Pages: [1] 2   Go Down

Author Topic: Prints do not match monitor  (Read 16167 times)

ChuckZ

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 59
    • http://czhike88.smugmug.com/
Prints do not match monitor
« on: October 31, 2009, 12:48:02 am »

I processed several images using Photoshop CS4 and Adobe1998 color space on a Dell 2408 monitor calibrated with Spyder2express .  Next I softproofed them with the appropriate profile from a print company called WHCC and then sent the image files to WHCC for printing.  When I received the prints, the colors looked right, but the prints were darker than what the softproof shows on the monitor screen.  When I talked to a rep at WHCC, she told me that I needed to turn the brightness on my monitor down so that what shows on the monitor will be the same as what the print looks like.  I turned the brightness all the way down, but the print was still darker.  Any thoughts?  Thank you.
Logged

Larry Berman

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Prints do not match monitor
« Reply #1 on: October 31, 2009, 05:29:02 am »

David Brooks wrote an article for Shutterbug on why prints tend to be too dark:
Prints Too Dark



Quote from: ChuckZ
I processed several images using Photoshop CS4 and Adobe1998 color space on a Dell 2408 monitor calibrated with Spyder2express .  Next I softproofed them with the appropriate profile from a print company called WHCC and then sent the image files to WHCC for printing.  When I received the prints, the colors looked right, but the prints were darker than what the softproof shows on the monitor screen.  When I talked to a rep at WHCC, she told me that I needed to turn the brightness on my monitor down so that what shows on the monitor will be the same as what the print looks like.  I turned the brightness all the way down, but the print was still darker.  Any thoughts?  Thank you.
Logged
Larry Berman
[url=http://bermangraphics.

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Prints do not match monitor
« Reply #2 on: October 31, 2009, 08:42:11 am »

Where you say you turned your display brightness "all the way down" - what specifically does that mean in CD/mm2? It could be that "all the way down" isn't down enough. Were the prints from the lab VASTLY darker than what you see on the display. Recall, there is some inherent additional luminosity apparent from the display relative to a print simply because the former is transmitted light and the latter reflected light. One never totally overcomes this divide, so you need to make some mental adjustment for it - but in today's calibrated and managed environments, not a whole lot, why I'm asking whethe the difference you see is very large. As well, it may be important to verify whether the profile the lab gave you provides a current and valid protrayal of their machines' actual behaviour. You also need to be sure you are softproofing with "Simulate Paper White" selected, so you can be sure to be capturing the effect of the paper white on the overall appearance of the softproof.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

jjlphoto

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 467
Prints do not match monitor
« Reply #3 on: October 31, 2009, 11:01:09 am »

Quote from: MarkDS
Where you say you turned your display brightness "all the way down" - what specifically does that mean in CD/mm2? It could be that "all the way down" isn't down enough.

This can be quite true. On my friend's iMac, turning it all the way down still makes for a display that is 175cd/m2 using GMB i1 Pro and Match3. I calibrate my Eizo to 100cd/m2, I make my own custom profiles for my Epson 3800, and for outsource prints, use MPIX. Prints are a dead match to my monitor all the time.
Logged
Thanks, John Luke

Member-ASMP

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Prints do not match monitor
« Reply #4 on: October 31, 2009, 12:53:54 pm »

Quote from: Larry Berman
David Brooks wrote an article for Shutterbug on why prints tend to be too dark:
Prints Too Dark

Brooks hasn’t got a clue here (at least in terms of his original article on the subject). I’ve told him so but he’s in denial. He’d prefer to come up with some color management conspiracy theories rather than just use those tools properly.

The first issue is, do you have your display calibration luminance set to match the viewing conditions of the print next to the display? Prints too “dark”? Maybe but 99 times out of 100, its just the prints are darker than the display because either the viewing booth is too low (assuming it can be adjusted which many allow) OR the display calibration luminance is too high.

Some LCD displays provide little (not enough) provisions for dialing down luminance. jjlphoto’s post is spot on. That’s a problem. The alternative, other than getting a decent display system is to raise the viewing conditions of the print viewing booth. The idea is to produce a match at this location only. Moving the print outside this environment, using differing illuminants in terms of color and intensity isn’t an issue, you simply can’t run back into the digital darkroom and look at the display, hoping for a match. If you dial in the digital darkroom conditions to produce a match, I can assure you that short of viewing the prints under some really odd conditions, you’ll like what you see (you’ll adapt to the environment).

lastly, its super critical to have an output profile for the print conditions, one that’s not only being used properly for soft proofing (using full screen mode and the simulate check boxes for comparing print to screen) but also being used for output. A lot of labs use this silly, half baked “color management” workflow where they provide a profile for soft proofing but demand you send the document for output in sRGB. If you can’t use the profile for conversions, control the rendering intent, pre or post edit the data based on that exact conversion, you’re fooling yourself into getting the much desired screen to print match.
« Last Edit: October 31, 2009, 12:57:32 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Prints do not match monitor
« Reply #5 on: October 31, 2009, 01:44:32 pm »

Quote from: digitaldog
its super critical to have an output profile for the print conditions, one that’s not only being used properly for soft proofing (using full screen mode and the simulate check boxes for comparing print to screen) but also being used for output
I don't want to hijack the thread but may I ask about the simulate check boxes...?
If, as you do it, the monitor already is set to paper white referring to the viewing booth... why then use the simualtion of anything (or: anything else than "black ink") if it spins the white point of the monitor then? The monitor already matches paper white... and this is why it is - IMO - not useful when softproofing spins the white point. If you handle a wide range of papers that's different but photographic papers often contain optical brighteners and therefore the measured white in the profiles is blueish. If paper simulation turns the screen blueish (in the case of perceptual and relcol RI) or yellowish (in the case of absolute colormetric RI) there must be something wrong as the monitor initially displayed the correct (paper-) white.
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Prints do not match monitor
« Reply #6 on: October 31, 2009, 02:05:08 pm »

No - "the viewing conditions of the print" Andrew refers to is the illuminant hitting the paper, not the paper colour. You still need to select "Simulate Paper White" to get a reliable soft-proof.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Prints do not match monitor
« Reply #7 on: October 31, 2009, 02:41:02 pm »

Quote from: MarkDS
No - "the viewing conditions of the print" Andrew refers to is the illuminant hitting the paper, not the paper colour. You still need to select "Simulate Paper White" to get a reliable soft-proof.
here he referrs to the illuminant. If I remember correctly he also adjusts his white point visually to paper white in the viewing booth .. as I do.
And then... no, paper simulation doesn't help at all if the respective paper contains optical brighteners.
Logged

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Prints do not match monitor
« Reply #8 on: October 31, 2009, 05:11:07 pm »

The simulate check boxes alter the dynamic range of the preview to match that of the print (assuming the profile is correctly defining this). It has nothing to do with the illuminant nor color of the viewing booth.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Prints do not match monitor
« Reply #9 on: October 31, 2009, 05:29:24 pm »

Quote from: digitaldog
The simulate check boxes alter the dynamic range of the preview to match that of the print (assuming the profile is correctly defining this). It has nothing to do with the illuminant nor color of the viewing booth.
in this post I illustrated different softproof modes. Only the first 2 images match the luminance level (under respective viewing conditions), the paper white and the differentition of the actual print.
-> http://luminous-landscape.com/forum/index....st&p=316182

In this post I showed a selection of paper profiles set to "paper simulation" (mode rel.col RI + BPC) -> http://luminous-landscape.com/forum/index....st&p=316544
With all these papers the white point in "paper simulation" is incorrect (either rel.col or absolut col.) as well as the differentiation of tonal values due to too high compression (resulting from a white that is measured as a blue and therefore at the same time darker than white).
Actually with all photographic papers this problem is present... as long as they do not contain optical brighteners (such as e.g. some Canson Infinity papers).

As to the dynamic range it is reflected if the monitor is adjusted to the luminance level of the paper (under resp. viewing conditions) and if you check simulation of "black ink" only in softproof mode. "Black ink" boosts the black point of the display so that the dynamic range of the print is matched on the display.

Logged

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Prints do not match monitor
« Reply #10 on: October 31, 2009, 05:44:00 pm »

Quote from: tho_mas
With all these papers the white point in "paper simulation" is incorrect (either rel.col or absolut col.) as well as the differentiation of tonal values due to too high compression (resulting from a white that is measured as a blue and therefore at the same time darker than white).
Actually with all photographic papers this problem is present... as long as they do not contain optical brighteners (such as e.g. some Canson Infinity papers).

As to the dynamic range it is reflected if the monitor is adjusted to the luminance level of the paper (under resp. viewing conditions) and if you check simulation of "black ink" only in softproof mode. "Black ink" boosts the black point of the display so that the dynamic range of the print is matched on the display.

By its very nature, the paper simulation is an Absolute colorimetric intent with BPC off going to the display:

•Simulate Paper Color: Convert using the absolute colorimetric
intent (no Black Point Compensation).
•Simulate Paper Color and Simulate Black Ink Off: Convert using the
relative colorimetric intent with Black Point Compensation.
•Simulate Black Ink: Convert using the relative colorimetric intent
without Black Point Compensation.


Something “off” could be the profile itself of course, the quality of the AtoB and BtoA tables. It should be partially the display profile. OBA’s could affect what you’re actually seeing depending on the illuminant (Fluorescent lights can be problematic here). IOW, there are a number of things that could be “off” here.

In terms of the dynamic range and the display, unless you’ve got one of the rare “smart monitors” that provide full control over this (by adjusting black and white target calibration), you’re pretty much stuck altering this by using the simulate command (or sometimes called the Make my image look like crap buttons). If you’ve got a display that’s pumping out an 800:1 contrast ratio, the only way to show you what 300:1 looks like is to use the simulate options which isn’t as ideal as actually putting the display into a 300:1 ratio.
« Last Edit: October 31, 2009, 05:45:16 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Prints do not match monitor
« Reply #11 on: October 31, 2009, 06:27:06 pm »

Quote from: digitaldog
there are a number of things that could be “off” here.
of course. But at the first stage the "as white" measured value is off. Optical brighteners reflect the light of the measurement device blueish. And  the trouble starts from there.
In any case there is literally no photographic paper (profile) that produces an accurate paper simulation. At least I've seen none by now.

Quote
the only way to show you what 300:1 looks like is to use the simulate options which isn’t as ideal as actually putting the display into a 300:1 ratio.
I for myself am working with something around 340:1. However even on a cheap monitor I'd prefer to use "black ink" simulation (or no simulation at all) over "paper simulation".
Color managment works really accurate in most of the cases. But when it comes to the simulation of print media it sucks (often), sadly.
Logged

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Prints do not match monitor
« Reply #12 on: October 31, 2009, 07:09:44 pm »

Quote from: tho_mas
of course. But at the first stage the "as white" measured value is off. Optical brighteners reflect the light of the measurement device blueish.

Again, it depends on the Spectrophotometer used (does it have UV filtration or not), does the software compensate for this etc. OBA’s are by and large, not a good thing.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Prints do not match monitor
« Reply #13 on: October 31, 2009, 09:45:41 pm »

Quote from: tho_mas
But when it comes to the simulation of print media it sucks (often), sadly.
Actually, for me it doesn't. It's pretty good. And the paper I'm using most often has no OBAs (Baryta-base Ilford Gold Fibre Silk)
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Prints do not match monitor
« Reply #14 on: November 01, 2009, 05:40:40 am »

Quote from: digitaldog
Again, it depends on the Spectrophotometer used (does it have UV filtration or not)
the de facto standard is the i1pro I think... as for standarized profiles AFAIK the measurement device must be without UV cut filter.

Quote from: MarkDS
the paper I'm using most often has no OBAs
well, then yes, might be one of the rare cases where it is okay.

Still: if you adjust the white point and the luminance level of the display to match paper white in the viewing booth... then any color spin or darkening of the display in softproof mode set to simulate "paper" necessarily leads to a wrong representation. The display already matched paper white - so if the white representation changes it must be wrong. This is actually self evident. And this applies to the most photographic papers and as well to the standarized offset printer profiles (gracol, swop, ISOcoated ...).
So basically the recommendation should be: do not use paper simulation unless the actual print really matches the monitor preview better with paper simulation than without.
Why, as mentioned above, should I edit in fullscreen mode? All the whites on my monitor represent paper white so pallettes or menus doe not distract or hinder adaption.
« Last Edit: November 01, 2009, 05:44:33 am by tho_mas »
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Prints do not match monitor
« Reply #15 on: November 01, 2009, 07:00:33 am »

Quote from: digitaldog
By its very nature, the paper simulation is an Absolute colorimetric intent with BPC off going to the display:

•Simulate Paper Color: Convert using the absolute colorimetric
intent (no Black Point Compensation).
•Simulate Paper Color and Simulate Black Ink Off: Convert using the
relative colorimetric intent with Black Point Compensation.
•Simulate Black Ink: Convert using the relative colorimetric intent
without Black Point Compensation.


Something “off” could be the profile itself of course, the quality of the AtoB and BtoA tables. It should be partially the display profile. OBA’s could affect what you’re actually seeing depending on the illuminant (Fluorescent lights can be problematic here). IOW, there are a number of things that could be “off” here.

In terms of the dynamic range and the display, unless you’ve got one of the rare “smart monitors” that provide full control over this (by adjusting black and white target calibration), you’re pretty much stuck altering this by using the simulate command (or sometimes called the Make my image look like crap buttons). If you’ve got a display that’s pumping out an 800:1 contrast ratio, the only way to show you what 300:1 looks like is to use the simulate options which isn’t as ideal as actually putting the display into a 300:1 ratio.

Now I am becoming confused. I was under the impression that Black Point Compensation took care of the limited dynamic range of the print as compared to the monitor. The referenced white paper states: "Adobe Systems implemented BPC to address this conversion problem by adjusting for differences between the darkest level of black achievable on one device and the darkest level of black achievable on another [the print and the monitor]".

Black point compensation is usually used with the relative colorimetric rendering intent. It is not available with absolute colorimetric. With perceptual rendering, BPC should not be needed if one has a good profile, but is available for use with malformed profiles. If you have a good profile and use perceptual rendering with BPC is the black point mapped twice?

Also, if you are using relative colorimetric with BPC, what does simulate paper black do? Is it needed?
Logged

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799
Prints do not match monitor
« Reply #16 on: November 01, 2009, 07:14:13 am »

Quote from: bjanes
Now I am becoming confused.
yes, it's confusing, but it's correct.
You have to think laterally here (I always forget about that as well): the softproof simulates a certain output (paper) profile. Now this paper profile is displayed WITH BPC to the monitor if "simulation" is deselected. I.e. the paper contrast range is displayed within the contrast range of the monitor - not within the contrast range of the paper itself. So from paper to monitor there is a BPC.

Quote
Also, if you are using relative colorimetric with BPC, what does simulate paper black do? Is it needed?
With "black ink" selected the (paper-) image is displayed without BPC on the monitor... i.e. without stretching the paper-contrast ratio to that of the monitor.
With "black ink" selected you actually have a relative colormetric view of the paper on the monitor.
So in short: "black ink" is boosting the black point of the display to that of the paper. The result is that you are viewing the contrast range of the paper itself not that of the monitor.
I use "black ink" all the time, I never use "paper simulation" (see above).
« Last Edit: November 01, 2009, 07:15:01 am by tho_mas »
Logged

ChuckZ

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 59
    • http://czhike88.smugmug.com/
Prints do not match monitor
« Reply #17 on: November 01, 2009, 09:35:34 am »

Quote from: MarkDS
Where you say you turned your display brightness "all the way down" - what specifically does that mean in CD/mm2? It could be that "all the way down" isn't down enough. Were the prints from the lab VASTLY darker than what you see on the display. Recall, there is some inherent additional luminosity apparent from the display relative to a print simply because the former is transmitted light and the latter reflected light. One never totally overcomes this divide, so you need to make some mental adjustment for it - but in today's calibrated and managed environments, not a whole lot, why I'm asking whethe the difference you see is very large. As well, it may be important to verify whether the profile the lab gave you provides a current and valid protrayal of their machines' actual behaviour. You also need to be sure you are softproofing with "Simulate Paper White" selected, so you can be sure to be capturing the effect of the paper white on the overall appearance of the softproof.
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Prints do not match monitor
« Reply #18 on: November 01, 2009, 09:36:45 am »

Quote from: tho_mas
the de facto standard is the i1pro I think... as for standarized profiles AFAIK the measurement device must be without UV cut filter.

 well, then yes, might be one of the rare cases where it is okay.

Still: if you adjust the white point and the luminance level of the display to match paper white in the viewing booth... then any color spin or darkening of the display in softproof mode set to simulate "paper" necessarily leads to a wrong representation. The display already matched paper white - so if the white representation changes it must be wrong. This is actually self evident. And this applies to the most photographic papers and as well to the standarized offset printer profiles (gracol, swop, ISOcoated ...).
So basically the recommendation should be: do not use paper simulation unless the actual print really matches the monitor preview better with paper simulation than without.
Why, as mentioned above, should I edit in fullscreen mode? All the whites on my monitor represent paper white so pallettes or menus doe not distract or hinder adaption.

It may be less rare than you think.

And even with the display condition you propose, which may not be easy to achieve in practice - there is still the DR simulation you get activating <Simulate Paper White> which your procedure will not necessarily portray.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com
Prints do not match monitor
« Reply #19 on: November 01, 2009, 09:40:50 am »

Ref post #18 - Chuck - whatever you intended to say did not reproduce. Suggest you re-post.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."
Pages: [1] 2   Go Up