RE: The color of gray
RE: The color of gray
- Subject: RE: The color of gray
- From: Roger Breton <email@hidden>
- Date: Sun, 31 Aug 2014 08:30:07 -0400
Peter,
In my opinion, your understanding of "RGB Gray" being made of equal 'equal RGB pixel values' is correct.
This is the case for all "synthetic" or "reference" RGB spaces such as ProPhotoRGB, AdobeRGB, sRGB, ColormatchRGB, eciRGB, LstarRGB, BetaRGB and so on.
As you noted, the concept carries through a calibrated RGB monitor. In principles, given a reliable instrument, equal calibrated RGB pixel values ought to appear gray to a Standard Observer. That's the promise of 1931 colorimetry. As far as the underlying color management system is concerned, "gray" maps out to a* and b* being equal 0 in Lab space.
If your instrument says your Eizo white is achieving a white close to 5000K or D50 then, in principles, it ought to evoke the appearance of 5000K or D50 white in ProPhotoRGB or eciRGB or BetaRGB or ColormatchRGB or any other D50-based RGB space. That's my take on the subject.
As for AdobeRGB being viewed on your 5000K or D50 calibrated monitor being incorrect, well, strictly speaking, in my opinion, this is the case.
Two things. Ideally, one would want to test the following :
- take two identical monitor models;
- calibrate one to D50 and the other to D65;
- create a new ProPhotoRGB document in in Photoshop;
- fill the document with RGB = 128,128,128 on both monitors;
- on the D50 monitor, we know that there will be no chromatic adaptation involved;
- on the D65 monitor, we know that colors will be chromatically adapted to D50;
- will the two monitor images appear 'similar'?
That's the $10,000 question.
Personally, if I was to base my own color workflow on AdobeRGB, I would calibrate my monitor to D65 in a heartbeat. It Is clear that, given a D65 system, no chromatic adaptation is involved whatsoever, 'equal RGB pixels values' ought to appear gray.
Last but not least, will AdobeRGB 1998 and ProPhotoRGB gray to print gray with different color with absolute colorimetric rendering? You don't have to be near your printer to answer that question. A quick visit to Photoshop reveals that RGB 128,128,128 maps out to L54 with adobeRGB and L61 with ProPhotoRGB, because of the difference in gamma (2.2 vs 1.8). But both map out to a* and b* = 0. This difference in Lab will carry to your printer, regardless of the rendering intent but both will appear the same 'gray', just with different Lightness value. But one will not appear warm gray while the other appear cool gray, if that's your question.
Best / Roger
-----Original Message-----
From: colorsync-users-bounces+graxx=email@hidden [mailto:colorsync-users-bounces+graxx=email@hidden] On Behalf Of Peter Miles
Sent: 31 août 2014 03:26
To: ColorSync
Subject: The color of gray
Hi List members.
I'm in the process of doing some spring cleaning. I'm pulling out some my old ideas that around color that I haven't examined for a while and asking myself if they are really correct. I'd appreciate it if you would read the following and kindly point out any misunderstandings / half-understandings on my part.
The Color of Gray
I am a bit suspicious of my current ideas around the concept 'gray' after asking myself if the following was really true...
"That in an editing RGB color spaces, such as AdobeRGB 1998 and ProPhoto RGB, that 'equal RGB pixel values equal gray'.
QUESTION 1
My current understanding is that this statement is true. But they are gray only relative to the white point of the colorspace. Because AdobeRGB 1998 is D65 and ProPhotoRGB is D50 the color of ProPhotoRGB 'gray' is different to the color from Adobe RGB 1998 'gray'.
I attempted to test this (expecting to fail) by making up some gray ramps in photoshop in both ProPhotoRGB and AdobeRGB. I wasn't too surprised to see no color difference between the grays of ProPhotoRGB and AdobeRGB 1998. My understanding for this failure to see a difference is that the 'whites' of both colorspaces are being scaled relatively colorimetrically to my monitors white. Is this right?
QUESTION 2:
So am I correct in the following
My Eizo monitor being set to 5000K will show me the color of the gray in ProPhotoRGB (D50 white) more or less correctly, in an absolute sense. And when my Eizo monitor is set to 5000K, the monitor color displayed for my AdobeRGB 1998 grays is incorrect. In an absolute sense. In that the grays will be displayed warmer than they should be given that Adobe RGB white is D65.
QUESTION 3:
I'm not near my printer to test this just now, but I'd expect AdobeRGB 1998 and ProPhotoRGB gray to print gray with different color with absolute colorimetric rendering?
Thanks for your time. Regards
Peter Miles
_______________________________________________
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
_______________________________________________
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