Early last year, I wrote a solution to the Minimum Enclosing Circle problem in Ruby. I figured this code was "real world" enough to compare performance of Ruby 1.8.7 to Ruby 1.9.1. There are a lot of numbers flying around claiming Ruby 1.9 is anywhere from 5 - 10 times faster than 1.8.7 and I have verified this on very small code samples.

The MEC solution is more real world. Here are some benchmark numbers from running the code under both 1.8.7 and 1.9.1. 

$ ruby -v
  ruby 1.8.7 (2009-06-12 patchlevel 174) [x86_64-linux]
$ ruby mec.rb benchmark
                     user     system      total        real
points: 1        0.000000   0.010000   0.010000 (  0.004344)
points: 10       0.130000   0.000000   0.130000 (  0.157543)
points: 100      0.960000   0.060000   1.020000 (  1.044081)
points: 1000    14.710000   1.460000  16.170000 ( 16.820545)
points: 10000  186.660000  14.360000 201.020000 (218.940879)
$ rvm 1.9.1
$ ruby -v
ruby 1.9.1p243 (2009-07-16 revision 24175) [x86_64-linux]
$ ruby mec.rb benchmark
                     user     system      total        real
points: 1        0.000000   0.000000   0.000000 (  0.003518)
points: 10       0.060000   0.000000   0.060000 (  0.074752)
points: 100      0.590000   0.000000   0.590000 (  0.589993)
points: 1000     7.820000   0.000000   7.820000 (  7.922892)
points: 10000  106.140000   0.620000 106.760000 (113.783464)

As you can see, I get anywhere from 39% - 47% better performance. Call Ruby 1.9.1 twice as fast as Ruby 1.8.7 in a real world code base. I think this is what we can expect out of real world code bases when transitioning to 1.9.1

As a side note, I am using Ruby Version Manager to switch between my system installed Ruby (1.8.7) and 1.9.1. Kudos to Wayne Seguin, author of rvm. It is a beautiful package.