Ruby dynamic calls benchmark

gistfile1.rb

n = 100000

# taken from [here](http://blog.lojic.com/2008/12/22/ruby-dynamic-method-invocation-performance/), 
# but the original one use the same string all the time, which doesn't reflect real world situation

bm(12) {|x|
  x.report("orig") { n.times { test = "stormy weather"; test.length } }
  x.report("call") { n.times { test = "stormy weather"; m = test.method(:length); m.call } }
  x.report("send") { n.times { test = "stormy weather"; test.send(:length) } }
  x.report("eval") { n.times { test = "stormy weather"; eval "test.length" } }
}

#                   user     system      total        real
# orig           0.020000   0.000000   0.020000 (  0.017626)
# call           0.080000   0.010000   0.090000 (  0.093148)
# send           0.030000   0.000000   0.030000 (  0.023334)
# eval           0.990000   0.000000   0.990000 (  1.005901)
#
# we can see method.call is more than 5 times slower
#