Formulae for conversion between NCS and RGB (CMY, etc.)

[Chris Nellist]

I understand that the Dulux paint colour codes are based on the NCS colour system. I'd like to convert between this and the RGB/CMY systems, so as to be able, for example, to identify the Dulux colour most closely matching a given 3-byte hex code (e.g. in CMY, primary magenta as 00FF00).

I have been unable to find the mathematical specification of the NCS.

I have read the material at the NCS website and would be grateful if someone could tell me whether

a) the definitions of the elementary hues R, Y, etc. in NCS are the same as in CMY and RGB?

b) the gradation between two neighbouring elementary colours, e.g. R and Y, is linear (in CMY terms)?

c) a colour such as S 2030-Y90R can be made by subtractively mixing - 20% black (255/255/255 in CMY) - 30% of the hue defined by Y90R - 50% white (0/0/0 in CMY)?

If the answer to each of these questions is yes, conversion is simple, but I suspect there is more to it than this. I know that some NCS codes do not exist, i.e. some spaces in triangles in the NCS colour atlas for given hues are left blank. I am not sure whether this is because the colours are imaginary or just because the required pigments do not exist.

If there *is* more to it than this, could someone post sufficient information to allow conversion between NCS and CMY (and therefore RGB), given standard illumination conditions? If the answer to c) above is 'yes', all I need is a formula for converting between the two codes for a given hue.

Obviously a formula for converting between NCS and any colour system from which I can get to RGB will be just as good! :)

Failing this, does anyone know the RGB codes for the elementary colours (red, green, blue, yellow) used in NCS? Or the NCS codes for the primary colours in RGB or CMY? I have looked on loads of colour conversion sites, but haven't been able to find even this information, anywhere!

Many thanks in advance for help with this.

Yours sincerely,

Chris

[G&M]

I think they also apply some scrambling. I've tried to decode the numbers on the samples without success. Lawrence LB once went to a place that would take a colour, scan it in and match it exactly but I can't find anybody to offer this service.

[Danny Rich]

"Chris Nellist" wrote in message news:Xns958075D8DD40Bpnellist@195.92.193.157...

This is because there is no mathematical specification - the NCS system is a perceptual space.

No - RGB & CMY are device space coordinates and RYGB are perceptual, unique hues.

No they are not since RGB/CMY are not linear in perceptual color space.

The NCS is not a substractive space - it is a perceptual space in which blackness and whiteness are attributes and not phytsical quantities,

These spaces are left blank because these perceptual colors cannot be reliably created in the paint medium used to create the example pages.

The Scandinavian Color Institute offers a program that can generate standard RGB values for a given input notation. They have a program which you can license that contains a combination Look-Up Table and analytical model that is based on instrumental readings of the example chips.

This is because the conversion between NCS and any other color order system is considered commercially proprietary. Some things in life are unfortunately, not free.

Danny Rich

[G&M]

Provided it is not for commercial gain, in the EU it is legal to make public such information. For instance the yanks tried to prosecute the Norwegian guy who posted how to decode DVD codings and were laughed out of court.

[Chris Nellist]

"G&M" wrote in news:ckpaf1$uq1$ [e-mail]:

I think it is legal to do it even for commercial gain, e.g. to test a patch of NCS colour and sell someone the info as to what its RGB code is.

The program EasyRGB gives NCS codes for colours that harmonise with an input colour of specific RGB code, but unfortunately doesn't give the NCS code for the input colour itself!!

It would be a good thing if even just the RGB codes for the NCS elementary colours were publicly known. (Hint hint to anyone reading this who has got the information!)

Chris