> >On 10 March 2015 at 00:56, < varnays...@mail.ru > wrote: >> >>I used a nytprof to check performance, function call reduces it. >Its not just "reduces it" that matters. Its whether reducing it in the grand >scale of things is really important. ( vs other parts of the system / general >usecases ) > >If for instance the total time saved over a week amounts to 1 second, its >probably not worth the several hours spent optimising it. > >I'd conjecture you might already have proven that, but with profiling, that's >important enough a concern that we should ask that question =) > > >-- >Kent > >KENTNL - https://metacpan.org/author/KENTNL
Execution time of inline variant approximately 17.6ms, subroutine solution slower about 26%. My script is captcha generator. Random sign used very often during execution. I think 26% is a lot enough in this case. K ind regards , Art Varnayskiy .
