Why are mov ah,bh and mov al, bl together much faster than single instruction mov ax, bx?

Question

I\'ve found that

mov al, bl
mov ah, bh

is much faster than

mov ax, bx

Can anyone explain me why? I\'m run

Answer 1

In 32-bit code, mov ax, bx needs an operand-size prefix, whereas byte-sized moves don't. Apparently modern processor designers do not spend much effort at getting the operand-size prefix to decode quickly, though it surprises me that the penalty would be enough to do two byte-sized moves instead.