[racket] generator performance

A few minutes ago, Matthew Flatt wrote:
> 
> Using `call/cc' for generators is effectively a hint to the run-time
> system that the continuation doesn't need to compose. That hint is
> useful only because of the way that continuations are implemented
> internally.

So `call/cc' is faster than one of the delimited ones?

-- I tought that the main cost would be in the amount of stack saved
which would make delimited ones faster, and it sounds like an
obviously bogus intuition...  Is there some simple explanation to
understand relative costs without knowing about the bits?

-- 
          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
                    http://barzilay.org/                   Maze is Life!