Pro-10 and a Pro-100 (pigment vs dye), subscribed to Adobe, bought VueScan, got sample paper packs from different sources, a Rotatrim, and even sourced an old English Deckle cutter that nearly perfectly matches the jagged cuts of the old B&W photos – it's surprisingly awesome btw. Anyway, I tried to jump right in and just push the buttons and make it work and everything looked like crap. I spent months researching everything I could understand, with a lot of misinformation and red herrings. I was very disappointed with the X-Rite results, but I think some of that was my ignorance, broken printer profiles, and a lack of documentation. I spent months reading, experimenting, re-reading, playing with Argyll CMS (which really brings a lot of capability to my i1Studio!), etc. without much luck, landing me here.
I have a computer science degree, I write software, repair computers, and consult on other IT-related areas so I'm comfortable with the command line, working with numbers, writing scripts to get things done, etc. But this color management stuff is new to me obviously, and I'm not an artist and have trouble speaking in color. I would like to think that there's an ideal workflow for my reproduction project, but there's a surprising lack of documentation for scanner-based reproduction work.
Like I mentioned back on page 1, most of the stuff I came across specifically said to never use Absolute, but in reading some other threads on this forum I gave it a shot and it helped quite a bit with what I was trying to do. Now, it's possible it's still wrong and it's just compensating for some other broken part of my workflow which I'm willing to accept.
One of the threads I gathered info on using Absolute was this: Getting Blue Colors "right" - http://forum.luminous-landscape.com/index.php?topic=106697.0 There were a few choice quotes from that thread:
If a friend gives you a 4x5 borderless print and asks you to enlarge it to an 8x10 the proper way to do it is to scan it with a scene referred profile. Resize it. Then print it using Absolute Colorimetric Intent. As long as the print media has a gamut that is as large as that in the scanned image there is no art to it. Just the science and the numbers. The OP is scanning something and wishes to reproduce it. A good scanner profile and AC printing is the best way to do it.
Graeme Gill had a post in that thread also about tracking the color through the workflow, which I'd like to do though I had trouble interpreting how to perform the exact steps. I might circle around to this again when I get back to scanning.
Doug also had good advice at the end of that thread:
Divide up the issues and address one at a time. Trying to get everything "right" without using instruments to measure each aspect separately, is a hard and frustrating process.
Which is what I'm attempting to do currently is break things up into manageable chunks. Thanks to the help I've received here I have improved my computer to printer results, at least numerically. Though I still have a follow-up about printing in absolute:
1. Given a hypothetically perfect paper profile: if printing an image using Absolute to two different papers that have a large enough gamut for the image, would the output look identical? I ask because even now that my paper profiles have dEs of around 1.0, there's still a visible difference in color temp between my test prints on different papers. They're very close, but I can tell that one looks warmer and the other colder.
Image white patch: 95.190, -1.030, 2.930
Plus Glossy II absolute: 94.190, -1.968, 3.164
Plus Glossy II white point: 94.5493467, -0.813282, -2.3624075
Pecos Gloss absolute: 94.770, -1.265, 3.539
Pecos Gloss white point: 94.959037, -0.0461172, -1.3828287
The Glossy II print looks a tad yellower to my eyes. The Glossy II is glossier if that matter.
2. To see what the affect would be with Relative (because it seems like it's the thing to use for most other circumstances), the same white patch for the Pecos Gloss shifts to 90.39, -1.093, 1.484. Despite the dramatic drop in lightness, the actual print seems to have more contrast. Maybe I don't need to get distracted by that just yet, but it wasn't what I was expecting.
At any rate, where I'm at now seem so look like the right direction for printing. Based on data I've collected today I think it's still primarily my scanning that's off by quite a bit. Not sure if I should continue on this thread with that topic or start a new one. I suppose it would still technically involve absolute intents.
-
Hi, Brian:
Kudos to you for making so much effort to reproduce the family photo album...hope your siblings and relatives appreciate the pains involved (not to mention the time and the money) to get them just right. My patience would have run out on day 3. I have postponed my own project to do something similar indefinitely for the fear of amount of work involved.
A couple of general points and a (wild) idea:
Given that the original photographs would have been printed on a wide variety of papers each with its own texture, reflectivity characteristics, and base color, spanning decades in time and technology - would it not be practically impossible to reproduce them all on a couple of inkjet papers using a generalized workflow?
Also, you are comparing the output against the original side-by-side - which is the most stringent test. If you show a reproduction to someone without the original, would they point out the discrepancy in the color of the sky or the grass, or the blue of the eyes or the tone of the sepia. If not, does it matter how close you are to the original? I am not saying you should lower your standards, but are you trying to achieve the impossible?
Now about the idea that has occurred to me before (what is that saying...a little knowledge is a dangerous thing) but haven't tried to see for myself: I am wondering since your final goal is to produce inkjet prints of the originals, why not "lump" the two separate profiles (one for the scanner and the other for the printer/paper) into a single unified profile that takes one straight to the print from the scanner. The potential profiling process could go this way: print the i1Studio target, then instead of measuring those output patches, scan them without using scanner profile, take the file back to a non-color managed software (like ACPU) and print the patches again. Now you run the i1 to create the combined profile. Finally for the reproduction, take the scanned image directly to Photoshop and print with this combined profile. Most likely I am overlooking the a pitfall or two in the process. Just some food for thought!
:Niranjan.
-
1. Given a hypothetically perfect paper profile: if printing an image using Absolute to two different papers that have a large enough gamut for the image, would the output look identical? I ask because even now that my paper profiles have dEs of around 1.0, there's still a visible difference in color temp between my test prints on different papers. They're very close, but I can tell that one looks warmer and the other colder.
They should look the same under the following conditions.
Use the same spectro to measure colors on the original as well as prints and use that same spectro when creating profiles for the paper. There are differences between spectros and using difference ones for measuring color and creating profiles is best for making reproductions. Differences are exacerbated when using an OEM profile on a paper with OBAs as it is unlikely to be an M2 profile and measuring colors with an M2 profile.
The prints should not have an unprinted border. If they do you will adapt to the white borders. This will cause the prints to look somewhat different even if the body of the print is colorimetrically the same. If the originals have a border scan the complete originals including the border and trim the prints.
Texture can alter colors because the original and repro may reflect specular components differently. Spectros measure color that is reflected from light at 45 degrees. This minimizes effects of luster/glossy/matte surfaces but only if the objects are also illuminated at 45 degrees.
Beware of colors near the paper's max luminance. This is often an area where the paper is out of gamut with the image. For instance a paper with a white point of Lab(95 0 -2) can print Lab(95,0,0) quite well since only a slight amount of yellow is needed. OTOH, a paper with a white point of Lab(95,0,2) cannot because the mix of cyan and magenta needed to decrease b* will drop the luminance and you will get something like Lab(94,1,1) as it is out of gamut. Out of gamut colors are mapped to the closest gamut boundary.
Another issue with repro work occurs if the original has significant fluorescence. If so you will need to find a photo paper that has similar fluorescence if the prints may be viewed by window light or outside. However, you should still use M2 (uV cut) profiles because the scanner has little/no uV. Similar substrate fluorescence will do the best job of making the prints look close to the originals when viewed outside.
-
Given that the original photographs would have been printed on a wide variety of papers each with its own texture, reflectivity characteristics, and base color, spanning decades in time and technology - would it not be practically impossible to reproduce them all on a couple of inkjet papers using a generalized workflow?
That was my original assumption. Once I standardized on using the ColorChecker images to test, and learned how to measure and compare my results (thanks to Doug), I realized I had a more basic problem on my hands. Now, I do suspect there could be some photos that a generic workflow won't apply perfectly to, but given the 100s that are here, I'm hoping that most will be "good enough". To compound the problem though, the originals that my family want reproduced the most are super deep glossy prints. They look pretty good printed to glossy paper, but of course there's a catch :) these photos have writing on the back that they want preserved too, so I had to limit myself to papers that supported printing on the back too… so I'm going to have to accept that my prints might have a different look if I'm using papers with different characteristics. I think I just need to stay away from any papers that are too different, otherwise I'll run into issues with smaller gamuts that would complicate matters more.
Also, you are comparing the output against the original side-by-side - which is the most stringent test. If you show a reproduction to someone without the original, would they point out the discrepancy …. If not, does it matter how close you are to the original? I am not saying you should lower your standards, but are you trying to achieve the impossible?
I agree with all of this. :) I will say, I was about to give up many times, the learning process can be discouraging, and I'm often interrupted by actual work I need to do – this stuff is time consuming! I had things to where I would classify it as "acceptable" and at least better than amateur work but then I finally got my prints dialed in and now I'd say they're like 92% there! So now I'm quite excited that I might be able to squeeze out just that little bit more if I can get my scanning dialed in too. When I print the ColorCheckers, any difference side-by-side on different papers is imperceptible (to me) under normal lighting – aside from glossiness. If I hold them under different lights and squint I might be able to pick them apart. The ColorCheckers aren't a great stress test for clipping or other gamut issues (I don't think), but they've made a good practical test that's relatively easy for me to measure (now).
The prints should not have an unprinted border. If they do you will adapt to the white borders. This will cause the prints to look somewhat different even if the body of the print is colorimetrically the same. If the originals have a border scan the complete originals including the border and trim the prints.
I did figure that out eventually. I didn't believe it could make that much of a difference, but I'm now in the habit of always cutting off the excess paper borders and it helps. They're close enough now that I sometimes mistake the prints for the original unless I'm really looking for it. So I'd say that's progress!
-
I will say, I was about to give up many times, the learning process can be discouraging, and I'm often interrupted by actual work I need to do – this stuff is time consuming!
Discouraging, but ultimately rewarding.
I've always found I learn the most from what doesn't work. One gets to break it down to find the whys. The result is knowledge.
-
They should look the same under the following conditions. …
Looks like I'm now doing things correctly, given what I have to work with.
Beware of colors near the paper's max luminance. This is often an area where the paper is out of gamut with the image. For instance a paper with a white point of Lab(95 0 -2) can print Lab(95,0,0) quite well since only a slight amount of yellow is needed. OTOH, a paper with a white point of Lab(95,0,2) cannot because the mix of cyan and magenta needed to decrease b* will drop the luminance and you will get something like Lab(94,1,1) as it is out of gamut. Out of gamut colors are mapped to the closest gamut boundary.
Hopefully if I'm sticking with glossy papers with a large gamut that are relatively bright, my originals should hopefully fit within in the gamut of my new papers well enough for now. If/when I start branching out or bumping into related problems I'll need to come back to this. I understand it at a high level, would like to know it from a more detailed level some time.
Discouraging, but ultimately rewarding.
I've always found I learn the most from what doesn't work. One gets to break it down to find the whys. The result is knowledge.
I've learned a lot from your help in a short period of time. My early trial-and-error raised a ton of questions, now I have a direction and feel like I'm progressing. Quite a bit has been over my head (even in this thread), but that's fine because as I learn I can go back and absorb the things I missed the first time around. I hope once I get this project moving along a bit I can really dive into this subject even more and try to learn more of the technical aspects.
I think that right now, for all practical purposes, and given my current resources, I should probably consider my printing to be close enough and get on with the next step. :)
-
STEP 2: Scanning
Now that I feel more confident that my printer output is on target, the next step is to look into my scanning. To try to keep things consistent and relevant with the previous discussions, my test photo is one of my printed ColorCheckers that I've already taken measurements of.
I'm not sure where I should begin as there's so many variables, but I think I'll start with some basic measurements and then break it down from there as necessary?
Test Input
- Source photo is the same BabelColor ColorChecker TIFF in native Lab color space that I've been using throughout my testing.
- I have Photoshop's color settings set to absolute colorimetric.
- I printed the ColorChecker to Canon Photo Paper Plus Glossy II using an absolute colorimetric intent and a custom profile that I've recently made using i1Studio.
Scanner profiling
- I have multiple options for profiling targets: I have a Digital ColorChecker SG (DCCSG), 3 IT8 reflective targets, and a ColorChecker 24 Mini. (I'd love to try making my own reference files, but frankly it's a pain doing spot-checks with the i1Studio unless the color area is large enough
- For software I have: Epson Scan, VueScan Pro, and SilverFast 8. (Let's forget I mentioned SilverFast unless it can work miracles.)
- For doing the profiling I have i1Studio and Argyll CMS.
Process
- Scan target using Epson Scan set to "No Color Correction" which disables everything. Output as 48bit untagged TIFF.
- Create profiles with i1Studio and Argyll.
- Scan my printed ColorChecker in Epson Scan using the exact same settings.
- Open the untagged TIFF in Photoshop and *assign* the profiles.
- Record the Lab values from the info panel
Findings
i1Studio doesn't give me any diagnostic output. But with Argyll I can run profcheck to output dE values of the profile vs the target reference values:
Argyll -aX -qu DCCSG: Avg dE(76) = 0.64, max dE(76) = 2.8.
Argyll -aX -qu DCCSG: Avg dE(2000) = .45, max dE(2000) = 2.17.
Argyll -aX -qu IT8: Avg dE(76) = 0.64, max dE(76) = 2.8.
Argyll -aX -qu IT8: Avg dE(2000) = .21, max dE(2000) = 1.52.
Things get a bit messy when I apply the profiles to the scanned image of the ColorChecker though:
Argyll -aX -qu DCCSG: Avg dE(76) = 3.9894, max dE(76) = 10.806.
i1Studio DCCSG: Avg dE(76) = 3.925, max dE(76) = 9.555.
Argyll -aX -qu IT8: Avg dE(76) = 3.117, max dE(76) = 7.451.
(The avg and max don't really tell the whole story so I attached the individual color values and dEs.)
Over the last few months I've done a lot of reading about scanning. I understand the basics that scanners see differently than people or spectrophotometers, and different papers being scanned can lead to different results. But I have a feeling that I should be able to achieve better results that I am. If not, please let me know. I'd be curious if anyone's done similar testing and what kind of results you achieve.
I also get different results if I start with different scan settings. For example if I go with a VueScan 48bit RAW vs 24bit RAW vs Epson Scan. Given the shear number of combinations between settings and profile creation, I'd love to find a way to streamline my testing process. (Or if someone can just tell me what's best! Ha.) For example, when I use Argyll I have a script to output every combination of profiles, output profcheck results, and output TIFFs with embedded profiles all in one go. If I could take the RGB color values of my Epson Scan scan (attached) and run them through each profile that I create (including the X-Rite profiles) using some command line tool that would be great. I tried using a couple Argyll tools, but probably used them wrong as the output seemed like nonsense.
So, much like my printing analysis, I'd like to know if my testing is sound, if I should be able to achieve better results, and if so, the most efficient way that I can test so that I can ask smart questions and not waste too much time. :)
Thanks!
-
Looks like I found an answer to one of my questions. If I want to convert my scanned RGB values to Lab through a profile, it seems I can use icclu (part of Argyll). I had tried it before, but I was missing the scale parameter.
# Convert color list
input_colors="Epson Scan Scanned RGB.txt"
output="Results"
profile="_Epson v800 DCCSG Epson Scan 2018-05-11 QU (XYZ cLUT M+S) -aX.icc"
icclu -s255 -ff -ia -pl -v2 "$profile" < "$input_colors" > "$output Forward Absolute Lab.txt"
icclu -s255 -ff -ir -pl -v2 "$profile" < "$input_colors" > "$output Forward Relative Lab.txt"
icclu -s255 -ff -ip -pl -v2 "$profile" < "$input_colors" > "$output Forward Perceptual Lab.txt"
Aside from very minor variations the results match up with what I see in Photoshop's info panel and correspond to the intent I choose in Photoshop's color settings. I could whip up a program to take the output and run it to a spreadsheet. It'd be great to find a way to automate the dE calculation between the results and the original.
-
Looks like I found an answer to one of my questions. If I want to convert my scanned RGB values to Lab through a profile, it seems I can use icclu (part of Argyll). I had tried it before, but I was missing the scale parameter.
Aside from very minor variations the results match up with what I see in Photoshop's info panel and correspond to the intent I choose in Photoshop's color settings. I could whip up a program to take the output and run it to a spreadsheet. It'd be great to find a way to automate the dE calculation between the results and the original.
You might want to start a new topic on profiling scanners. It's a fairly complex area because there is little to no info on the spectral characteristics of the light profiles use to scan. The same problem applies to cameras. It's a specialized art. Long ago I made scanner profiles but don't currently have a good scanner. There are people here that have done significant work with scanners.
Also, you might want to look into getting Matlab for processing files. They have a low cost version for non-commercial home use. It's really hard to beat. I use it extensively for working with images, CGAT files and such and have made scripts and functions that calculate dE76 as well as dE00. It think Matlab and the Imaging toolboc can be had for about $200. Would be happy to share these with you if you take the plunge.
-
Unlike some camera profiles, scanner ICC profiles are easy to make and are effective. The target plays a role (the gamut of the scanner profile can't exceed the target for one). Scanners and cameras don't really have a color gamut. At least some scanners with software that support profiles allow an idealized behavior of the data with controls over the rendering of the scan into either the scanner RGB or to some RGB working space. Output referred,one light source, so much simpler than a camera.
-
Unlike some camera profiles, scanner ICC profiles are easy to make and are effective. The target plays a role (the gamut of the scanner profile can't exceed the target for one). Scanners and cameras don't really have a color gamut. At least some scanners with software that support profiles allow an idealized behavior of the data with controls over the rendering of the scan into either the scanner RGB or to some RGB working space. Output referred,one light source, so much simpler than a camera.
Scanner profiles are almost always scene referred, not output referred, since they are used to replicate images that don't have the dynamic range of typical photos which look better output referred.
-
Discouraging, but ultimately rewarding.
I've always found I learn the most from what doesn't work. One gets to break it down to find the whys. The result is knowledge.
Agree wholeheartedly.