Comments on Bill the Lizard: SICP Exercise 1.16: Fast Exponentiation

geeknanny, You're absolutely right. The tail-...

2010-02-18T09:21:49.617-05:00

geeknanny,
You're absolutely right. The tail-call version will be more efficient in both time and (stack) space for most exponents. (They'll run virtually the same for exponents that are powers of 2.)

I hadn't noticed this before, thanks for pointing it out.

pbewig's ipow function isn't tail-recursiv...

2010-02-18T00:00:19.220-05:00

pbewig's ipow function isn't tail-recursive, is it? The else condition calls ipow and then multiplies the result by b. Rewriting the function to be tail-recursive is the whole point of the exercise, yes?

pbewig, There's a lot of really useful stuff i...

2010-01-11T19:51:23.728-05:00

pbewig,
There's a lot of really useful stuff in your Standard Prelude. That will probably be the first place I look for hints if I get stuck on any SICP exercises. I can already tell it will come in handy when I get to the next chapter, which deals with data abstraction.

The ipow function in my Standard Prelude does this using exactly the same algorithm:

(define (ipow b e)
  (cond ((zero? e) 1)
        ((even? e) (ipow (* b b) (/ e 2)))
        (else (* b (ipow (* b b) (/ (- e 1) 2))))))

Comments on Bill the Lizard: SICP Exercise 1.16: Fast Exponentiation

geeknanny, You're absolutely right. The tail-...

pbewig's ipow function isn't tail-recursiv...

pbewig, There's a lot of really useful stuff i...

The ipow function in my Standard Prelude does this using exactly the same algorithm:(define (ipow b e) (cond ((zero? e) 1) ((even? e) (ipow (* b b) (/ e 2))) (else (* b (ipow (* b b) (/ (- e 1) 2))))))

The ipow function in my Standard Prelude does this using exactly the same algorithm:

(define (ipow b e)
(cond ((zero? e) 1)
((even? e) (ipow (* b b) (/ e 2)))
(else (* b (ipow (* b b) (/ (- e 1) 2))))))