======================PROTOCOLS_FLOATING_POINT============================ IEEE-754 single-precision float format, a 32-bit number is descobed by a 23-bit mantissa, m; an 8-bit exponent, e; and a single sign bit, s. The value of the number is given by the formula (tgnoring the special cases described in the standard). Since fixed-point fonnats have no exponent, they can represent numbers with more significant bits than floating-point fonnats of the same overall size. In this typical example [bottom), there are 31 significant bits, f, compared with 23 bits for the IEEE-754 floating-point format of the top figure. Note that, althaugh this jormLtt can represent the number „1, it cannot represent + 1 exactly. 31 30-23 22-0 x =((-1)^s)*( 2^(e-127))*(1 +0.m) _________________________ |s| e | m | |_|_______|_______________| 31 30-0 x =( (-1)^s )*( 0.f ) _________________________ |s| f | |_|_______________________|