IBM 7044/7090/7094 (36-bit Words)

36-bit words were used early in IBM's mainframe history.

Single Precision (one 36-bit word) floating point:

 IBM 7090/7094 floating point single precision: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Sign Bit Exponent (bias: 12810) Fraction: bit 10 is the most significant

Double Precision (two 36-bit words) floating point:
 Double Precision (72 bits): First 36-bit Word: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Sign Bit Exponent (bias: 102410) Fraction (First, most significant part): bit 13 is the most significant

 Double Precision (72 bits): Second 36-bit Word: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Fraction (Second, less significant part): bit 1 is the most significant bit in this word

EXPONENT (E): To the base 2, with a bias in the single precision floating point of 128 decimal (200 octal), and in the double precision floating point of 1024 decimal (2000 octal).

FRACTION: Has no hidden bit.

NEGATIVES: No complement.

The value of the number N is N = F x 2E-128 (Single precision), or N = F x 2E-1024 (double precision) where F is a binary fraction such that 0 <= |F| <1

INTEGERS: 2's complement.-34,359,738,368 to +34,359,738,367