It's not faster, if you run the code, you'll notice that the time varies a lot between one and the other, sometimes one is faster and sometimes the other, and endswith is written in C for performance.

In the first case you are comparing a character to a string. In the second case you are creating a completely new string only to compare it while this solution just iterates over the characters (which is of course faster than creating a new string).

It's really quite simple, actually : look at the fork.

Edit : similar solution (except that I used 128 instead of 1000)