n-body benchmark N=50,000,000

Each chart bar shows how many times more Code, one ↓ n-body program used, compared to the program that used least Code.

These are not the only programs that could be written. These are not the only compilers and interpreters. These are not the only programming languages.

Column × shows how many times more each program used compared to the benchmark program that used least.

    sortsortsort 
  ×   Program Source Code CPU secs Elapsed secs Memory KB Code B ≈ CPU Load
1.0Hack #3 21 min21 min55,3441080  1% 0% 0% 100%
1.0PHP #3 11 min11 min3,3401082  0% 0% 0% 100%
1.1Ruby #2 10 min10 min7,5201137  0% 0% 0% 100%
1.1Ruby JRuby #2 6 min6 min691,3921137  0% 1% 1% 100%
1.1Fortran Intel #4 22.4722.485121172  0% 0% 0% 100%
1.1C gcc 20.8120.824361173  0% 1% 1% 100%
1.1C gcc #6 20.5920.604321180  1% 0% 1% 100%
1.1Python 3 15 min15 min5,5721181  1% 0% 0% 100%
1.1Lua #2 7 min7 min1,0401193  0% 0% 0% 100%
1.1Lua 8 min8 min1,0361201  0% 0% 0% 100%
1.1C gcc #3 20.6620.674361208  1% 0% 0% 100%
1.1OCaml 21.8621.877201239  0% 1% 0% 100%
1.2F# Mono 25.0925.1033,0161259  0% 0% 1% 100%
1.2C gcc #2 22.2022.214361263  0% 1% 1% 100%
1.2Dart #2 30.7030.7117,9281274  1% 1% 0% 100%
1.2JavaScript V8 35.9035.9110,4921287  1% 0% 1% 100%
1.2Hack 13 min13 min56,1161288  1% 0% 0% 100%
1.2C# Mono #6 24.4624.4721,9921289  0% 0% 0% 100%
1.2Fortran Intel #3 21.5321.545121299  0% 0% 0% 100%
1.2Lua #4 7 min7 min1,0401305  0% 0% 0% 100%
1.2C# Mono #3 21.7521.7622,3041305  1% 1% 0% 100%
1.2Pascal Free Pascal 34.7534.7681308  1% 0% 1% 100%
1.2Go 22.9622.961,0361310  1% 0% 0% 100%
1.2Rust #2 26.1926.204,8241321  1% 0% 0% 100%
1.2Pascal Free Pascal #2 37.0137.0281322  1% 0% 0% 100%
1.3Scala 26.4826.4926,0401358  0% 0% 0% 100%
1.3Lisp SBCL 42.8542.878,6561363  1% 0% 0% 100%
1.3Rust 20.9921.004,8241371  1% 0% 0% 100%
1.3Fortran Intel 20.0620.065121389  0% 0% 0% 100%
1.3Lisp SBCL #2 26.6726.688,6561398  0% 1% 0% 100%
1.3Erlang HiPE #3 119.80119.8412,1081399  0% 0% 0% 100%
1.3Perl #2 17 min17 min2,6201401  0% 0% 0% 100%
1.3C# Mono 37.9737.9822,4041403  1% 0% 0% 100%
1.3C# Mono #2 22.6722.6822,1041410  1% 1% 1% 100%
1.3Pascal Free Pascal #3 35.4035.4181418  1% 1% 0% 100%
1.3Dart #3 28.3528.3618,7761420  0% 1% 0% 100%
1.3Java  #2 22.6622.6719,5681424  0% 0% 0% 100%
1.3C++ g++ #4 20.9320.943601428  1% 1% 0% 100%
1.3C gcc #5 20.9320.943601429  1% 0% 0% 100%
1.3Hack #5 10 min10 min56,0801432  1% 0% 0% 100%
1.3Clojure #2 27.9027.9260,2841440  1% 0% 0% 100%
1.4C# Mono #4 30.8130.8222,3521461  1% 0% 1% 100%
1.4C# Mono #5 31.5931.6022,7641485  1% 0% 0% 100%
1.4C gcc #4 10.1910.194321490  1% 0% 0% 100%
1.4Fortran Intel #2 19.9519.965121496  0% 0% 0% 100%
1.4Racket 103.92104.0128,5361496  0% 1% 1% 100%
1.4JavaScript V8 #2 36.7036.7110,0201527  0% 0% 1% 100%
1.4C# Mono #7 28.1428.1522,5721539  1% 0% 1% 100%
1.4C++ g++ #8 9.359.354361544  1% 1% 1% 100%
1.4C++ g++ #7 9.799.794361545  0% 1% 1% 100%
1.5Ada 2005 GNAT 25.9225.931,5001608  1% 0% 0% 100%
1.5Smalltalk VisualWorks 192.75192.8042,8801652  0% 0% 0% 100%
1.5Fortran Intel #5 9.059.055121659  1% 0% 1% 100%
1.5C++ g++ 19.5419.544361659  1% 0% 1% 100%
1.5C++ g++ #6 19.3419.354321668  1% 1% 1% 100%
1.6Racket #2 53.9153.9427,5281689  0% 1% 1% 100%
1.6Haskell GHC 27.4027.412,3601706  1% 0% 0% 100%
1.6Ada 2005 GNAT #3 23.2723.281,5001740  1% 0% 0% 100%
1.6C++ g++ #5 13.4513.451,0681749  1% 0% 1% 100%
1.6C++ g++ #3 9.929.928281763  0% 1% 0% 100%
1.7Haskell GHC #2 22.6522.662,3441874  0% 1% 1% 100%
2.0Clojure 27.6627.6758,8642162  0% 1% 1% 100%
2.2Ada 2005 GNAT #5 15.4315.441,4922427  1% 0% 0% 100%
"wrong" (different) algorithm / less comparable programs
1.4JavaScript V8 #3 33.3133.329,7761505
2.1C++ g++ #2 13.3913.401,0682288

 n-body benchmark : Double-precision N-body simulation

You can write your own program for this task and contribute to the benchmarks game by following these general instructions.

More specifically:

ndiff -abserr 1.0e-8 program output N = 1000 with this output file to check your program is correct before contributing.

We are trying to show the performance of various programming language implementations - so we ask that contributed programs not only give the correct result, but also use the same algorithm to calculate that result.

Each program should model the orbits of Jovian planets, using the same simple symplectic-integrator - see the Java program.

For background information see N-body problem. Useful symplectic integrators are freely available, for example the HNBody Symplectic Integration Package.

Thanks to Mark C. Lewis for this benchmark.

Revised BSD license

  Home   Conclusions   License   Play