On Thu, Sep 20, 2012 at 10:01:28PM -0700, Dan Dennedy wrote:
> But then we need to map [0, 15000], which only makes the bottom 6.6%
> of the [0,1] range useless.

I'm not worried about the “missing” range. You could argue exactly the same
for 0..15000 as 0..1; floats work that way. 1 K is _more_ than enough
precision for this anyway.

> OK?

Sounds good. I've been picky about having full sentences in all comments
(capitalized first letter, period after) in the file, but I realize this is
far from the typical frei0r style, so I guess it won't hurt having two
comments in there without =)

/* Steinar */
--
Homepage: http://www.sesse.net/