Ask HN: Y ALU as floating point method of handling the decimal point not devised

Why is CPU/ALU with a floating point method of handling the decimal point position not devised/implemented ? all simply just treated as integer with a byte or few bits is required to manage or keep tracking the decimal point position, let it be B e.g. 0.2 x 0.7 = 0.14 it’s simply 2 x 7 = 14 . B = 0 987 x 0.07 =69.09 it’s simply 987 x 7 =6909 . B =2 IMHO this CPU/ALU method of microprogramming just far simpler