so what would be the best utilization for speed with C embedded in Perl?

Roughly - any time you have to perform expensive operations on data of a known type. Perl's data types (scalars, arrays, hashes) are, at the C level, data structures with a copious amount of flags and magic stuff dangling; using embedded C you avoid shuffling around that in the algorithm you want to implement. Input and output still have to be perl variables.

_($_=" "x(1<<5)."?\n".q·/)Oo.  G°\        /
                              /\_¯/(q    /
----------------------------  \__(m.====·.(_("always off the crowd"))."·
");sub _{s./.($e="'Itrs `mnsgdq Gdbj O`qkdq")=~y/"-y/#-z/;$e.e && print}