Closed NilesDavis closed 2 years ago
Thank you for this detective work.
The underlying principle was to take the spectral power distribution in the smallest resolution available from official CIE documents. This boils down to the state of affairs you mention above. The exact sources are given in https://luox.app/about#illuminant-data.
I hope this clarifies things.
Okay this is the way you went. So to be consistent I would recommend to use the 1nm dataset from ISO/CIE 11664-2:2020 for D50 as well. Thanks for the quick exchange!
At the time of writing, this document was not available unfortunately. Could you potentially create a pull request with this change? Thanks.
Thank you, much appreciated @NilesDavis.
I found that in the dataset for standard illuminants (reference_spectra.json) that 2 datasets are taken from other sources. So C, D50, D55 and D75 are taken from the CIE 015:2018 as they are in 5nm resolution as given in the standard. Standard illuminants A and D65 are in 1nm and exactly match the dataset from the ISO/CIE 11664-2:2020. What is the reason for this? Shouldn't the 5nm CIE15 data be the reference for all standard illuminants?