The Computer Language
24.11 Benchmarks Game

simple

These are naive un-optimised single-thread mandelbrot programs. They were transliterated from the same original program. However some language implementations don't buffer stdout by default.

The crowd-sourced Ruby multithread programs are about 7x faster than the simple Ruby program. The exhaustively-optimised + multicore + vector-instruction C programs are more-like 25x faster than the simple C program.

  source secs mem gz    
  Intel C 23.62 20,580 433    
  C gcc #2 25.16 19,772 412    
  C gcc 25.17 19,772 433    
  C clang #2 25.28 19,772 412    
  C clang 25.91 19,772 433    
  Swift 26.46 19,436 400    
  C# aot 26.59 19,988 493    
  Go #2 26.67 19,808 500    
  C# .NET 27.23 32,036 493    
  Go 27.30 19,808 469    
  Java 27.93 45,008 445    
  Node.js 70.99 56,080 380    
  PHP #3 127.57 19,804 418    
  Dart 142.15 19,904 462    
  PHP 167.49 19,804 390    
  Java -Xint 8 min 36,372 445    
  Ruby yjit 20 min 20,992 341    
  Python 3 #3 22 min 19,316 392    
  Toit 24 min 68,864 387    
  Python 3 #2 31 min 19,316 337    
  Matz's Ruby 2h 39 min 19,680 341    

Transliterations are obviously not idiomatic — does that matter for your purposes? When the archive Ruby #2 and Python #2 programs are most idiomatic they are furthest from identical — are they similar-enough to be comparable for your purposes? (The archive Python #2 program uses a built-in complex-number type.)