Code:
$ python
Python 2.5.1 (r251:54863, Oct 5 2007, 13:50:07)
[GCC 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> 2**32
4294967296
^ bytes of memory addressable with a 32bit cpu. Now if we try that with 64bits
Code:
>>> 2**64
18446744073709551616L
So in more convenient units: 4gb vs 16exabytes
It also means that for floating point calculations you have double the number of bytes for a floating point number representation, which is rather handy for stuff that uses a lot of floating point (few desktop apps do).
Also, because memory addresses are now 64bits, 64bit binaries are larger in filesize, because every memory address is twice the length.
Is 64bit faster? no.