From: Ken Kunz <kennethkunz@...>
Date: 2005-06-22T07:20:36+09:00
Subject: Re: 1. Ruby result: 101 seconds , 2. Java result:9.8 seconds, 3. Perl result:62 seconds

Here's a version that skips even numbers:

class Integer
  def is_prime?
    return false if self < 2
    return false if self % 2 == 0
    max_factor = Math.sqrt(self).to_i
    3.step(max_factor, 2) {|i| return false if self % i == 0 }
    return true
  end
end

And here's a benchmark comparison:

require 'benchmark'

Benchmark.bm(12) do |x|
  x.report("original:")  { 2.upto(50000) {|i| is_prime(i) } }
  x.report("Florian's:") { 2.upto(50000) {|i| i.prime? } }
  x.report("Ken's:")     { 2.upto(50000) {|i| i.is_prime? } }
end

                  user     system      total        real
original:   171.767000   0.020000 171.787000 (174.627000)
Florian's:    2.804000   0.010000   2.814000 (  2.814000)
Ken's:        1.041000   0.000000   1.041000 (  1.042000)

What's more important, how fast a program _runs_, or how fast I can
_write_ it?  With Ruby, I can almost always write something that is
"fast enough", and I can write it a whole lot quicker than in Java or
C.  The above (semi-) optimized version only took 2 minutes to write.
(And as a side note... it was more fun to write it in Ruby :) )

And when you need some extra umph... there's always YARV or ruby2c...
and when you really need it (rarely), manually optimized C.

Cheers,
Ken