- From: Phil Green <green@colourspace.demon.co.uk>
- Date: Mon, 2 Apr 2018 22:04:13 +0100
- To: public-colorweb@w3.org
- Message-ID: <4b34769c-07d8-79f1-52b1-e5b993ec81ea@colourspace.demon.co.uk>
Hi Chris This discrepancy has been known for some time, and was discussed in a paper I did at Electronic Imaging in about 2004. The values in the ICC.1 specification [96.42, 100, 82.49] were those for the 1931 observer and D50 illuminant as originally published in CIE Publication 15. Unfortunately there was an error in the Z value, which was corrected in CIE 15:2004 to 82.51. These values and the precision of two significant places remain the same in the new version currently in press. ICC has discussed this on a number of occasions and it has been decided that it would not be appropriate to change the D50 PCS illuminant value for ICC.1, since this would require a change in CMMs, extensive modification of installed code, and potential interoperability problems with existing profiles. ICC.2, however, supports the use of the corrected values as a custom PCS. Note that values [0.9642,...] arise since by convention ICC encodes CIE XYZ values normalised so that Y=1.0 rather than 100. Phil Green ICC Technical Secretary On 02/04/2018 21:19, Chris Lilley wrote: > > Hi folks, > > It was recently pointed out[0] that the XYZ values for the D50 > illuminant given on, for example, Matlab[1] or Bruce Lindbloom's site[2] > > |[0.96422, 1.00000, 0.82521]| > > differ from the once specified by the ICC > > |[0.9642, 1.0000, 0.8249]| > > || > > "In ICC v4, the requirement was introduced that the media white point > of a Display class profile shall be equal to D50 (i.e. [96.42, 100, > 82.49])" [3]. > > I assume the authoritative source is the CIE. I will check my books > this evening, but can anyone shed light on the discrepancy (and the > correct value)? > > [0] https://github.com/w3c/csswg-drafts/issues/2492#issuecomment-377913660 > [1] > https://www.mathworks.com/help/images/ref/whitepoint.html?s_tid=gn_loc_drop > [2] http://www.brucelindbloom.com/index.html?Eqn_ChromAdapt.html > [3] http://www.color.org/whyd50.xalter > > -- > Chris Lilley > @svgeesus > Technical Director @ W3C > W3C Strategy Team, Core Web Design > W3C Architecture & Technology Team, Core Web & Media
Received on Monday, 2 April 2018 21:05:31 UTC