This is well documented: https://www.agner.org/optimize/blog/read.php?i=49#49, https://www.agner.org/optimize/blog/read.php?i=49#112.
It goes very far back to MMX: https://yro.slashdot.org/comments.pl?sid=155593&cid=13042922
tldr: Intel's compiler doesn't optimize using standardized instructions on non-Intel hardware.