Hi, That's exactly my point, which you seem to be missing. I'm nowhere trying to be smart, but only to show that even for a simplistic problem as the classical 3-poles Hanoi Tower, getting an iterative solution than does *better* than the iterative counterpart, that is NOT mimic a call stack in any way (since in that event, you better off with the recursive approach in the first place), is a hell of a proposition, as you can see (my personal solution, not yet published, does not use a stack, is sound by design, but utterly complex). Therefore this is my point : general recursive functions (whether implemented thru the built-in support of the language or thru functional continuations to render the stack explicit and provide potential for applying various traversal strategies) are not only powerful in their abilities to capture complex algorithms, but sometimes unavoidable when the iterative counterpart (which does not mimic a stack) is too complex or even not known (egAckerman iterative is not known). In conclusion, one should not conceive recursion as limited to what built-in support actual languages offer, but as an abstract conceptual tool, as it is with pure FP. This is the point I wanted to emphasize on a simple example. Cheers