Comment by CanSpice

I think Richard Feynman is a good counter-example to the "single function machines" hypothesis.