OUT-OF-DATE! Read http://shootout.alioth.debian.org/      

- all benchmarks - binary-trees chameneos-redux fannkuch fasta k-nucleotide mandelbrot meteor-contest n-body nsieve nsieve-bits partial-sums pidigits recursive regex-dna reverse-complement spectral-norm startup sum-file thread-ring - all languages - Ada 2005 GNAT BASIC FreeBASIC C CINT C GNU gcc C Intel C# Mono C++ GNU g++ C++ Intel CAL Clean D Digital Mars Eiffel SmartEiffel Erlang HiPE F# Mono Forth bigForth Forth GNU GForth Fortran G95 Fortran Intel Groovy Haskell GHC Icon Io Java 1.4 -server Java 6 -client Java 6 -server Java 6 -Xint Java 6 -Xms64m Java GNU gcj JavaScript Rhino JavaScript SpiderMonkey Lisaac Lisp SBCL Lua Lua LuaJIT Mercury Mozart/Oz Nice Oberon-2 OO2C OCaml Pascal Free Pascal Perl PHP Pike Prolog SWI Prolog YAP Python CPython Python IronPython Python Psyco Rebol Ruby 1.9 Ruby JRuby Ruby MRI Scala Scheme Chicken Scheme Ikarus Scheme PLT Smalltalk GNU Smalltalk Squeak Smalltalk VisualWorks SML MLton SML SML/NJ Tcl Zonnon Mono

 recursive benchmark hypotheses non fingo

↓ Naive recursive-algorithms: ack fib tak, N=11.

Which programs used least Code? Which programs use highly optimised assembly code libraries? Which programs make use of all the processor cores?

		sort	sort	sort	sort
×	Program & Logs	CPU secs	Elapsed secs	Memory KB	Code B	~ CPU Load
1.0	C GNU gcc	2.36		432	547
1.0	C++ GNU g++ #2	2.40		1,008	566
1.0	Lisaac	2.45		524	590
1.1	Ada 2005 GNAT #2	2.51		548	598
1.1	C GNU gcc #2	2.54		484	425
1.1	Oberon-2 OO2C	2.54		1,784	505
1.1	C++ GNU g++	2.59		916	345
1.1	Eiffel SmartEiffel	2.63		584	579
1.4	Mercury	3.23		3,720	545
1.4	C Intel	3.24		692	547
1.4	C Intel #2	3.28		700	425
1.6	Clean #4	3.80		596	422
1.6	C++ Intel #2	3.83		724	566
1.7	Mercury #2	3.94		3,740	696
1.7	Clean #3	4.10		600	427
1.7	C++ Intel	4.11		512	345
1.8	Pascal Free Pascal	4.14		460	462
2.0	Java 1.4 -server	4.81		10,464	497
2.5	CAL	5.80		14,320	665
2.7	D Digital Mars #2	6.31		1,196	349
2.8	Scala #2	6.53		20,184	411
2.9	Java 6 -server	6.76		12,044	427
2.9	Java 6 -Xms64m	6.82		12,180	427
3.0	Fortran Intel	7.17		880	620
3.2	OCaml #2	7.47		1,120	389
3.3	F# Mono	7.70		7,500	461
3.4	BASIC FreeBASIC #2	7.94		1,068	449
3.4	SML MLton #2	7.97		1,384	453
3.4	Java GNU gcj	7.98		15,828	497
3.4	Haskell GHC	8.05		1,464	447
3.6	Lisp SBCL #3	8.45		17,460	523
3.7	Scheme Chicken	8.68		2,384	467
3.7	Forth bigForth	8.77		872	413
3.9	C# Mono	9.22		5,060	435
4.0	SML SML/NJ	9.42		5,244	585
4.0	Java 6 -client	9.43		10,716	427
4.6	Scheme Ikarus	10.81		20,880	564
4.6	Nice	10.86		24,328	377
4.6	Nice #2	10.89		23,792	393
4.7	Fortran G95	10.98		1,244	620
8.1	Scheme PLT	19.07		6,988	579
8.4	Scheme PLT #2	19.82		7,708	427
11	Erlang HiPE #2	26.45		4,856	392
15	Python Psyco	36.12		3,420	354
18	Forth GNU GForth	43.30		956	493
21	Lua LuaJIT	48.67		2,912	330
24	Java 6 -Xint	55.66		10,520	427
38	Mozart/Oz	89.35		5,192	574
40	Mozart/Oz #2	93.06		6,100	500
47	Prolog YAP	111.65		22,820	493
49	Lua	114.38		2,280	330
66	Pike	154.91		5,312	403
69	Smalltalk VisualWorks	161.83		32,764	509
80	Python IronPython	187.65		19,588	360
146	Prolog SWI	5 min		3,320	587
238	Groovy	9 min		63,848	361
290	Python CPython	11 min		14,296	360
305	Ruby JRuby #2	11 min		49,096	301
373	PHP #2	14 min		18,088	315
381	Perl #2	14 min		10,836	380
543	Tcl	21 min		24,908	461
629	Ruby MRI #2	24 min		48,360	301
636	Smalltalk Squeak	24 min		20,736	505
7,140	Io	4h 40 min		58,540	331
	C CINT	Timed Out			403
	Erlang HiPE	Failed			356
	Erlang HiPE #3	Failed			373
	Icon	Failed			311
	JavaScript Rhino	Failed			306
	JavaScript SpiderMonkey	Failed			306
	Rebol	Failed			504
	Ruby 1.9 #2	Failed			301
	Scala	Failed			420
	Smalltalk GNU	Failed			510
	Zonnon Mono	Failed			493

 recursive benchmark : Naive recursive-algorithms: ack fib tak

diff program output N = 3 with this output file to check your program is correct before contributing.

Each program should use the same naïve recursive-algorithms to calculate 3 simple numeric functions: ackermann, fibonnaci and tak.

Ack(x,y)
  x = 0     = y+1
  y = 0     = Ack(x-1,1)
  otherwise = Ack(x-1, Ack(x,y-1))

Fib(n)
  n < 2     = 1
  otherwise = Fib(n-2) + Fib(n-1)

Tak(x,y,z)
  y < x     = Tak(Tak(x-1.0,y,z),Tak(y-1.0,z,x),Tak(z-1.0,x,y))
  otherwise = z

For this benchmark, the fibonnaci and tak implementations should either provide separate functions - one for integer calculation and one for double calculation - or provide a function that uses integer calculation with integer parameters and double calculation with double parameters.

The Ackermann benchmark is described in Timing Trials, or, the Trials of Timing: Experiments with Scripting and User-Interface Languages.

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

For a different version of the Fibonacci function see Eric W. Weisstein, "Fibonacci Number." From MathWorld--A Wolfram Web Resource.
http://mathworld.wolfram.com/FibonacciNumber.html

The tak benchmark is described in Performance and Evaluation of Lisp Systems, Richard P. Gabriel, 1985, page 81. (1.1MB pdf)

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