Lol, yes I'm aware of the three packages currently featured on BT4. ALL of these employ statiscal analysis to break WEP keys (WPA cracking even with rainbow tables is essentialy brute-forcing from a large wordlist), so the time taken is wholly dependant upon the processing power* available to the 'penetration tester'.
What I was alluding to was the fact that encryption algorithm for some common routers is known. As such there is no need to perform repeated calculations to determine the secret key, just one is required provided the default SSID has not been changed.
*other factors are involved in the real world such as: distance to target, PPS Injection rate, and MAC address filtering that hinder the IV collection procedure before one can even start breaking the key.. Of course with WPA only the handshake is required, then it becomes a simple mater of checking permutations against the wordlist. Of course in a real world test you have no idea what the passphrase is so a very large wordlist is needed.