in reply to Procedural Perl slower than OO Perl!!!
For whatever it may be worth, I still have a (prized...) copy of the First Edition (the FORTRAN edition) of Kernighan & Plauger’s The Elements of Programming Style, in which the following sage maxim first appeared: Don’t “diddle” code to make it faster: find a better algorithm.
And so, I very-earnestly offer the following opinion: “if whatever-you-are-doing seems to be ‘human-visibly slow,’ find a better algorithm.”
In other words, don’t be concerned about what language it is. (Believe it or not, it is in fact a perfectly safe assumption that all language implementors do know what they are doing.) If you find yourself saying ... “language-X is slower than language-Y”, it probably actually has nothing to do with either of the two languages, and everything to do with your choice of algorithm.
I mean... let’s face it ... when whatever-we-are-doing finally makes its way down to the hardware level (as, in fact, everything-we-are-doing eventually does do...), everything happens as “bezillions of operations-per-second.” Therefore, if somehow we still see “a human-noticeable difference in speed,” it very-simply is not realistic to attribute this difference to some discrepancy in technical language-compiler implementations . . .