/mobile Handheld Friendly website

 mandelbrot benchmark N=16,000

Each chart bar shows how many times slower, one ↓ mandelbrot program was, compared to the fastest program.

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
1.0C gcc #4 22.625.6627,076799  100% 100% 100% 100%
1.0C gcc #3 22.785.7331,740763  99% 99% 100% 99%
1.0C gcc #7 22.845.7531,760994  99% 99% 99% 100%
1.0C gcc 23.425.8726,804822  100% 100% 100% 100%
1.0C++ g++ #7 23.475.9032,0881017  99% 99% 99% 100%
1.0C gcc #6 23.485.8826,560879  100% 100% 100% 100%
1.1Ada 2005 GNAT #3 24.826.2629,9241805  99% 100% 99% 99%
1.2Java  #2 27.176.9573,488796  97% 97% 99% 98%
1.2C++ g++ #6 27.967.0231,7601035  99% 100% 100% 99%
1.3Scala #4 28.457.36201,012786  96% 96% 99% 97%
1.3C gcc #9 29.977.7831,724694  95% 100% 95% 95%
1.3Java  #6 30.447.7669,188802  98% 97% 99% 98%
1.4C++ g++ #9 30.928.0432,088726  100% 95% 95% 95%
1.4C# Mono #6 31.207.9549,152986  98% 99% 98% 98%
1.4C++ g++ #8 31.3011.5332,088742  100% 36% 36% 100%
1.4C# Mono #4 31.978.1349,812872  98% 100% 98% 98%
1.5Clojure #7 34.729.60134,8081195  94% 89% 88% 92%
1.5C++ g++ #5 35.0635.0732,036590  0% 1% 100% 0%
1.6F# Mono #4 35.949.2454,572869  98% 97% 98% 97%
1.8C++ g++ #2 40.1340.1432,036687  0% 1% 100% 0%
1.9Java  #3 42.2510.7269,932903  98% 98% 99% 98%
1.9Java  43.3943.3615,652665  23% 78% 1% 1%
2.0C# Mono #3 45.1411.4349,216701  99% 99% 99% 99%
2.0F# Mono #3 45.6211.5850,372800  98% 98% 98% 100%
2.0C# Mono 45.9611.6349,420798  99% 100% 99% 99%
2.1Haskell GHC #2 48.0612.0433,272782  100% 100% 100% 100%
2.2Go #6 48.6412.1730,116700  100% 100% 100% 100%
2.2Fortran Intel #4 48.7318.4160,388611  96% 34% 96% 40%
2.2C gcc #2 49.9449.96328400  0% 1% 100% 1%
2.3Scala 50.9212.9249,556724  100% 98% 98% 98%
2.3F# Mono #2 51.2913.7818,2681043  99% 91% 95% 88%
2.3Scala #2 52.5052.4218,636454  5% 95% 0% 1%
2.3Dart 52.6219.24133,852901  39% 38% 99% 98%
2.5Clojure #5 55.9454.3389,6521069  3% 5% 84% 12%
2.5Fortran Intel #5 56.8956.9132,608508  0% 0% 100% 0%
2.5C# Mono #2 57.0357.0514,912484  100% 1% 1% 0%
2.6Ada 2005 GNAT #2 58.0758.09968572  0% 1% 100% 0%
2.6Clojure #6 58.3616.0886,3761069  92% 92% 90% 90%
3.0Lisp SBCL 66.9018.7535,3482459  80% 79% 98% 99%
3.3Lisp SBCL #3 75.6329.0335,404888  95% 33% 33% 100%
3.4Racket #3 76.6876.6617,636797  0% 1% 100% 0%
3.5Racket #4 78.2419.7651,112796  99% 99% 100% 99%
3.5Pascal Free Pascal #3 78.7478.768530  95% 0% 0% 6%
3.7Pascal Free Pascal 83.7031.6431,900820  94% 41% 95% 36%
3.7Pascal Free Pascal #2 83.7131.6031,900817  95% 35% 48% 88%
3.8OCaml #6 85.8085.841,648444  0% 95% 5% 0%
4.2Racket #2 95.7895.7616,036585  1% 0% 0% 100%
4.5Rust 101.23101.26672633  1% 0% 0% 100%
4.7Racket 106.66106.6415,044517  0% 1% 100% 0%
6.0Pascal Free Pascal #5 135.8434.1830,632829  100% 100% 100% 100%
6.6OCaml 150.0337.713,736710  100% 100% 100% 99%
8.1C++ g++ #3 183.59183.97884414  28% 36% 32% 7%
11Pascal Free Pascal #4 240.0760.1530,340748  100% 100% 100% 100%
17Erlang HiPE 6 min107.28662,920534  94% 92% 91% 92%
46Ruby JRuby #3 17 min17 min619,024307  24% 30% 26% 23%
51PHP #3 19 min290.9875,728863  100% 100% 100% 100%
53PHP 19 min19 min3,124443  29% 18% 2% 53%
81Python 3 #2 30 min8 min7,488777  100% 94% 93% 94%
106Python 3 #5 39 min9 min28,064449  100% 100% 100% 100%
164Perl 1h 01 min15 min37,660579  100% 100% 100% 100%
174Ruby #3 1h 05 min1h 05 min5,284307  28% 22% 21% 30%
194Ruby #2 1h 13 min18 min88,412925  100% 100% 100% 100%
300Ruby JRuby #2 1h 53 min28 min668,756925  99% 99% 99% 99%
Fortran Intel Bad Output485
Fortran Intel #3 Bad Output754
Fortran Intel #2 Bad Output813
Python 3 #3 Failed304
Scala #3 Failed796
"wrong" (different) algorithm / less comparable programs
0.3C gcc #5 7.531.9015,9001026
0.5C++ g++ 10.652.6725,4601239
0.8C gcc #8 17.644.4228,568859
7.0Python 3 #6 157.5440.9446,1601380

 mandelbrot benchmark : Generate Mandelbrot set portable bitmap file

cmp program output N = 200 with this 5KB 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.

Mandlebrot output N=200,converted to PNG

Each program should plot the Mandelbrot set [-1.5-i,0.5+i] on an N-by-N bitmap. Write output byte-by-byte in portable bitmap format.

For more information see Eric W. Weisstein, "Mandelbrot Set." From MathWorld--A Wolfram Web Resource.
http://mathworld.wolfram.com/MandelbrotSet.html

Thanks to Greg Buchholz for this benchmark.

Revised BSD license

  Home   Conclusions   License   Play