Well use a binary floating point format with a sign bit, 4 digit exponent, and 8 bit significant. These fields are in most to least significant bit order. Lets figure out the Hexadecimal encoding for Pi, which is 11.0010010000111111... in binary.