Abstract: A format conversion system for use in digital systems converts digital data words from a fixed point format to a floating point format and vice versa. A two's complement circuit tests the sign bit of a first fixed point digital data word and provides the complement if the word is negative. A code establishing circuit receives the first fixed point digital data word or its complement, detects the number of O's to the left of the most significant one bit and establishes a first code representative thereof. A second code establishing circuit receives the exponent of a first floating point digital data word and provides a second code representative thereof. A mantissa/fixed point determining circuit receives either the first fixed point digital data word, or the mantissa in reverse order, of the first floating point digital data word.