That first question is answered in the thread:

> Yes, when using either 32-bit or 64-bit number, the lowest portion can output a zero, as I suggested on the attached example file as a modification to fix the problem. But a true zero (fitting the requested size), on AMD, never happens.

Another person (on page 2) confirms those results on an older AMD processor (but failed to reproduce on a very new one, 9950X3D).