Pages: [1]   Go Down

Author Topic: gamut calculation from custom color list possible?  (Read 3851 times)

smilem

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 297
    • Color Management Services
gamut calculation from custom color list possible?
« on: August 17, 2013, 09:44:16 am »

Hello, is it possible to calculate gamut size from a custom measured color list in colorthink
or any other tool?

Would be very handy for tracking and evaluation of lightfastness of prints. And for monitoring gamuts of various devices without the need to build any profiles or even have a software package to do so.
Logged

xpatUSA

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 390
    • Blog
Re: gamut calculation from custom color list possible?
« Reply #1 on: August 17, 2013, 11:50:11 am »

Would it be possible to define what you mean by "gamut size"?

Apparently there are several usages, see Andrew Rodney's post here:

http://www.luminous-landscape.com/forum/index.php?topic=57625.msg466022#msg466022

I use ColorThink V2.3 and mostly view the relative "size" of image gamuts and color-spaces in the 3D Lab view. For me, "size" of a gamut includes volume and shape. Simply boiling it down to single number, e.g. % [area] of NTSC, is an over-simplification. Some may disagree with that, for example those accustomed to viewing the 1931 CIE xyY diagram to compare color gamuts.
« Last Edit: August 20, 2013, 12:51:00 am by xpatUSA »
Logged
best regards,

Ted

smilem

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 297
    • Color Management Services
Re: gamut calculation from custom color list possible?
« Reply #2 on: August 17, 2013, 12:58:08 pm »

Would it be possible to define what you mean by "gamut size"?

gamut size as in colorthink, the single number used to judge the amount of colors a device is capable printing. I understand that this is in direct relationship with the amount of colors and color shades I will use so the goal is comparison as long as same patch chart is used.

I understand that i will be not comparable to colorthink but it should be able to go in the same direction, if I compare 2 devices and the second has lesser gamut then this analysis should also show that dev2 is less then dev1.
Logged

xpatUSA

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 390
    • Blog
Re: gamut calculation from custom color list possible?
« Reply #3 on: August 17, 2013, 09:54:17 pm »

gamut size as in colorthink, the single number used to judge the amount of colors a device is capable printing. I understand that this is in direct relationship with the amount of colors and color shades I will use so the goal is comparison as long as same patch chart is used.
The Color List in ColorThink V2.3 on my computer appears to do nothing. The stupid manual is for the Mac and is written in double dutch, although it does show a single number in the list header bar. Is that the one you're using to determine gamut? In the context of the Color List, I have no idea what that number, for example, "875" colors is supposed to represent. That is to say, I don't know how the infinite quantity of different colors in any color space gets quantized in ColorThink down to a single number and how significant that number is with reference to printing.

However, as I don't print, I can be of little help in what you're trying to do. My personal feeling is that the use of a single number to compare gamuts, no matter how that number was determined, could be quite misleading.
« Last Edit: August 18, 2013, 11:04:12 pm by xpatUSA »
Logged
best regards,

Ted

MarkM

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 428
    • Alaska Photographer Mark Meyer
Re: gamut calculation from custom color list possible?
« Reply #4 on: August 20, 2013, 08:22:03 pm »

The answer is no in almost all cases. The number reported as gamut size is the volume of a three-dimensional solid. The problem you are describing is like trying to calculate how much an invisible container can hold given the coordinates of a few points that are inside or on the edge of the container. You might be able to do this if you had a lot of points that were carefully picked (this is how profiles are made) or knew they were a random sampling inside and outside the gamut (see this post for how that might work). Also, if you knew something about the shape of the container and knew where the points fell relative to the container. For example, you could calculate the volume if you knew the container was a cube and you had the coordinates of two vertices.

But with just a handful of colors and no more information, you can't really determine the volume.
Logged
Pages: [1]   Go Up