The IEEE floating point number precisions supported.
32 bit IEEE floating point number format
64 bit IEEE floating point number format