The talkorigins article uses the phrase “possible functional cytochrome c protein sequences.”, in an article on “functional” redundancy.
”functional” being the key word in both.
“functional” does not necessarily refer to just ‘optimal’, but would also include any others that would ‘work’.
Thus, even the sourced talkorigins article shows that Yockey was referring to “functional” not ‘optimal’ for the calculated figure.
On the previous page Cofty wrote:
”The number of sequences that will produce something 'good enough' is orders of magnitude greater than the - astonishingly massive - number of sequences that result in the optimum version.”
I haven’t seen any separate calculations for “optimum versions”.