Pages: 1 ... 4 5 [6]   Go Down

Author Topic: I tried a Custom Profile  (Read 10180 times)

howardm

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1984
Re: I tried a Custom Profile
« Reply #100 on: February 21, 2019, 11:07:04 am »

there are sorta 'magic' values of # of patches for i1Profiler that yields an increased # of grey-ish patches.  I believe they are
all perfect cube numbers (12^3 = 1728). 

I then may optimize w/ several hundred near-greys (+/-2 a* or b*) for every L* slice (so something like 9 colors of near-grey for every L* slice)

The difference is small but definitely there, particularly in the clarity of the lower quarter tones.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20646
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: I tried a Custom Profile
« Reply #101 on: February 21, 2019, 11:51:25 am »

there are sorta 'magic' values of # of patches for i1Profiler that yields an increased # of grey-ish patches.  I believe they are
all perfect cube numbers (12^3 = 1728). 

I then may optimize w/ several hundred near-greys (+/-2 a* or b*) for every L* slice (so something like 9 colors of near-grey for every L* slice)

The difference is small but definitely there, particularly in the clarity of the lower quarter tones.
I use a target in i1P with a 5x5 grid (-2,-1,0,1,2 - a* and b*) at each L*, for a total of 2500 patches. Not a burden to measure on an Isis XL.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bwana

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 309
Re: I tried a Custom Profile
« Reply #102 on: February 21, 2019, 12:33:21 pm »

there are sorta 'magic' values of # of patches for i1Profiler that yields an increased # of grey-ish patches.  I believe they are
all perfect cube numbers (12^3 = 1728). 

I then may optimize w/ several hundred near-greys (+/-2 a* or b*) for every L* slice (so something like 9 colors of near-grey for every L* slice)

The difference is small but definitely there, particularly in the clarity of the lower quarter tones.

Ok I get it. Printers use color inks to make grey tones so a profiled printer for moonochrome allows you to eliminate all casts.

@howardm  have you documented your process anywhere? As I am a novice, I understand that you are talking with Lab values, but do not know the specific steps to create such a patch set w iprofiler.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20646
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: I tried a Custom Profile
« Reply #103 on: February 21, 2019, 12:50:44 pm »

Ok I get it. Printers use color inks to make grey tones so a profiled printer for moonochrome allows you to eliminate all casts.
It isn't a monochrome profile. It's a color profile that has good gray balance. A monochrome ICC profile converts color to monochrome. As outlined here by the ICC:
http://www.color.org/faqs.xalter

Q. Can I make a profile for converting colour to grey scale?

A. ICC defines a monochrome profile format  though that doesn?t mean that all profile making software supports it. Since profiles produce colour transformations by combining an input profile that goes to a standard colour space (PCS), and an output profile that goes from this space, an output profile that converts colour to greyscale would simply define the relationship between the L* channel and the device levels.

The creation of gray profiles is not a feature that many profiling applications have, but you can make one quite easily by custom settings in Photoshop's Color Settings dialog. When you save the settings an ICC profile is generated.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

howardm

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1984
Re: I tried a Custom Profile
« Reply #104 on: February 21, 2019, 01:17:30 pm »

Bwana,

I cheat :D

short answer/process:  I got a used iSis and created a 1728 patchset on a  8x19 piece of paper, then I went here and generated a large .cxf file that
would fit on 5x19 paper (I forget offhand).

http://www.russellcottrell.com/photo/optimization/cxfGeneratorLab.htm

Then I merged the 2 patchsets in PS so it all fits neatly on a 13x19.  Print the 13x19 and you now have 2 charts (requires cutting).  Generate the
profile w/ the 1728 and then optimize w/ the other (used as i1Profiler spot colors).

I also generated a somewhat smaller set so I can get everything on 3 USLetter sheets if need be.

I was somewhat surprised at the printed difference between the optimization chart from the PRO-100 (dye) and my 3880.
The 3880 was quite neutral whereas the -100 has this significant overall magenta tone (haven't noted the actual Lab #s for it though)
but once the profile is made and applied, the magenta pretty much is eliminated (as it should be).

All this is MUCH easier w/ an iSis unit (I'd never sit around long enough to scan all that manually).
Pages: 1 ... 4 5 [6]   Go Up