[Ecls-list] ECL shootout

Marko Kocić marko.kocic at gmail.com
Tue Jun 30 11:01:06 UTC 2009


And here are updated results with unithreaded ECL. Seems like threads
add big overhead to benchmarks.

2009/6/30 Marko Kocić <marko.kocic at gmail.com>:
> I downloaded cl-bench from
> http://common-lisp.net/project/cl-bench/running.shtml and was able to
> run it with latest ECL, SBCL and CCL versions with this small patch
> attached.
>
> Here are the results on Linux x86 box. If someone have both mingw and
> msvc installed I would like to see  those results compared too.
>
> In short ECL beats them all on bignum tests (expected since it uses
> GMP), but is overally much slower in other tests. I used threaded ECL.
>
> Regards,
> Marko
>
-------------- next part --------------
This is SBCL 1.0.29.50, an implementation of ANSI Common Lisp.
More information about SBCL is available at <http://www.sbcl.org/>.

SBCL is free software, provided as is, with absolutely no warranty.
It is mostly in the public domain; some portions are provided under
BSD-style licenses.  See the CREDITS and COPYING files in the
distribution for more information.
; loading system definition from
; /home/markko/cvstree/clbuild/systems/asdf-binary-locations.asd into
; #<PACKAGE "ASDF0">
; registering #<SYSTEM ASDF-BINARY-LOCATIONS {AD709B9}> as ASDF-BINARY-LOCATIONS
* Benchmark                 Reference  ECL N  Clozu  ECL 9
-------------------------------------------------------------------------------------
COMPILER                 [      1.40]   1.05   0.72   1.49
WALK-LIST/SEQ            [      0.02]   4.65   2.85   4.50
WALK-LIST/MESS           [      0.03]   3.90   1.80   3.90
BOYER                    [      2.95]   0.27   1.99   0.38
BROWSE                   [      0.23]   2.08   1.43   2.44
DDERIV                   [      0.20]   4.09   1.71   4.99
DERIV                    [      0.22]   3.98   1.69   4.79
DESTRUCTIVE              [      0.19]   3.17   1.76   3.42
DIV2-TEST-1              [      0.33]   3.60   1.48   4.15
DIV2-TEST-2              [      0.41]   3.75   1.50   4.62
FFT                      [      0.05]  24.60   2.86  26.72
FRPOLY/FIXNUM            [      0.24]   3.75   1.63  11.66
FRPOLY/BIGNUM            [      0.29]   2.09   1.81   4.14
FRPOLY/FLOAT             [      0.36]   2.52   2.55   7.66
PUZZLE                   [      0.22]  12.64   3.93  14.10
TAK                      [      0.21]   1.90   1.48   2.90
CTAK                     [      0.26]   4.19   2.07   5.45
TRTAK                    [      0.22]   1.84   1.49   2.79
TAKL                     [      0.38]   0.92   1.54   1.29
STAK                     [      0.38]   1.74   1.70  19.01
FPRINT/UGLY              [      1.54]   1.27   1.36   1.56
FPRINT/PRETTY            [      1.91]  17.15   1.09  23.27
TRAVERSE                 [      0.46]   2.03   1.55   2.90
TRIANGLE                 [      0.47]   6.37   1.60   8.15
RICHARDS                 [      0.48]   1.29   2.95   4.54
FACTORIAL                [      0.21]   1.64   2.68   1.73
FIB                      [      0.20]   2.87   0.66   3.41
FIB-RATIO                [      0.03]   3.64   1.42   3.94
ACKERMANN                [      2.58]   2.75   0.43   2.97
MANDELBROT/COMPLEX       [      0.38]   1.97   0.80   2.12
MANDELBROT/DFLOAT        [      0.02]  56.88   8.24  62.94
MRG32K3A                 [      0.71]   3.98  11.11   4.83
CRC40                    [      9.93]   6.04   2.00   6.49
BIGNUM/ELEM-100-1000     [      0.14]   0.47   5.41   0.51
BIGNUM/ELEM-1000-100     [      0.34]   0.36  15.64   0.36
BIGNUM/ELEM-10000-1      [      0.41]   0.35  15.99   0.36
BIGNUM/PARI-100-10       [      0.02]   0.58  80.58   0.42
BIGNUM/PARI-200-5        [      0.08]   0.48 343.11   0.48
PI-DECIMAL/SMALL         [      1.06]   3.58  25.38   3.63
PI-DECIMAL/BIG           [      1.06]   3.74  92.91   3.76
PI-ATAN                  [      0.85]   0.95   2.59   1.00
PI-RATIOS                [      1.51]   0.77   3.19   0.78
HASH-STRINGS             [      0.17]   3.98  34.46   4.45
HASH-INTEGERS            [      0.38]   2.11   8.65   2.29
SLURP-LINES              [      1.43]   3.42   1.49   3.93
BOEHM-GC                 [      0.79]   5.80   9.81   6.32
DEFLATE-FILE             [      0.22]   3.55   1.78   4.03
1D-ARRAYS                [      0.04]   4.91   0.77   5.59
2D-ARRAYS                [      0.48]  10.23   5.33  10.21
3D-ARRAYS                [      1.57]  18.02   3.50  18.39
BITVECTORS               [      0.85]   6.04  35.33   6.09
BENCH-STRINGS            [      0.70]  11.90   4.46  14.35
fill-strings/adjustable  [      7.83]   2.07   4.26   2.40
STRING-CONCAT            [NIL       ]  -1.00  -1.00  -1.00
SEARCH-SEQUENCE          [      0.15]  24.11  14.53  32.31
CLOS/defclass            [      0.81]   0.66   0.25   0.93
CLOS/defmethod           [      3.29]   0.08   0.02   0.11
CLOS/instantiate         [      5.34]  17.00   0.46  18.79
CLOS/simple-instantiate  [      0.15] 336.12  28.49 475.84
CLOS/methodcalls         [      0.98]   4.31   2.76   7.19
CLOS/method+after        [      2.32]   1.47   0.60   2.48
CLOS/complex-methods     [      1.61]   1.42   0.37   2.85
EQL-SPECIALIZED-FIB      [      0.26]   7.59   2.56  14.01
Reference time in first column is in seconds; other columns are relative
Reference implementation: SBCL 1.0.29.50
Impl ECL N: ECL NT 9.6.2
Impl Clozu: Clozure Common Lisp Version 1.4-dev-r12331M-trunk  (LinuxX8632)
Impl ECL 9: ECL 9.6.2
=== Test machine ===
   Machine-type: X86
   Machine-version: Intel(R) Pentium(R) M processor 2.13GHz



More information about the ecl-devel mailing list