The Pokémon Regularity Problem | Nathan Dalaklis 

Although that sounds probable to me, even if there is an arbitrarily large FSA that can handle solving the problem when we mathematically abstract that far away from it, the problem posed, at least in its original form, is an applied coding construction problem. So the “desired” solution wouldn’t be that abstract. So to be explicit, the problem is can we write a program P in language L that implements an FSA that can solve Pokémon Blue/Red/Green/Yellow in finite time with finite (and more realistically restricted) resources. If we rely on the memory of the console to store our program, the console itself would not be able to store the FSA used in the program as you have described as it would be too large to store. There are probably other considerations to make as well as one considers different implementations of the program with such a large FSA with finite resources since program memory utilization is dynamic in most cases (this is probably one of the reasons why the original talk on the problem mentions utilizing amazon’s hosting infrastructure to run some cases of the problem and how expensive it was 😅)