?

Log in

No account? Create an account
Rainbows Are Bad - The Mad Schemes of Dr. Tectonic [entries|archive|friends|userinfo]
Beemer

[ userinfo | livejournal userinfo ]
[ archive | journal archive ]

Rainbows Are Bad [Oct. 7th, 2011|01:10 pm]
Beemer
Rainbows are bad...

...at least when it comes to visualizing data.

This paper explains why. (And for all the non-scientists, it's got neat pictures and some cool stuff about how vision works.)

Don't use spectral colormaps. They are the devil.
LinkReply

Comments:
[User Picture]From: detailbear
2011-10-08 02:42 am (UTC)
So interesting. My use of colour at work (non-engineering) is all segmentation and highlighting. We're soon™ about to start visual representations of financial and productivity data. I need to summarize this for the VP-Programming.

Thanks!
(Reply) (Thread)
[User Picture]From: danthered
2011-10-08 04:46 am (UTC)
Thenk ya. I'm going to write a précis of this and put it, with a link in this week's Driving Vision News. Spectral colourmaps are frequently used to present beam isoplots. We're probably doing it wrong.
(Reply) (Thread)
[User Picture]From: thedragonweaver
2011-10-09 01:17 am (UTC)
Cross-field comment—one of the interesting artifacts of the pace of television technology is the use of black & white monitors in the studio. This allows the television crew to look at a luminance-based version of what they're filming, because some people's televisions, well, really suck.

The added value of this is that a picture that looks good or at least reasonably clear in black & white is far superior in terms of color picture quality to one that does not have a lot of variance. And it's good for people who have variants of color-blindness. And... this is why they tell you to not wear tan on television unless your skin is very dark.
(Reply) (Thread)