Pages: [1] 2   Go Down

Author Topic: Images printing too warm in spite of color management - color space problem?  (Read 10216 times)

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284

I have an image that has a a green color tone, but when I print it it is much more into yellow - I need to apply a strong orange filter to have the image look the same on screen as on paper.

Computer: MacBook Pro Retina 16gb

Display: NEC PA271W, calibrated with i1Display pro using NEC's Spectraview software.

Soft proofing in Photoshop.

Color Space: ProPhoto (is this the problem? Should I convert to sRGB before soft proofing and printing?)

Canned profiles used for both Ilford Galerie Smooth Gloss and Hahnemuehle Fine Art Baryta High Gloss.

Color are the same on both papers, brightness is OK (meaning the images are as bright as they appear in softproofing).


Is my having the images in ProPhoto RGB at print time the culprit?
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

Schewe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6229
    • http:www.schewephoto.com

Is my having the images in ProPhoto RGB at print time the culprit?

No...

So, does the image soft proofed look like the print?
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com

What did the images start as? Raw? TIFF? 8 bit, 16 bit? What do they look like in Camera Raw or in Lightroom compared with their appearance in Photoshop before softproofing?
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284

I am printing from a TIFF in ProPhoto RGB and chose high quality and 16bit. (I always edit 16bit in Photoshop).

The images looked the same at all stages.


I am looking at my newest, optimized print and it comes much closer to what I see on screen. I can detect green taking more the lead - even though I had not adjusted the colors.

What I have adjusted is contrast in the dark areas and increased sharpening.



I wonder if the perception of color is affected by contrast and sharpness.

After optimizing sharpening and increasing contrast in the darker areas I got richer blacks the image looked my closer to what I saw on screen.

Maybe we (or is it just me?) see low contrast and immediately have the perception of a warmer color temperature because the blacks are not as deep?

Particularly the blacks... it's as if they were priming the viewer's eyes when looking at the pictures. If they are muddy through lack of contrast and sharpening, the viewer has the impression of a warmer color temperature.

What do you think about this?
« Last Edit: April 26, 2013, 08:33:43 pm by The View »
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com


Maybe we (or is it just me?) see low contrast and immediately have the perception of a warmer color temperature?

Particularly the blacks... it's as if they were priming the viewer's eyes when looking at the pictures. If they are muddy through lack of contrast and sharpening, the viewer has the impression of a warmer color temperature.

What do you think about this?

It doesn't correspond with my visual perception. I don't see shifts in colour balance when I alter luminosity - but I do all this mostly in Lightroom to raw files. Increasing contrast "strengthens" colours and makes them look somewhat more saturated, but it should not actually alter colour balance to the extent that something looking cool on a display looks warm in a print. It tells me that something is amiss with profiling, and I suspect the display profile, because two separate printer profiles from two different vendors are producing the same impact on the print. If they are "wrong", at least they are consistently wrong, so I suspect they are OK. There could be something amiss in the chain between the display, the video card in your laptop and the monitor profiling. When you calibrated the display using that software in your laptop, were you able to do it via DDC, or was it a manual process? What white point did you select?
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284

It doesn't correspond with my visual perception. I don't see shifts in colour balance when I alter luminosity - but I do all this mostly in Lightroom to raw files. Increasing contrast "strengthens" colours and makes them look somewhat more saturated, but it should not actually alter colour balance to the extent that something looking cool on a display looks warm in a print. It tells me that something is amiss with profiling, and I suspect the display profile, because two separate printer profiles from two different vendors are producing the same impact on the print. If they are "wrong", at least they are consistently wrong, so I suspect they are OK. There could be something amiss in the chain between the display, the video card in your laptop and the monitor profiling. When you calibrated the display using that software in your laptop, were you able to do it via DDC, or was it a manual process? What white point did you select?

I calibrated to D65.

What do you mean by DDC? It sounds like a standard that enables communication between profiling tools and the hardware.

But for what I know, Spectraview II does hardware calibration.

There wasn't anything manual in it - I just clicked "calibrate" (after choosing the basic settings like D65 for white point, gamma 2.2, intensity 140 cd/2, monitor default contrast, and native color gamut (full) ).
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com

DDC  is the direct data channel allowing hardware calibration. So if you have that working, it's good. I think your problem is the D65. It's too cool relative how you probably view the prints. Your print viewing environment, even if Solux, would likely be no more than about 4800~5000K. Try re-calibrating the display at a D50 of D55 white point and see what happens. It will take you a bit of time to get comfortable with it. The warmer display will cohere better with usual indoor viewing conditions and incite you to make adjustments that somewhat cool what goes to the printer, hence under normal viewing conditions it will probably all end-up cohering. Give it a shot and see what happens.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284


The white point is a good tip - I'll do some tests and see where it gets me.

Thanks, Mark.
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

Schewe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6229
    • http:www.schewephoto.com

Try re-calibrating the display at a D50 of D55 white point and see what happens.

Yeah, ya know...that's a rabbit hole (you are welcome to go down it if you want–I wouldn't).

If you are using ACR/Photoshop/Lightroom, it really doesn't matter how you calibrate your display because the profile is used to correct white point and gamma for all your image displays in any color managed apps.

If your print is warmer than you image and soft proof, something else is going on...
Logged

Ernst Dinkla

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4005

The display, your viewing light, both require an adaption to one another in color temperature and luminosity. What suits does not have to have the same numbers in color temperature etc but the combination of both properties has to fit, I would take that 140 cd in the display down to 110 for instance when you go to 5000-5500K. The papers used are not that warm, slightly cooler than neutral actually and that paper white should be present in the softproof too if done properly.
What kind of viewing light are you using and what is the ambient light level?

--
Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
December 2012, 500+ inkjet media white spectral plots.
Logged

Mark D Segal

  • Contributor
  • Sr. Member
  • *
  • Offline Offline
  • Posts: 12512
    • http://www.markdsegal.com

Yeah, ya know...that's a rabbit hole (you are welcome to go down it if you want–I wouldn't).

If you are using ACR/Photoshop/Lightroom, it really doesn't matter how you calibrate your display because the profile is used to correct white point and gamma for all your image displays in any color managed apps.

If your print is warmer than you image and soft proof, something else is going on...

Any idea what that "something else" may be given what we've been told?

As for the rabbit hole - I don't agree. D65, relative to D50 for example, produces a cooler rendition of anything, softproofed or not. My own experience working between the printer and display indicates that this should be one of the first things to experiment with when confronted with this issue. Not a big deal. The OP will try it and see whether it floats his boat. Whether this is the exact problem or not I suspect at the end of the day the outcome will be a display profiling issue.
Logged
Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....."

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799

Soft proofing in Photoshop.
with or without "paper white" simulation? I assume you've checked "paper simulation" since this is the general recommendation...
These glossy papers contain quite a large amount of optical brighteners. In effect the actual white of the papers is represented pretty blueish in the paper profiles (the Hahnemuehle profile for Epson 11880 shows a Lab value of 97/2/-5 for white). Therefore softproofing with "paper simulation" enabled the image will always look much too blue on screen.... see attachment. (the complementary color of blue is orange ... so it comes to no surpirse that the image looks closer to the actual print when you apply a strong orange filter. Unfortunately applying color filters is not the desired workflow...)
As noted above, calibrate your monitor to around 5500K. For softproofing (with these papers) uncheck "paper simulation" but select "simulate black ink".
« Last Edit: April 27, 2013, 10:47:27 am by tho_mas »
Logged

Ernst Dinkla

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4005

with or without "paper white" simulation? I assume you've checked "paper simulation" since this is the general recommendation...
These glossy papers contain quite a large amount of optical brighteners. In effect the actual white of the papers is represented pretty blueish in the paper profiles (the Hahnemuehle profile for Epson 11880 shows a Lab value of 97/2/-5 for white). Therefore softproofing with "paper simulation" enabled the image will always look much too blue on screen.... see attachment. (the complementary color of blue is orange ... so it comes to no surpirse that the image looks closer to the actual print when you apply a strong orange filter. Unfortunately applying color filters is not the desired workflow...)
As noted above, calibrate your monitor to around 5500K. For softproofing (with these papers) uncheck "paper simulation" but select "simulate black ink".


Hence my question about the used viewing light but on teh other hand most of the time (custom) profiles are made with UV-cut spectrometers so the paper white softproof will not look blue. I think this HM profile is more the exception than the rule. I measured b -4 for the Baryta and -4.5 for the Ilford, cooler than neutral but there are papers out there with b -11. That HM profile could be the fly in the soup this time. Shows again that OBAs and color management do not go well together.

--
Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
December 2012, 500+ inkjet media white spectral plots.
Logged

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799

I think this HM profile is more the exception than the rule.
I'm not sure. In any case UV cut filters introduce other issues since we rarely view images under UV cut conditions ;-)
Printing with papers containing OBAs I feel much better to calibrate my monitor to match paper white in my viewing booth (both white point and brightness) and switch off paper white simulation when sofproofing (I do use the simulation of "black ink", though). Prinitng with papers that do not contain OBAs paper white simulation may or may not work well. Depends on the actual profile (and paper, of course) ...

Logged

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284


If you are using ACR/Photoshop/Lightroom, it really doesn't matter how you calibrate your display because the profile is used to correct white point and gamma for all your image displays in any color managed apps.

If your print is warmer than you image and soft proof, something else is going on...

Reading your post, adjusting the color temperature of the display would not help - my eyes would accommodate (and so would the soft proofing view)?

Then it must be the viewing conditions. I don't have a daylight balanced viewing light, and it was late at night after a whole day at the computer: warm light bulbs and tired eyes?

The prints look much better in daylight, but I'm not completely sure if I'm getting what I was working towards when I was editing the images.

The other moment is probably the difference of seeing an illuminated picture, like a "color slide", on the monitor, versus a picture that comes from reflected light on a paper surface.

I haven't done a lot of printing yet, so I probably have to develop the feeling how a picture translates from screen to paper.

I'll do an intentionally cooler print and see how I feel about that.
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284

Any idea what that "something else" may be given what we've been told?

As for the rabbit hole - I don't agree. D65, relative to D50 for example, produces a cooler rendition of anything, softproofed or not. My own experience working between the printer and display indicates that this should be one of the first things to experiment with when confronted with this issue. Not a big deal. The OP will try it and see whether it floats his boat. Whether this is the exact problem or not I suspect at the end of the day the outcome will be a display profiling issue.

Yes, I will create a 5750 K setting that Digital Dog is using (he says, he's only using this setting to adjust for the viewing conditions in his booth, though).

It's not a big deal using Spectraview II. One can switch the settings from the menu.

One probably has to try out all the things: viewing conditions, white point, and getting a feeling how your image style transfers from screen to paper.

I think I have to get more experience, also by playing with the above variables.
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284

with or without "paper white" simulation? I assume you've checked "paper simulation" since this is the general recommendation...
These glossy papers contain quite a large amount of optical brighteners. In effect the actual white of the papers is represented pretty blueish in the paper profiles (the Hahnemuehle profile for Epson 11880 shows a Lab value of 97/2/-5 for white). Therefore softproofing with "paper simulation" enabled the image will always look much too blue on screen.... see attachment. (the complementary color of blue is orange ... so it comes to no surpirse that the image looks closer to the actual print when you apply a strong orange filter. Unfortunately applying color filters is not the desired workflow...)
As noted above, calibrate your monitor to around 5500K. For softproofing (with these papers) uncheck "paper simulation" but select "simulate black ink".


Yes, I had both "simulate paper color" and "simulate black ink" checked.

I noticed that "simulate paper color" makes images looks cooler - more than the original edit. With that checked off I could adjust for the warming of the paper.

I'd like to put this out to everybody: do you check or not check "simulate paper color"?
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/

Reading your post, adjusting the color temperature of the display would not help - my eyes would accommodate (and so would the soft proofing view)?

To some degree yes, but if you're viewing a print and the display (giving a slight amount of time to adjust moving your head), you want a visual match. That means you may very well be calibrating to match a soft proof differently for one paper than another. That's when having a smart display that allows you to switch on-the-fly to account for contrast ratio and perhaps issues with paper/viewing conditions is so darn useful. Keep trying to tweak the calibration to produce a visual match since a visual mismatch isn't useful <g>.

You want the simulate boxes ON.

http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

The View

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1284

To some degree yes, but if you're viewing a print and the display (giving a slight amount of time to adjust moving your head), you want a visual match. That means you may very well be calibrating to match a soft proof differently for one paper than another. That's when having a smart display that allows you to switch on-the-fly to account for contrast ratio and perhaps issues with paper/viewing conditions is so darn useful. Keep trying to tweak the calibration to produce a visual match since a visual mismatch isn't useful <g>.

You want the simulate boxes ON.

http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml

Of course! As soon as my eyes go off the screen and to the print, that adjustment of the eyes is gone.

I'll check the 5750 K setting and see how it compares.

I will definitely need a color corrected light bulb to check my prints. Those regular light bulbs are way too warm.

Is there a color corrected bulb that works with a regular desk lamp? (So far I only found halogen bulbs) I'll check for one when I go to Samy's today.
Logged
The View of deserts, forests, mountains. Not the TV show that I have never watched.

tho_mas

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1799

check or not check "simulate paper color"?
trust your eyes!
make a bold white border aorund your image and enable softproof with paper white simulation checked. Now look at the paper. Does your Hahnemühle really looks that blueish? Now turn off paper white simulation. Which mode looks better (better = more like the white of the actual paper)?
« Last Edit: April 27, 2013, 01:47:59 pm by tho_mas »
Logged
Pages: [1] 2   Go Up