[tom_pinckney]

Worth checking out pysco as well http://psyco.sourceforge.net/

Though in practice I find that if performance really matters using numpy or writing a C extension module usually works out the best.

[pmiller2]

I like psyco simply because all you have to do is import the module and do `psyco.full()` to get an idea how well it's going to work with your code. Once I feel like I've gotten the most I can possibly get out of algorithmic and data structure optimization, psyco is almost always my next stop for further speed. Often, importing and activating psyco is all the optimization I end up needing.

[ominous_prime]

I would like to note though, that psyco development has been stopped for a couple years, and probably won't continue (unless a new developer thinks there's a future and takes it over). Further development is going into pypy.

Also, it can't run on 64-bit machines. If you want to run it on OS/X 10.6, you need to build your own 32 bit python.