Re: On ProPhotoRGB
Re: On ProPhotoRGB
- Subject: Re: On ProPhotoRGB
- From: Marco Ugolini <email@hidden>
- Date: Sun, 26 Feb 2006 19:28:47 -0800
In a message dated 2/26/06 2:17 PM, Karsten Krüger wrote:
> Did you do the same with ECIrgb and LStar-RGB ? Both are recomended
> workingspaces for digital photography when targetting print.
>
> [snip]
>
> They both have the same primaries and gammut, but differ internally. ECIrgb is
> has a gamma of 1.8 while LStar-RGB is linear to L*-axis. What is your opinion
> on them?
Hi Karsten.
>From what I can tell -- and confirming what you are saying -- both ECIRGB
and LStarRGB do indeed share the very same exact coordinates both for the
primaries (all 3 situated within the spectrum locus) and the white point,
but are not as large as ProPhotoRGB by a long shot.
ProPhotoRGB encompasses them completely, whereas DonRGB misses out a tiny
peak in the red-oranges.
Predictably enough, when viewed in ColorShop, both ECIRGB and LStarRGB
display exactly the same 6 hue line shapes. But the 10 steps in the 2D
cobweb view appear visibly more-evenly distributed in LStarRGB compared to
ECIRGB, due to the difference you mentioned (ECIRGB adopts a 1.8 gamma
curve, whereas LStarRGB uses a 256-point curve instead).
> They are supposed to match ISO cmyk gammut without too much loss.
I may be mistaken, but I am inclined to believe that most of these concerns
about working spaces being too large compared to the output on press are
misplaced, even if we do not use 16 bits and stay in 8 bits. The way I see
it (and I am willing to be proven wrong), if one works exclusively for
commercial print output, working spaces like AdobeRGB or ECIRGB-LStarRGB (or
probably PhotoGamutRGB as well) will be good choices, and will encompass all
the colors on press. That statement seems to hold true for the Europe ISO
Coated FOGRA27 profile as well. None of this is anything that sRGB can be
claimed to be able to do (no matter how much "certain people in certain
forums" -- harrumph! -- poo-poo such statements).
Whether or not there will be "too much loss" does *not* depend, in my
opinion, on how close the shape of the RGB working space is to that of your
CMYK commercial print output. As Bruce Fraser frequently points out, the
important thing is that the working space (source) maintain distinctions
among the colors in the file, without clipping them. If it does, then these
color differences will be mapped as distinct colors in the print output
space (destination) through the careful use of a suitable rendering intent
(sometimes RelCol will do, other times Perceptual will be better).
If instead one works for inkjet output (like many fine-art photographers
do), it is more prudent to work with larger-gamut working spaces, like
ProPhotoRGB or DonRGB (though, as you know, I'm trying to figure out a
couple of things about ProPhotoRGB at the moment). I say that because inkjet
printing technologies are constantly evolving, and we may regret having
thrown away colors that may become printable sooner than we think, or may
already be printable now on existing devices that we simply haven't heard of
yet. On the other hand, if none of that is of concern to fine-art
photographers (a hard-to-believe concept for me), then AdobeRGB or sRGB will
do (though that would certainly not be either my choice or advice).
Best regards.
--------------
Marco Ugolini
Mill Valley, CA
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden