Agreed. The article author claims that the "don't waste an element" code is also more efficient, but that claim seems to be based on a hard-on about the post-i

Agreed. The article author claims that the "don't waste an element" code is also more efficient, but that claim seems to be based on a hard-on about the post-increment operator, rather than any kind of dive into the cyclometric complexity, or even, y'know, just looking at the assembler output from the compiler.