floating-point

Non-integer number format

Floating-point describes a system for representing numbers in computing where the decimal point can move to accommodate very large or very small values.

Unlike fixed-point representation, floating-point allows the same number of digits to express vastly different magnitudes: floating-point arithmetic, floating-point precision.

The term reflects how the decimal "floats" to different positions depending on the value being stored.

errors

overflow

Exceed capacity and spill

  • underflowValue too small computed
  • NaNUnrepresentable numerical value
  • infinityUnlimited extent or quantity

exception · denormalization · rounding error · inexact

representation

mantissa

Logarithm's fractional component

fraction · encoding · format · standard · specification · radix

precision

accuracy

Degree of correctness

relative error

notation

scientific notation

Numbers times ten-powers

magnitude · scale

arithmetic

addition

Joining quantities mathematically

coprocessor · ALU · operation · calculation · comparison · commutativity · hardware · vectorization · instruction

Categories
Opposites
numeric representation type

floating-pointfixed-point

Element mapped to itself

integer · whole number · int · rational · decimal

exactness of representation

floating-pointexact

Perfectly correct, no deviation

precise · integer · symbolic · lossless · discrete

Family

float, floating, pointer, afloat, floated, floats, floater, point, punctuation

Familiarity
Top 500,000
Profile
Used worldwide · Timeless · Abstract · Rated G