CHROMiX

CGATS formatting for illumination

According to the .pdf documentation, ColorThinkPro appears to allow use of customised illuminants, providing that the file is in “CGATS text file format”. I can create CGATS files for reflected data (IT8.7/2 format), but cannot create an illuminant file because I do not know what the format should be, and cannot track down any info on the internet. (I have an Eye-one Pro Spectrophotometer, so extracting the spectral data for my light source from a “.cxf” file is no problem.)
Does anyone have an example illuminant file which shows the correct formatting?

Simon

Ok, so I have found a part of the answer…
Gretag_Macbeth use some files for “Light sources” (ProfileMaker5 Measure Tool uses these) which look very like an IT8.7 target file.
The data is spectral data 380nm to 730nm (as measured by Eye-One Pro). These files can be found (PM4 or PM5 users - Windows OS) in C:\Program Files\GretagMacbeth\ProfileMaker Professional 5.0.5\Measurement Files\Light Sources.
PM5 accepts multiple “sets”, but ColorThinkPro (v3.0) will only read the first set (it appears to ignore the others, which is fine for me).

So I have constructed single illuminant files for the light sources that I needed by:
1 Measuring the light source using Eye-One share and saving the data as a “.cxf” file
2 Using Excel to open the “.cxf” file and copying the spectral intensity values
3 Pasting (using “paste_special/transpose”) into a text file (in Excel).
4 Saving the file from Excel as text (".txt")
5 Opening the text file with Notepad (text editor)
6 Copy/paste into the “template” text file with a similar format to the GretagMacbeth “Light Source” files.

The reason that I dont do the whole thing in Excel is because Excel (not very helpfully) when reading the text file strips out the Quotation ( " ")marks which appear to be an integral part of the CGATS spec.

“MEASUREMENT_SOURCE…”, “SAMPLEID…” and “1 10D_Flash…” should be all on a single line (HTML will put phantom line breaks in when you read this).

A workable file format seems to be as follows:
LGOROWLENGTH 1
CREATED “02/03/2006” # Time: 18:10
INSTRUMENTATION “Eye-One Pro”
MEASUREMENT_SOURCE “Illumination=Unknown ObserverAngle=Unknown WhiteBase=Abs Filter=Unknown”
ILLUMINATION_NAME “Emission”
KEYWORD “SAMPLEID”
KEYWORD “SAMPLE_NAME”
NUMBER_OF_FIELDS 38
BEGIN_DATA_FORMAT
SampleID SAMPLE_NAME nm380 nm390 nm400 nm410 nm420 nm430 nm440 nm450 nm460 nm470 nm480 nm490 nm500 nm510 nm520 nm530 nm540 nm550 nm560 nm570 nm580 nm590 nm600 nm610 nm620 nm630 nm640 nm650 nm660 nm670 nm680 nm690 nm700 nm710 nm720 nm730
END_DATA_FORMAT
NUMBER_OF_SETS 1
BEGIN_DATA
1 10D_Flash 0.000929975 -0.0235814 0.147943 0.421343 0.536534 0.604945 0.660189 0.684064 0.766526 0.750329 0.751998 0.729285 0.6534 0.593494 0.617933 0.636181 0.634857 0.608826 0.586445 0.544186 0.567587 0.591566 0.538842 0.555021 0.504815 0.48403 0.442343 0.546393 0.547209 0.431345 0.384908 0.462377 0.465281 0.580814 0.328768 0.351967
END_DATA

What I don’t yet know - ("…part of the answer…") is what to do about scaling the values to avoid arithmetic errors. My data above corresponds to a light level (reported by Eye-one Share) of 43lux. Gretag Macbeth files appear to be 2,500 to 7,000 lux.

I also found that the Argyll site (LCMS) referred to, but did not define, a CGATS (style) format text file.
CGATS17-2005 (ANSI spec) hints at illumination capability (in the context of a user defined field) but does not define this particular format either in text or XML.

You are right about the file format. ColorThink Pro will use the “logo” file format that you have seen in ProfileMaker.

The “logo” format is the one used by GretagMacbeth in ProfileMaker, Eye-One Match, etc. It is very similar to the CGATS format but apparently not in full compliance. That is why you see a translator between logo format and CGATS format in the Argyll toolset.

I am working on support for .cxf files as well.

Also, when it comes to scaling the values you don’t need to worry. Spectral files for illuminants are “relative” values. So ColorThink Pro scales them all down to 0.0-1.0 based on the highest value in the curve.

regards,

Steve

Thanks Steve for the update.

Gretag Macbeth file format (.cxf) support would be very welcome when you get time for it…as extracting by the copy/paste and text editing from the .cxf file is just a bit clunky.

Simon