Path: utzoo!mnetor!uunet!husc6!bloom-beacon!mit-eddie!mit-amt!turk
From: turk@mit-amt.MEDIA.MIT.EDU (Matthew Turk)
Newsgroups: comp.graphics
Subject: Re: Naive Question About Primary Colors
Message-ID: <2192@mit-amt.MEDIA.MIT.EDU>
Date: 28 Mar 88 21:54:04 GMT
References: <7871@oberon.USC.EDU> <7779@apple.Apple.Com> <310@bdrc.UUCP>
Distribution: na
Organization: MIT Media Lab, Cambridge, MA
Lines: 35
Keywords: primary colors
Summary: "Well, almost", once again

In article <310@bdrc.UUCP>, jcl@bdrc.COM (John C. Lusth) writes:
> 
> A small point.  Given primary colors A, B, and C, and target color X,
> it is not necessarily true that some combination of the primaries will
> match color X.
> 
>        A + B + C matches X                  *** not necessarily true ***
>        
> However, it is true that for any X that some combination of X and one
> of the primaries can be matched be some combination of the other
> two primaries.
> 
>        A + B matches X + C                   *** always true ***
> 

Actually, neither are *always* true, but it is usually true that
one of them will be true!  Think of the primaries A, B, and C
as corners of a triangle.  Some perceivable colors are inside
the triangle and some are outside.  Any mixture of A, B, and C
will fall somewhere inside the triangle.  A mixture of any two
will fall on the line between the two (one of the triangle sides).

So if X is inside the color triangle, then the first statement
is true.  If X is outside the triangle the second statement is
sometimes true, and the rest of the time yet a third statement
will be true:  A matches X + B + C (when that is necessary is
left as a proof to the reader!).  Remember that adding A to
one side is conceptually the same as subtracting (B + C) from
the other.

Of course all this discussion is ignoring the fine print in
the "optional" section of the text book -- the part that says
it's not really true for all colors....

	Matthew