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