Yep, you guessed it Connors, this is for an encryption program.
I'm attempting to write an algorithm which will use a short password to generate a sequence of bytes which is as long as the file being encrypted. Each byte in the generated sequence will encrypt a byte in the plaintext. If the encrypting sequence is unique, non-repetitive, and seemingly random, then the ciphertext will be very hard to crack, at least in theory. A derivative sequence isn't as good as a truly random sequence, but it's much better than my previous method (Vigenere substitution).
I want to test the sequences to determine if they eventually repeat themselves. I also want to compare sequences to determine if different sequences can converge into a common sequence. A convergence would probably be a large chunk in both files, and it would have to continue to the end of the files. The longer the common sequence is, the more likely it's a true convergence.