A computer approximation to a real number.
Any fraction number used in a computer, (with the exception of a double floating point number). It is defined with 32-bits, but can be defined with less depending on the device doing the calculation. Some video cards use 16-bit floating point numbers, usually set up with a sign bit, ten bits for the mantissa (or base number), and five bits for exponent.
A decimal number. You set up an floating point number using float
Often referred to in mathematical terms as a "rational" number, this is just a number that can have a fractional part. See "Double Precision" and "Single Precision."
A number whose decimal point can be in any position, as opposed to a Fixed point decimal which is always in the same position. Fixed points : 2.00, 2.12 Floating points : 0.023, 00.23, 002.3