Floating point encodes an approximation of proper base 2 arithmetic. Or do you really believe that 0.1 + 0.2 isn't equal to 0.3 for some reason? Click to expand... It encodes (i) a base two mantissa; ...