String concat contains dummy comparision.

在 repository 中搜索 issue 所描述的 String concatenation 基准测试，并阅读其当前设置和已报告的案例。更新该基准测试，使其覆盖随机化的字符串长度和多字符串拼接，然后将结果与 issue 中的示例进行比较，以确认具有误导性的固定输入比较已得到解决。

You are doing your test without any randomization and length variety, so results are incorrect completely.

You simply just testing how fast are 'foo' and 'bar' concatenation, and only

Live case scenario looks more like this:

# strings length is a ( 100 + rand( 100 ) ), i.e. short strings max difference around double 

 String#append:   439150.7 i/s
            String#+:   419672.1 i/s - same-ish: difference falls within error
  "#{'foo'}#{'bar'}":   365799.6 i/s - 1.20x  slower
       String#concat:   353912.5 i/s - 1.24x  slower

 # strings length is a  (1000 + rand( 1000) ) , i.e. longer strings max difference around double

            String#+:   298633.5 i/s
   String#append:   284566.8 i/s - same-ish: difference falls within error
  "#{'foo'}#{'bar'}":   267548.4 i/s - 1.12x  slower
    String#concat:   215419.4 i/s - 1.39x  slower

# string length is a ( 100 + rand(1000) ), i.e. highly variable length
             String#+:   388205.2 i/s
   String#append:   357669.3 i/s - 1.09x  slower
  "#{'foo'}#{'bar'}":   331617.7 i/s - 1.17x  slower
    String#concat:   286717.8 i/s - 1.35x  slower

# string length is a 1 + rand 1000, i.e. random string
             String#+:   385180.8 i/s
   String#append:   373926.8 i/s - same-ish: difference falls within error
  "#{'foo'}#{'bar'}":   341496.6 i/s - 1.13x  slower
   String#concat:   297988.1 i/s - 1.29x  slower

OK

# This is completely oposite to
         "foo" "bar":  5369594.5 i/s
 "#{'foo'}#{'bar'}":  5181745.7 i/s - same-ish: difference falls within error
   String#append:  3075719.2 i/s - 1.75x slower
    String#concat:  3016703.5 i/s - 1.78x slower
            String#+:  2977282.7 i/s - 1.80x slower

Also you didn't test on multiple string concat, where actually "#{'foo'} #{'bar'} #{'shines'}".