I was unhappy with the current implementation of the color schemes. This provides a way for a program to dynamicly scale the number of different colors on plots, and it ensures there won't be any color overlap.
Ryan, it sounds like you are going to be disappointed in this 'maximum' color separation too.
Open the file up, and you will see that this one is not 'maximum' for small numbers of line colors. Depending on how you quantify color separation, even large numbers of line colors could be argued to be non-maximum in some metrics.
I suspect CMAPLINE provides results more in line with your desires.
Your function is nice. I have been trying to do something similar, in order to create a maximum number of distinct line colors to use as the axes ColorOrder. I haven't been able to figure out an algorithm which produces maximum color variation, and at the same time convince myself that the algorithm really gives the 'maximum' color variation.
So, my question is, how did you come up with the RGB curves, and if/how do you know that it is really optimal in terms of maximizing color variation?
03 Oct 2008
21 Aug 2008
Yes, thank you. I saw those and wasn't happy. The pick of the week description does an excellent job of pointing out some of the utility with this file.