Re: In search of a D50 Editing colorspace
Re: In search of a D50 Editing colorspace
- Subject: Re: In search of a D50 Editing colorspace
- From: Peter Miles <email@hidden>
- Date: Wed, 03 Sep 2014 20:02:46 -0700
I am not talking about comparing two displays.
I'm talking about comparing two real-world objects together. An object and a print of that object, viewed together in a GTI viewing booth.
One example of the what I am talking about happened to involve comparing a real colour checker chart and a printed copy of that colour checker chart. Viewing them both together in a GTI booth.
The situation I find myself in that if I were to enter D50 as my viewing condition for the Colorchecker in spectra shop (which I assume isn't _wildly_ different to my GTI booth), the the neutral 8 patch RGB values I get from spectrashop for AdobeRGB is R 211, G 200, B 174.
This is correct, assuming the D65 of AdobeRGB were to be actually used in the adobe's Absolute Colorimetric print pipeline. But as we now know, it isn't used. So if I were to actually print R 211, G 200, B 174, the print and the real world colour checker chart would NOT match in my GTI viewing booth.
I's not until I 'LIE' in spectrshop and say that the chart is being illuminated by D65 I get neutral 8 RGB as R201 G202 B200. And I actually get a print that is beginning to resembling the actual chart.
From the previous discussion, this situation is understandable given that Adobe is substituting a relative colorimetric rendering from the RGB space when the user requests an absolute rendering.
As far as I can see, the problems caused by Adobe doing this, to a large extent become mute if the source RGB space is D50.... and I'm printing into a CMYK colorspace.
Regards
Peter
________________________________
From: Lars Borg <email@hidden>
To: ColorSync <email@hidden>
Sent: Thursday, 4 September 2014, 13:52
Subject: Re: In search of a D50 Editing colorspace
Peter,
Note that ICC's Absolute colorimetric is illuminant-relative colorimetry.
So a display's max R=G=B maps to a second display's max R=G=B even if the
color temps of the displays differ, so won't look the same.
This is different from CIE absolute colorimetric, which is truly absolute
(same XYZs), and cannot reasonably be reproduced in print as paper doesn't
have a built-in illuminant. :-)
CIE absolute colorimetric is often used for cinema projectors.
Absolute colorimetric is useful if you're printing something on two
different paper stocks.
If done right, the prints should match under the same (D50 calibration)
light source.
They need not match under D65, though.
Absolute colorimetric isn't adding any value for displays, as they are
their own light source, so illuminant-relative colorimetry becomes
identical to media-relative colorimetry, i.e. relative colorimetric.
You can also use Lab as color space. Beware that Lab is also based on the
D50 illuminant.
Lars
_______________________________________________
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