cEd wrote: > Hello, > > I'm wondering how to compile python to get good performance. > Because when I compare this ugly code which find prime number: > > # prime_number.py > start=30000 > for is_first in range(start): > found = 0 > is_first+=1 > for i in range (2, is_first): > if not (is_first%(i)): > found = 1 > break > if not found: > print is_first > > > between : > - the python distributed with my ubuntu > #time python prime_number.py > /dev/null > real 0m12.237s > user 0m12.129s > sys 0m0.024s > > - and the one compiled by my self > time my_python_compiled prime_number.py > /dev/null > real 0m42.193s > user 0m41.891s > sys 0m0.044s > > so which option should I give or which flag ???
I doubt that there is such a flag. There must be a different reason for this. Can you give us the python versions for each, and architecture (32/64 bit)? Diez -- http://mail.python.org/mailman/listinfo/python-list