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.

    sort sortsort
  ×   Program Source Code CPU secs Elapsed secs Memory KB Code B ≈ CPU Load
1.0C gcc #4 22.405.6128,988799  100% 100% 100% 100%
1.0C gcc #7 22.905.7532,940994  100% 100% 99% 100%
1.0C++ g++ #7 22.975.7733,5921017  99% 99% 99% 100%
1.1C gcc #3 24.846.2432,424763  100% 100% 100% 100%
1.1C gcc 24.996.2627,752822  100% 100% 100% 100%
1.1C gcc #6 25.096.2827,760879  100% 100% 100% 100%
1.2Rust #2 25.626.5357,5041080  100% 100% 96% 97%
1.2Go #3 27.106.8133,680894  100% 100% 100% 100%
1.2C++ g++ #6 27.596.9332,8881035  100% 100% 100% 100%
1.3Java  #2 27.797.1287,832796  97% 99% 97% 97%
1.3Scala #4 27.687.1592,200786  96% 96% 99% 97%
1.3Java  #6 29.037.4487,332802  99% 97% 97% 97%
1.4C gcc #9 29.637.6932,464694  95% 100% 95% 95%
1.4C# Mono #6 30.667.8874,868986  97% 97% 99% 97%
1.4C++ g++ #9 31.008.0633,444726  95% 95% 95% 100%
1.4C# Mono #4 31.478.0974,852872  97% 97% 97% 99%
1.4Ada 2005 GNAT #3 32.258.1331,7561805  100% 99% 100% 99%
1.7F# Mono #4 36.359.6080,748869  94% 94% 97% 94%
1.7Clojure #7 34.909.66133,6881195  88% 96% 89% 90%
1.9Java  #3 42.4910.8086,168903  98% 98% 98% 100%
1.9Go #2 43.4110.8931,656837  100% 100% 100% 100%
2.0C++ g++ #8 30.9611.4233,504742  100% 100% 37% 36%
2.0C# Mono #3 45.0011.4874,564701  98% 98% 99% 98%
2.1F# Mono #3 45.1211.5575,924800  97% 100% 97% 97%
2.1C# Mono 46.0111.7474,324798  99% 98% 97% 98%
2.1Go #6 48.0212.0332,264700  100% 100% 100% 100%
2.3Scala 50.9812.9660,476724  98% 98% 100% 98%
2.5Haskell GHC #2 55.7313.9734,212782  100% 100% 100% 100%
2.5F# Mono #2 51.9914.1943,6481043  96% 96% 89% 86%
2.8Scala #5 43.7815.9387,072589  71% 66% 77% 62%
2.8Clojure #6 58.8015.9493,5041069  90% 96% 91% 93%
3.3Fortran Intel #4 48.9018.7865,080611  93% 34% 93% 42%
3.4Go 75.7818.9932,112823  100% 100% 100% 100%
3.5Racket #4 78.1019.7355,048796  99% 100% 99% 99%
3.6Dart 54.9220.2366,540901  98% 37% 98% 38%
3.8OCaml 84.9321.321,740710  100% 100% 100% 100%
4.0Lisp SBCL 80.2522.2956,0002459  98% 82% 100% 82%
5.2Lisp SBCL #3 75.4828.9746,592888  38% 95% 34% 95%
5.6Pascal Free Pascal #2 83.6231.4931,948817  95% 36% 41% 95%
5.6Pascal Free Pascal 83.6031.5131,948820  36% 100% 36% 95%
6.1Pascal Free Pascal #5 135.8434.0430,560829  100% 100% 100% 100%
6.1C++ g++ #5 34.1634.1833,220590  0% 0% 0% 100%
6.8C++ g++ #2 38.0538.0633,448687  0% 0% 100% 0%
7.8Java  43.7643.7323,428665  1% 99% 2% 1%
8.9C gcc #2 49.9950.01664400  1% 0% 0% 100%
9.4Scala #2 52.5552.4826,496454  1% 68% 2% 31%
9.5Ada 2005 GNAT #2 53.4753.481,624572  1% 0% 0% 100%
9.7Clojure #5 56.3254.5793,7481069  61% 21% 13% 10%
10Fortran Intel #5 56.5956.6132,600508  1% 0% 1% 100%
13Pascal Free Pascal #4 289.8772.5930,980748  100% 100% 100% 100%
14Racket #3 76.7676.7119,720797  1% 52% 1% 49%
14Pascal Free Pascal #3 78.7278.758530  1% 100% 0% 0%
18Racket #2 100.56100.5521,512585  1% 6% 1% 94%
21Racket 116.23116.2119,752517  1% 100% 0% 0%
23C# Mono #2 127.47127.5238,896484  19% 31% 25% 27%
29Erlang HiPE #2 10 min162.01635,812613  98% 97% 97% 97%
30Erlang HiPE 10 min169.77883,836534  95% 97% 94% 95%
31C++ g++ #3 171.26171.321,136414  24% 21% 19% 37%
46OCaml #6 259.44259.50864444  27% 1% 13% 62%
49Python 3 #7 18 min277.2728,372686  100% 100% 100% 100%
50PHP #3 18 min280.48107,708863  100% 100% 100% 100%
76Python 3 #2 27 min7 min47,620777  100% 98% 98% 98%
114Python 3 #5 42 min10 min28,112449  100% 100% 100% 100%
122Ruby JRuby #4 44 min11 min2,253,832849  98% 98% 98% 98%
166Perl 1h 02 min15 min44,184579  100% 100% 100% 100%
188Ruby #5 1h 10 min17 min51,4081019  100% 100% 100% 100%
196Ruby JRuby #3 18 min18 min648,516307  26% 32% 20% 25%
198PHP 18 min18 min3,104443  8% 35% 29% 29%
233Ruby #2 1h 27 min21 min86,532925  100% 100% 100% 100%
247Ruby #6 1h 28 min23 min112,484909  97% 97% 97% 97%
322Ruby JRuby #2 1h 58 min30 min714,896925  99% 99% 99% 99%
Fortran Intel Bad Output485
Fortran Intel #3 Bad Output754
Fortran Intel #2 Bad Output813
Ruby #3 Timed Out1h 00 min307
Ruby #4 Timed Out1h 00 min866
Ruby JRuby #6 Failed885
Ruby JRuby #5 Failed1019
Rust Make Error1323
Scala #3 Failed796
"wrong" (different) algorithm / less comparable programs
0.3C gcc #5 7.601.9216,9241026
0.5C++ g++ 11.472.8828,1081239
0.8C gcc #8 17.534.3931,700859
7.1Python 3 #6 152.0339.8050,2201380

 mandelbrot benchmark : Generate Mandelbrot set portable bitmap file

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

More specifically:

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