Well that's exactly what the late Bruce Fraser stated.
"A handy rule of thumb, though, is that for glossy stocks, the default view with Paper White and Ink Black unchecked will be the closest, while for uncoated stocks, the view with Paper White checked will generally be the most accurate."
and earlier in his same article with regards to having Paper White checked:
"As a result, your first reaction when checking paper white may be that your image just died before your eyes. I've become accustomed to looking away from the monitor when I check Paper White so that I don't see the change happen. This simple trick makes it a lot easier to accept the paper color displayed on the monitor as a true white. "
Andy, shouldn't "Simulate Paper White" only be unchecked for glossy paper and checked for matte finish?
Doesn't matter, it should be one if and when you wish to see an accurate soft proof taking the dynamic range of the paper, not the display into account. Its not gloss or matt paper specific in terms of when you'd use it. You should always use it when comparing display to final print.
Doesn't matter, it should be one if and when you wish to see an accurate soft proof taking the dynamic range of the paper, not the display into account. Its not gloss or matt paper specific in terms of when you'd use it. You should always use it when comparing display to final print.
So you are in agreement with one of the guys over at POTN (Canon Photography on the Net) who says it should always be checked.
The more I look at it, when compared to the final print, the following seems to produce the best results (for my setup):
Intent: Relative Colormetric (using perceptual gives too light and too red a representation of the final print).
Black Point Compensation: Unchecked. Checking it gives a strange result with the blacks much lighter than seen on the final print.
Simulate Paper Color: Maybe it depends on the room light when viewing the prints. However, the more I view things with it checked the more it appears to be a better match.
So you are in agreement with one of the guys over at POTN (Canon Photography on the Net) who says it should always be checked.
The more I look at it, when compared to the final print, the following seems to produce the best results (for my setup):
Intent: Relative Colormetric (using perceptual gives too light and too red a representation of the final print).
Black Point Compensation: Unchecked. Checking it gives a strange result with the blacks much lighter than seen on the final print.
Simulate Paper Color: Maybe it depends on the room light when viewing the prints. However, the more I view things with it checked the more it appears to be a better match.
My monitor is calibrated using a Spyder 3 Pro.
For these settings to be meaningful at all to anyone, you have to say what final print these are consistent with: what printer, what paper, what printer profile, what rendering intent and then any print driver settings that you have tweaked. All of this is very specific to a given printer target. Are you printing these at home? Smugmug's lab? With what paper, what paper profile, what rendering intent?
For these settings to be meaningful at all to anyone, you have to say what final print these are consistent with: what printer, what paper, what printer profile, what rendering intent and then any print driver settings that you have tweaked. All of this is very specific to a given printer target. Are you printing these at home? Smugmug's lab? With what paper, what paper profile, what rendering intent?
I'm soft proofing for Kodak Endura lustre printed by EZ Prints through Smugmug.
So you are in agreement with one of the guys over at POTN (Canon Photography on the Net) who says it should always be checked.
The more I look at it, when compared to the final print, the following seems to produce the best results (for my setup):
Intent: Relative Colormetric (using perceptual gives too light and too red a representation of the final print).
Black Point Compensation: Unchecked. Checking it gives a strange result with the blacks much lighter than seen on the final print.
Simulate Paper Color: Maybe it depends on the room light when viewing the prints. However, the more I view things with it checked the more it appears to be a better match.
It should be check when comparing prints to display or when you wish to edit the image based on what the print should look like.
Rendering intent is solely image specific. One might look better using RelCol, another Perceptual.
But when I check it the soft proof looks way off the printed image through EZ Prints. The blacks aren't nearly as deep as prior to soft proofing or on the finished print.
But when I check it the soft proof looks way off the printed image through EZ Prints. The blacks aren't nearly as deep as prior to soft proofing or on the finished print.
Prints do not have the same dynamic range as your screen. The deepest black on the paper is not nearly as dark as the deepest black either in sRGB or on your screen. If you don't use black point compensation everything in your image darker than the darkest black the printer can create gets mapped to pure black which means you lose shadow detail. If you use black point compensation, then the blacks get compressed (rather than clipped) to match the dynamic range of the print preserving details in the shdows.
Creating an sRGB image which looks good on screen and prints well is tricky because of the smaller dynamic range of print. Black point compensation works resonably well for many images but if you are willing to prep two different files, one for the screen and one for print, you can do better. When I am specifically prepping an image for print (and not for screen display) I turn off black point compensation and use the highlight/shadow tool along with some careful curves to recover shadow detail while viewing the image in soft proof mode. Done carefully, I can preserve local contrast in my shadows while compressing the image to fit the range of my paper.
Prints do not have the same dynamic range as your screen. The deepest black on the paper is not nearly as dark as the deepest black either in sRGB or on your screen. If you don't use black point compensation everything in your image darker than the darkest black the printer can create gets mapped to pure black which means you lose shadow detail. If you use black point compensation, then the blacks get compressed (rather than clipped) to match the dynamic range of the print preserving details in the shdows.
Creating an sRGB image which looks good on screen and prints well is tricky because of the smaller dynamic range of print. Black point compensation works resonably well for many images but if you are willing to prep two different files, one for the screen and one for print, you can do better. When I am specifically prepping an image for print (and not for screen display) I turn off black point compensation and use the highlight/shadow tool along with some careful curves to recover shadow detail while viewing the image in soft proof mode. Done carefully, I can preserve local contrast in my shadows while compressing the image to fit the range of my paper.
Thanks LiquidAir.
The black point compensation I'm referring to is the soft proof checkbox. And when I soft proof it's to specifically fine tune an image for print.
The colors between my editted and soft proofed image look pretty much identical, but the shadows/highlights/contrast are a tad different and are easy to adjust using curves. I think I'll just make an EZ Prints adjustment layer and start storing my "for print" copies in a separate folder on my Smugmug site.
But when I check it the soft proof looks way off the printed image through EZ Prints. The blacks aren't nearly as deep as prior to soft proofing or on the finished print.
Then there's something screwy with either the profile or your display calibration/profile.
BPC either does nothing (because no compensation is necessary, the conversions are correct) OR it corrects an error in black point mapping. It should always be used.
Then there's something screwy with either the profile or your display calibration/profile.
BPC either does nothing (because no compensation is necessary, the conversions are correct) OR it corrects an error in black point mapping. It should always be used.
I only see a difference in how deep the blacks look with black point compensation on or off when using relative colormetric. Using perceptual checking the black point compensation box does nothing.
I only see a difference in how deep the blacks look with black point compensation on or off when using relative colormetric. Using perceptual checking the black point compensation box does nothing.
Which points a suspecting finger at the profile. BPC controls (fixes) an issue with the mapping of source black to destination black. There is some gamut scaling difference in the two intents, you often see a difference in overall "density" between the two but you need to examine the mapping of source black (0/0/0) and how it maps to destination black which should also be 0/0/0. BPC should ensure that's happening. If you don't see this when the option is off, the profile is suspect.
Which points a suspecting finger at the profile. BPC controls (fixes) an issue with the mapping of source black to destination black. There is some gamut scaling difference in the two intents, you often see a difference in overall "density" between the two but you need to examine the mapping of source black (0/0/0) and how it maps to destination black which should also be 0/0/0. BPC should ensure that's happening. If you don't see this when the option is off, the profile is suspect.
Maybe I'm misunderstanding you but I don't believe this is correct. Using perceptual BPC is supposed to have little or no effect since that intent already maps source white to destination white and source black to destination black. I believe Adobe gives the option to use BPC with perceptual in case there is a problem with a profile, but normally it shouldn't be needed. In my case using perceptual I see zero difference with BPC checked or unchecked, which is good.
Using relative colormetric there can be some noticeable difference between BPC being checked or unchecked ... which is what I see, and is supposedly normal.
Also, I see no difference in deeper blacks between perceptual and relative colormetric when BPC is checked. Unchecked it's a different story. The issue I'm struggling with is that the prints I get look closer to relative colormetric with BPC off than to perceptual or relative colormetric with BPC on (simulate paper white also on for all cases). With perceptual selected and simulate paper white off it's closer for deeper blacks but not as good for middle tones.
Maybe this is the best simulation I can do with a laptop monitor (Dell Inspiron 1720) ...
I believe Adobe gives the option to use BPC with perceptual in case there is a problem with a profile, but normally it shouldn't be needed.
Correct, it shouldn't be needed if the profile is doing what it should in terms of mapping source to destination black. BPC is essentially a "fix" for profiles that don't map black correctly. Therefore, it should always be on. If the profile is doing things correctly, it isn't "necessary" and has no downside/effect but if the profile isn't built correctly, it will.
Correct, it shouldn't be needed if the profile is doing what it should in terms of mapping source to destination black. BPC is essentially a "fix" for profiles that don't map black correctly. Therefore, it should always be on. If the profile is doing things correctly, it isn't "necessary" and has no downside/effect but if the profile isn't built correctly, it will.
Thanks for the link, it's an interesting read, especially this bit:
"In most cases, doing RGB to RGB conversions without Black Point Compensation will produce desirable prints. However, depending on the profile, doing a conversion from RGB to RGB with Black Point Compensation can produce poor output with washed out blacks. It appears that this problem with some RGB profiles is dependent on the software that is used to generate the profile. Apparently there is a “Black Tag” feature in ICC profiles that in some cases can be used or unused depending on the software that actually creates the profile. For this reason, there is no hard and fast rule that says we should or should not use Black Point Compensation with RGB output. Our recommendation is to turn off Black Point Compensation with RGB output profiles or if possible, try a test with Black Point Compensation
Thanks for the link, it's an interesting read, especially this bit:
Fortunately, that's less and less an issue with RGB profiles since those old PS5 days. Modern ICC profile packages should be doing this "right". Of course, there could certainly be legacy profiles floating around.
Hi Andy - So I've ordered the calibration print and am now following your soft-proofing tutorial --- but I'm stuck in Photoshop.
I have Photoshop Elements 4.0 and don't see the "Proof Setup" under the "View" menu. Is it only CS2 that has this option?
Am I s*** out of luck?
And dont' worry, I know I'm not --- cause I'm asking Mr. Problem Solver himslef. (c:
Thanks
Hi
I found a software product called Elements+ ($12) which enables many of the PS functions not otherwise available in Elements, and one of them is softproofing. I got it principally for the softproofing feature and it seems to show the effect of the ezprint profile as descibed in the tutorial.
soft proofing again
[Unfortunately, since the lab isn't going to apply that profile (or maybe the desired rendering intent you set), its only somewhat useful.]
Not sure what you mean here.
However imperfect softproofing may be, and I am new to this myself, my point was that there is a way to enable the soft proof functionality in Elements, via Elements+, and use the smugmug tutorial.
You say that 'the lab isn't going to apply that profile'. Isn't the point of providing the ezprint profile for download that this file will be used by the lab in printing.
I am new to all this so maybe I am missing something important. But hoping to learn from you more experienced folks.
Not sure what you mean here.
However imperfect softproofing may be, and I am new to this myself, my point was that there is a way to enable the soft proof functionality in Elements, via Elements+, and use the smugmug tutorial.
You say that 'the lab isn't going to apply that profile'. Isn't the point of providing the ezprint profile for download that this file will be used by the lab in printing.
They demand you send the document in sRGB. So they are applying the profile (maybe?) and you have no control over the rendering intent, nor can you use Adobe ACE, Black Point Compensation or post edit the document after conversion, assuming its even being converted based on the profile you're using.
The proper workflow for full color management capability is to supply the profile, then let YOU convert the document, after which, the numbers go straight to the printer. The sRGB color space is totally out of the picture here.
So you MUST convert to sRGB then setup the soft proof to even get the correct soft proof since they demand sRGB from the get go.
The real question should be, why supply the profile when its not being used, at least by the creator of the image?
They demand you send the document in sRGB. So they are applying the profile (maybe?) and you have no control over the rendering intent, nor can you use Adobe ACE, Black Point Compensation or post edit the document after conversion, assuming its even being converted based on the profile you're using.
The proper workflow for full color management capability is to supply the profile, then let YOU convert the document, after which, the numbers go straight to the printer. The sRGB color space is totally out of the picture here.
So you MUST convert to sRGB then setup the soft proof to even get the correct soft proof since they demand sRGB from the get go.
The real question should be, why supply the profile when its not being used, at least by the creator of the image?
Thanks for taking the time to clarify this. So much to learn.
The real question should be, why supply the profile when its not being used, at least by the creator of the image?
The profile can be used to see what colors the service is capable of rendering. While I agree that you would have a lot more control if you could apply the profile yourself and supply the image that way, isn't it at least partly useful to have some idea of what colors they can produce so you can check/tweak your image if required?
Hi, I'm totally new to calibration/printing, so pardon the ignorance. I've gone through this thread, got down the proofs/baypoint icc stuff on PS no problem. It was suggested to order test prints from bayphoto and EZ prints so I did. My question is what next? I use only Macs. I would assume the next step is to hold the test print up to the image on the screen and adjust the built-in calibration tool (System Preferences-->Displays-->Color-->Calibrate) until they match as closely as possible. Am I correct or what is everyone else doing with regard to test prints?
Thanks,
Hi, I'm totally new to calibration/printing, so pardon the ignorance. I've gone through this thread, got down the proofs/baypoint icc stuff on PS no problem. It was suggested to order test prints from bayphoto and EZ prints so I did. My question is what next? I use only Macs. I would assume the next step is to hold the test print up to the image on the screen and adjust the built-in calibration tool (System Preferences-->Displays-->Color-->Calibrate) until they match as closely as possible. Am I correct or what is everyone else doing with regard to test prints?
Thanks,
I am new to this also and have tried my hand at the comparison you mention. It is problematic with normal room lighting. Ideally this should be done with a 'lightbox' that allows for a more accurate matching. Check out this url by Andrew Rodney about the process. http://www.ppmag.com/reviews/200411_rodneycm.pdf
I am new to this also and have tried my hand at the comparison you mention. It is problematic with normal room lighting. Ideally this should be done with a 'lightbox' that allows for a more accurate matching. Check out this url by Andrew Rodney about the process. http://www.ppmag.com/reviews/200411_rodneycm.pdf
Thanks, very helpful on the comparison. Regarding the actual calibration of the monitor, I'd be curious to hear how other Mac users have done this. The calibration assistant isn't really designed to be used with a print comparison, and the color profiles are rather limited. I've read good and bad about the calibration software like Pentone, lots of people say it causes a red hue.
Chris
Thanks, very helpful on the comparison. Regarding the actual calibration of the monitor, I'd be curious to hear how other Mac users have done this. The calibration assistant isn't really designed to be used with a print comparison, and the color profiles are rather limited. I've read good and bad about the calibration software like Pentone, lots of people say it causes a red hue.
Chris
Most people who really want to do this buy a screen calibration/profiler like i1 Display 2 or one of the Spyder products. In addition to tweaking some monitor settings, it creates a color proflie that describes what color your monitor produces so that color-managed software can display accurate colors on your monitor.
Thanks, very helpful on the comparison. Regarding the actual calibration of the monitor, I'd be curious to hear how other Mac users have done this. The calibration assistant isn't really designed to be used with a print comparison, and the color profiles are rather limited.
Eyeball calibration doesn't work (and its not consistent), you need hardware to do this properly.
Comments
Andy, shouldn't "Simulate Paper White" only be unchecked for glossy paper and checked for matte finish?
My Legacy SmugMug Site (not ready to migrate yet)
Portfolio • Workshops • Facebook • Twitter
Well that's exactly what the late Bruce Fraser stated.
"A handy rule of thumb, though, is that for glossy stocks, the default view with Paper White and Ink Black unchecked will be the closest, while for uncoated stocks, the view with Paper White checked will generally be the most accurate."
and earlier in his same article with regards to having Paper White checked:
"As a result, your first reaction when checking paper white may be that your image just died before your eyes. I've become accustomed to looking away from the monitor when I check Paper White so that I don't see the change happen. This simple trick makes it a lot easier to accept the paper color displayed on the monitor as a true white. "
My Legacy SmugMug Site (not ready to migrate yet)
Doesn't matter, it should be one if and when you wish to see an accurate soft proof taking the dynamic range of the paper, not the display into account. Its not gloss or matt paper specific in terms of when you'd use it. You should always use it when comparing display to final print.
Author "Color Management for Photographers"
http://www.digitaldog.net/
So you are in agreement with one of the guys over at POTN (Canon Photography on the Net) who says it should always be checked.
The more I look at it, when compared to the final print, the following seems to produce the best results (for my setup):
Intent: Relative Colormetric (using perceptual gives too light and too red a representation of the final print).
Black Point Compensation: Unchecked. Checking it gives a strange result with the blacks much lighter than seen on the final print.
Simulate Paper Color: Maybe it depends on the room light when viewing the prints. However, the more I view things with it checked the more it appears to be a better match.
My monitor is calibrated using a Spyder 3 Pro.
My Legacy SmugMug Site (not ready to migrate yet)
For these settings to be meaningful at all to anyone, you have to say what final print these are consistent with: what printer, what paper, what printer profile, what rendering intent and then any print driver settings that you have tweaked. All of this is very specific to a given printer target. Are you printing these at home? Smugmug's lab? With what paper, what paper profile, what rendering intent?
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
I'm soft proofing for Kodak Endura lustre printed by EZ Prints through Smugmug.
My Legacy SmugMug Site (not ready to migrate yet)
It should be check when comparing prints to display or when you wish to edit the image based on what the print should look like.
Rendering intent is solely image specific. One might look better using RelCol, another Perceptual.
Black Point Compensation should always be on!
Author "Color Management for Photographers"
http://www.digitaldog.net/
But when I check it the soft proof looks way off the printed image through EZ Prints. The blacks aren't nearly as deep as prior to soft proofing or on the finished print.
My Legacy SmugMug Site (not ready to migrate yet)
Prints do not have the same dynamic range as your screen. The deepest black on the paper is not nearly as dark as the deepest black either in sRGB or on your screen. If you don't use black point compensation everything in your image darker than the darkest black the printer can create gets mapped to pure black which means you lose shadow detail. If you use black point compensation, then the blacks get compressed (rather than clipped) to match the dynamic range of the print preserving details in the shdows.
Creating an sRGB image which looks good on screen and prints well is tricky because of the smaller dynamic range of print. Black point compensation works resonably well for many images but if you are willing to prep two different files, one for the screen and one for print, you can do better. When I am specifically prepping an image for print (and not for screen display) I turn off black point compensation and use the highlight/shadow tool along with some careful curves to recover shadow detail while viewing the image in soft proof mode. Done carefully, I can preserve local contrast in my shadows while compressing the image to fit the range of my paper.
Thanks LiquidAir.
The black point compensation I'm referring to is the soft proof checkbox. And when I soft proof it's to specifically fine tune an image for print.
The colors between my editted and soft proofed image look pretty much identical, but the shadows/highlights/contrast are a tad different and are easy to adjust using curves. I think I'll just make an EZ Prints adjustment layer and start storing my "for print" copies in a separate folder on my Smugmug site.
My Legacy SmugMug Site (not ready to migrate yet)
Then there's something screwy with either the profile or your display calibration/profile.
BPC either does nothing (because no compensation is necessary, the conversions are correct) OR it corrects an error in black point mapping. It should always be used.
Author "Color Management for Photographers"
http://www.digitaldog.net/
I only see a difference in how deep the blacks look with black point compensation on or off when using relative colormetric. Using perceptual checking the black point compensation box does nothing.
My Legacy SmugMug Site (not ready to migrate yet)
Which points a suspecting finger at the profile. BPC controls (fixes) an issue with the mapping of source black to destination black. There is some gamut scaling difference in the two intents, you often see a difference in overall "density" between the two but you need to examine the mapping of source black (0/0/0) and how it maps to destination black which should also be 0/0/0. BPC should ensure that's happening. If you don't see this when the option is off, the profile is suspect.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Maybe I'm misunderstanding you but I don't believe this is correct. Using perceptual BPC is supposed to have little or no effect since that intent already maps source white to destination white and source black to destination black. I believe Adobe gives the option to use BPC with perceptual in case there is a problem with a profile, but normally it shouldn't be needed. In my case using perceptual I see zero difference with BPC checked or unchecked, which is good.
Using relative colormetric there can be some noticeable difference between BPC being checked or unchecked ... which is what I see, and is supposedly normal.
http://partners.adobe.com/public/developer/en/pdf/AdobeBPC.pdf
Also, I see no difference in deeper blacks between perceptual and relative colormetric when BPC is checked. Unchecked it's a different story. The issue I'm struggling with is that the prints I get look closer to relative colormetric with BPC off than to perceptual or relative colormetric with BPC on (simulate paper white also on for all cases). With perceptual selected and simulate paper white off it's closer for deeper blacks but not as good for middle tones.
Maybe this is the best simulation I can do with a laptop monitor (Dell Inspiron 1720) ...
My Legacy SmugMug Site (not ready to migrate yet)
Correct, it shouldn't be needed if the profile is doing what it should in terms of mapping source to destination black. BPC is essentially a "fix" for profiles that don't map black correctly. Therefore, it should always be on. If the profile is doing things correctly, it isn't "necessary" and has no downside/effect but if the profile isn't built correctly, it will.
Here's a real old PDF on the subject:
http://www.digitaldog.net/files/Black_Point_Compensation.pdf
Author "Color Management for Photographers"
http://www.digitaldog.net/
Thanks for the link, it's an interesting read, especially this bit:
My Legacy SmugMug Site (not ready to migrate yet)
Fortunately, that's less and less an issue with RGB profiles since those old PS5 days. Modern ICC profile packages should be doing this "right". Of course, there could certainly be legacy profiles floating around.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Hi
I found a software product called Elements+ ($12) which enables many of the PS functions not otherwise available in Elements, and one of them is softproofing. I got it principally for the softproofing feature and it seems to show the effect of the ezprint profile as descibed in the tutorial.
http://rwmj.smugmug.com/
Unfortunately, since the lab isn't going to apply that profile (or maybe the desired rendering intent you set), its only somewhat useful.
Author "Color Management for Photographers"
http://www.digitaldog.net/
[Unfortunately, since the lab isn't going to apply that profile (or maybe the desired rendering intent you set), its only somewhat useful.]
Not sure what you mean here.
However imperfect softproofing may be, and I am new to this myself, my point was that there is a way to enable the soft proof functionality in Elements, via Elements+, and use the smugmug tutorial.
You say that 'the lab isn't going to apply that profile'. Isn't the point of providing the ezprint profile for download that this file will be used by the lab in printing.
I am new to all this so maybe I am missing something important. But hoping to learn from you more experienced folks.
Thanks
http://rwmj.smugmug.com/
They demand you send the document in sRGB. So they are applying the profile (maybe?) and you have no control over the rendering intent, nor can you use Adobe ACE, Black Point Compensation or post edit the document after conversion, assuming its even being converted based on the profile you're using.
The proper workflow for full color management capability is to supply the profile, then let YOU convert the document, after which, the numbers go straight to the printer. The sRGB color space is totally out of the picture here.
So you MUST convert to sRGB then setup the soft proof to even get the correct soft proof since they demand sRGB from the get go.
The real question should be, why supply the profile when its not being used, at least by the creator of the image?
Author "Color Management for Photographers"
http://www.digitaldog.net/
Thanks for taking the time to clarify this. So much to learn.
http://rwmj.smugmug.com/
The profile can be used to see what colors the service is capable of rendering. While I agree that you would have a lot more control if you could apply the profile yourself and supply the image that way, isn't it at least partly useful to have some idea of what colors they can produce so you can check/tweak your image if required?
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
Thanks,
Crescent City Prints
Facebook Fan Page
Blog
I am new to this also and have tried my hand at the comparison you mention. It is problematic with normal room lighting. Ideally this should be done with a 'lightbox' that allows for a more accurate matching. Check out this url by Andrew Rodney about the process.
http://www.ppmag.com/reviews/200411_rodneycm.pdf
http://rwmj.smugmug.com/
Thanks, very helpful on the comparison. Regarding the actual calibration of the monitor, I'd be curious to hear how other Mac users have done this. The calibration assistant isn't really designed to be used with a print comparison, and the color profiles are rather limited. I've read good and bad about the calibration software like Pentone, lots of people say it causes a red hue.
Chris
Crescent City Prints
Facebook Fan Page
Blog
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
Eyeball calibration doesn't work (and its not consistent), you need hardware to do this properly.
Author "Color Management for Photographers"
http://www.digitaldog.net/