optimising render code

hi

i am trying to find the fastest way to render objects in a stack. After 
some tests i found that given a unidimensional array with 1000 numbers 
the following functions behave the oposite i would expect

def test1():
        for x in range(10000):
                for z, n in enumerate(stack):
                        y = n*999 # some calculation

def test2():
        for x in range(10000):
                z = 0 # opengl Z render position
                for n in dol(stack):
                        z+=1 # from botton to top
                        y = n*999 # some calculation

Basically the 2nd seems to perform far faster than the first, unless i 
am doing something wrong. I have been checking out some articles about 
optimising performance in python etc... But i am a bit confused so far.

I was wondering if there is some good but clear game where i could check 
techniques to improve this.

thanks!

--
altern