#+TITLE: Normalized Range
In =sndkit=, a =normalized range= is any real number within
the inclusive range 0 and 1.

Normalized ranges are ideal ways to abstract away units when
you don't want to think about them.

Usually, values in a normalized range are destined to be
@!(ref "scale" "rescaled later")!@, such as @!(ref "oscf"
"table lookup oscillators")!@.

In IEEE floating point, there are sometimes some numerical
advantages to keeping things in a smaller range, as values
in this range have more equidistance spacing compared to
using relatively larger values. This spacing is referred
to as the magnitude of a
[[https://en.wikipedia.org/wiki/Unit_in_the_last_place][unit in last place, or "ulp"]].
More information about the nuances of floating point
precision can be found in =Physically Based Rendering= by
Pharr, Jakob, and Humphreys in Chapter 3.9 (Managing
Rounding Error).