[part 5, final] So here we have a big difference between an optimizing compiler and what an assembler program would do.
When I'm writing in assembler, I try to write the code so that it will still work when changed (within limits). But what the compiler generated was the code for this exact situation: packed fields of this exact size, with this exact edit mask. It wouldn't work for anything else, and would need to be radically different for a more complicated edit mask. My point is, the compiler can do a better job of efficient code than I can, because I'm trying to write structured and maintainable code, but in assembler.
