

Extendedprecision floating point numbers have the following format on the Intel386 microprocessor:
79  78  64  63  0 

SIGN  EXPONENT  FRACTION  
^  
binary point 
Field  Position  Full name 

sign  79  sign bit (0==positive, 1==negative) 
exponent  7864  exponent (biased by 16,383) 
fraction  630  fraction (bit 63=1, followed by the 
decimal point. Bits 620 are the fraction. 