Yes, there is a valid distinction to be drawn; you would need a cryptologist to tell you how meaningful the difference is.
A "real salt" as you've described is used to "perturb the encryption algorithm". I vaguely understand that, but not well enough to describe properly. Suffice it to say that with a real salt, the original password text is enciphered but with different outputs depending on how the algorithm incorporates the salt input (the algorithm has (at least two) inputs, the salt and (separately) the original password text).
A "fake salt" involves modifying the original password text with the salt before enciphering it with the crypto algorithm. For example, if my password is "nezperce", and my salt is "qp", then the algorithm will be handed the modified password text "qpnezperce" to encode. If Alice also tries to use password "nezperce", but has salt "w4", then the algorithm will encode the modified password text "w4nezperce" for her. As a result, her enciphered hash will differ from mine, even though we're both using the same password.
Both methods provide the primary benefit of the salt; to ensure that the same password isn't always enciphered the same way. Using salts increases the difficulty of mounting a pre-computed hash attack (formerly I said dictionary or brute force attack, see comments).
I believe that there are other advantages to using a "real salt", such as increased computational overhead (which slows down attacks). But, again, you'd need someone with more crypto skill than I to know whether that's true or not.
I think the advantage of a "fake salt" is that it's easier to implement and requires less crypto savvy. I know that the place I've most commonly seen fake salts is web applications managing their own auth database.