Diverging colormap design

[rkern]

Kenneth Moreland (of CoolWarm fame) gives an algorithm for designing perceptual diverging colormaps that should be straightforward to implement here. He uses CIELAB as the core perceptual colorspace, so it would be interesting to see it reimplemented with CAM02-UCS instead.

[njsmith]

FTR I think that @kthyng may also be working on extending the tool so it can be used to easily design diverging colormaps.

[kthyng]

Yes, still working on that (did some more today). My confusion with Moreland's coolwarm colormap is how narrow the range of lightness is. That property seems to wipe out a lot of detail that can be seen in maps over larger lightness ranges, but I suspect he chose it for a particular reason.

[njsmith]

For coolwarm, IIRC having a narrow range of lightness was an explicit
design goal, since they wanted a colormap that would be suitable for
overlaying on 3d-rendered objects, where you want to mostly reserve the
lightness channel for lighting/reflectance/shadow information that gives
you cues to shape. Different tools for different purposes, I guess...

On Wed, Jul 15, 2015 at 4:43 PM, Kristen Thyng notifications@github.com
wrote:

Yes, still working on that (did some more today). My confusion with
Moreland's coolwarm colormap is how narrow the range of lightness is. That
property seems to wipe out a lot of detail that can be seen in maps over
larger lightness ranges, but I suspect he chose it for a particular reason.

—
Reply to this email directly or view it on GitHub
#4 (comment).

Nathaniel J. Smith -- http://vorpus.org

[rkern]

Or the application of hill shading, to bring it back to a 2D application.

[kthyng]

Ok, got it.

[njsmith]

Current master now has support for designing diverging colormaps.