Pages: [1]   Go Down

Author Topic: A Question About Gamut Examination in Color Think  (Read 2652 times)

mhmacdougall

  • Newbie
  • *
  • Offline Offline
  • Posts: 3
A Question About Gamut Examination in Color Think
« on: April 25, 2015, 04:08:20 pm »

In preparing to print an image with a very light and uniform sky, I found it difficult to make soft proof adjustments which would bring the proofed sky close to the original. Neither Photoshop’s soft proof gamut warning nor Print Setting gamut warning showed the sky as out of gamut. However, when I graphed a sample of the sky color together with the paper profile in Color Think, that color seems considerably out of gamut (as shown below).

I’m missing something here, and hoping that someone will point out the error of my ways …

Thanks,
Mac

Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: A Question About Gamut Examination in Color Think
« Reply #1 on: April 25, 2015, 04:57:46 pm »

The OOG overlay in both Photoshop and Lightroom are kind of buggy, not accurate. Ignore them. As for soft proofing, you might want to use ColorThink and plot the display profile against the image, it's possible the soft proof isn't as effective as it could be due to the limitations of the display gamut. Or it could be an issue in the proofing part of the ICC profile that controls the soft proof.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

mhmacdougall

  • Newbie
  • *
  • Offline Offline
  • Posts: 3
Re: A Question About Gamut Examination in Color Think
« Reply #2 on: April 25, 2015, 06:36:46 pm »

Thanks, Andrew -- that was interesting! I've been coasting along assuming that AdobeRGB was more or less encompassed by my NEC PA242 monitor, but not so! Doing the graph you suggested shows a number of image points falling in the gap between monitor space and working space.

Lessons learned so far: never assume anything; forget PS OOG, when gamut's a concern, check image vs. profile in Color Think.

Thanks again,
Mac
Logged

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1852
    • Frank Disilvestro
Re: A Question About Gamut Examination in Color Think
« Reply #3 on: April 27, 2015, 07:05:48 pm »

While I agree that PS gamut waring is not accurate (to the mathematically obsessed), I think there must be something wrong in your settings. I created a document with AdobeRGB color space and filled with the color referenced in your graph (162,199,255). When I check for gamut warning against Epson Ultra Premium Lustre profile, it shows clearly as out of gamut. I even tried with an older version of PS (CS3) and it does the same.

Note that small changes in saturation / lightness may place the sample inside the color space.

If you could share a crop of the image and the profile maybe we can check if there is something wrong.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: A Question About Gamut Examination in Color Think
« Reply #4 on: April 27, 2015, 07:10:50 pm »

While I agree that PS gamut waring is not accurate (to the mathematically obsessed), I think there must be something wrong in your settings.
I'm not referring to anything mathematical, I'm quite mathematically challenged. :-\ I'm referring to taking as an example, an sRGB document, soft proofing sRGB and asking to see the OOG overlay. There should be none. But often there is. That shouldn't happen.

I also don't believe treating something 1% OOG and 90% using the same ugly overlay that blocks the underlying image is at all useful for editing images.

An OOG overlay could be darn useful if implemented in a more robust fashion, like we see in ColorThink Pro. There you can apply three colors and you can decide what range they represent. Adobe could do the same, have a way to apply differing opacity of the overlay etc.
« Last Edit: April 27, 2015, 07:12:28 pm by digitaldog »
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1852
    • Frank Disilvestro
Re: A Question About Gamut Examination in Color Think
« Reply #5 on: April 27, 2015, 07:29:49 pm »

I'm not referring to anything mathematical, I'm quite mathematically challenged. :-\ I'm referring to taking as an example, an sRGB document, soft proofing sRGB and asking to see the OOG overlay. There should be none. But often there is. That shouldn't happen.

I also don't believe treating something 1% OOG and 90% using the same ugly overlay that blocks the underlying image is at all useful for editing images.

An OOG overlay could be darn useful if implemented in a more robust fashion, like we see in ColorThink Pro. There you can apply three colors and you can decide what range they represent. Adobe could do the same, have a way to apply differing opacity of the overlay etc.

Couldn't agree more! Sorry if my comment about being mathematically obsessed (which sometimes I am) offended anybody , I forgot to put the smiley  :).

Having said that, The color referenced by the OP should have been shown as OOG by Photoshop (regardless of how lousy this is implemented in PS)

mhmacdougall

  • Newbie
  • *
  • Offline Offline
  • Posts: 3
Re: A Question About Gamut Examination in Color Think
« Reply #6 on: April 27, 2015, 08:58:14 pm »

Thanks for your interest! The tests I described above were done is the same way as yours. I created a small Adobe RGB document and filled it with RGB = 162/199/255. In Customize Proof Conditions, I set Device to Simulate to the profile " Epson Stylus Pro 3880_3885_3890 PremiumLuster", Rendering Intent to Perceptual, checked Black Point Compensation, and checked Simulate Paper Color. If Gamut Warning in the View Menu then is turned on, no gamut warning appears. If I attempt to print that document using that profile and click on Gamut Warning in the Printer Settings menu, no gamut warning appears.

The profile in question is the standard Epson-supplied profile for Premium Luster paper on the 3880: the header fields as displayed by Color Think are shown in the attachment. If this is the same profile you are using, perhaps I should try replacing my current 3880 profile package.  Any suggestions you have would be greatly appreciated.

Thanks,
Mac
Logged

fdisilvestro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1852
    • Frank Disilvestro
Re: A Question About Gamut Examination in Color Think
« Reply #7 on: April 28, 2015, 09:49:18 am »

Hi,

I checked again with the correct profile for the 3880 since I had one for the 2880. Now I have the same behaviour as yours, where the gamut warning does not show OOG. I checked the same file with Ligthroom soft-proof and again it does not show any OOG with that color.

I found a possible explanation of how the Gamut warning works here (It is in Italian). Apparently it calculates how distant a color is from the gamut border and it shows the warning above a treshold which is not disclosed by PS. The author used a deltaE=13 to get the same results than PS but with a very limited set of patches, so I would think that the actual threshold is less than that. According to this discussion in the Colorsync user mailing list, the threshold is between 5 - 8 (deltaE)

Conclusion: That specific color is out of gamut of the paper profile used but within the threshold of PS' gamut warning, which might be unacceptable for critical work.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: A Question About Gamut Examination in Color Think
« Reply #8 on: April 28, 2015, 10:01:19 am »

Conclusion: That specific color is out of gamut of the paper profile used but within the threshold of PS' gamut warning, which might be unacceptable for critical work.
IOW, the OOG overlay in both Photoshop and Lightroom are kind of buggy, not accurate. Ignore them.  ;D
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".
Pages: [1]   Go Up