At Sun, 4 Nov 2012 16:38:25 -0800 (PST), Hugh Aguilar wrote:
> Has anybody done any benchmarks comparing Racket, Gambit, Chicken, or
> any other Scheme, for speed?
As it happens, as a sanity check on various changes that I've made
recently, I've recently re-run a bunch of conventional Scheme
benchmarks on Racket v5.3.1.4, Gambit v4.6.6, and Chicken v4.8.0 in
64-bit mode on Mac OS X using a 1.8GHz MacBook Air. A table of results
is enclosed.
For more information on the benchmarks, as well as more careful but
out-of-date measurements, see
http://blog.racket-lang.org/2010/01/benchmarks.html
Naturally, I will stress that all benchmark result are of limited use.
Still, I think the general conclusion that "the fast Lisps are all
pretty comparable" is about right in practice.
| | fastest | chicken | gambit | racket |
| conform | 129 | 261 ms | 12.72 | 1.32 | 27.16 | 1 | 1 | 1.82 |
| cpstak | 178 | 2727 ms | 1.50 | 1.94 | 1 | 1 | 1.74 | 1.60 |
| ctak | 177 | 138 ms | 1.50 | 1 | 1 | 1.74 | 1.81 | 43.79 |
| dderiv | 244 | 2052 ms | 1.34 | 5.76 | 1 | 1 | 1.18 | 1.70 |
| deriv | 203 | 1561 ms | 1.60 | 8.44 | 1 | 1 | 1.57 | 1.82 |
| destruct | 25 | 173 ms | 14.04 | 2.18 | 22.32 | 1 | 1 | 1.96 |
| div | 242 | 4811 ms | 1.19 | 1 | 1 | 1.04 | 1.06 | 1.12 |
| dynamic | 231 | 1011 ms | 23.87 | 1.45 | 217.90 | 19.06 | 1 | 1 |
| dynamic2 | 649 | 1173 ms | 8.27 | 1.05 | 76.91 | 16.52 | 1 | 1 |
| earley | 373 | 360 ms | 3.04 | 2.30 | 21.39 | 1.01 | 1 | 1 |
| fft | 253 | 3471 ms | 1.33 | 1.98 | 3.14 | 1.01 | 1 | 1 |
| graphs | 420 | 4062 ms | 1.30 | 1.43 | 6.26 | 1.48 | 1 | 1 |
| lattice | 36 | 1878 ms | 13.27 | 1.25 | 18.61 | 1 | 1 | 1.01 |
| lattice2 | 276 | 5498 ms | 1.68 | 1.18 | 2.41 | 1.15 | 1 | 1 |
| maze | 69 | 408 ms | 13.76 | 1.31 | 71.88 | 1.50 | 1 | 1 |
| maze2 | 337 | 4984 ms | 2.69 | 2.56 | 14.11 | 3.13 | 1 | 1 |
| mazefun | 283 | 8247 ms | 1.89 | 1.89 | 6.15 | 1.46 | 1 | 1 |
| nboyer | 410 | 2362 ms | 1.53 | 2.87 | 3.32 | 1 | 1 | 1.37 |
| nestedloop | 258 | 7804 ms | 1.21 | 3.55 | 1.37 | 1.04 | 1 | 1 |
| nfa | 208 | 5801 ms | 1.26 | 1.22 | 1 | 1 | 1.83 | 1.31 |
| nothing | 84 | 0 ms | 2.47 | 1 | 1 | 1 | 3.14 | 1 |
| nqueens | 239 | 4670 ms | 1.07 | 1.89 | 1 | 1.14 | 1.04 | 1 |
| nucleic2 | 1305 | 10293 ms | 2.09 | 1.07 | 44.29 | 1.27 | 1 | 1 |
| nucleic3 | 663 | 9736 ms | 0 | - | 0 | - | 1 | 1 |
| paraffins | 242 | 3271 ms | 1.85 | 5.16 | 14.22 | 1 | 1 | 2.22 |
| peval | 102 | 288 ms | 16.39 | 1.94 | 68.41 | 1 | 1 | 1.27 |
| puzzle | 269 | 5572 ms | 1.92 | 3.08 | 4.20 | 1.08 | 1 | 1 |
| ray | 502 | 6448 ms | 0 | - | 0 | - | 1 | 1 |
| sboyer | 342 | 3476 ms | 1.83 | 2.43 | 3.98 | 1 | 1 | 1.21 |
| scheme | 137 | 194 ms | 14.28 | 1 | 104.64 | 1.47 | 1 | 1.22 |
| scheme-c | 338 | 527 ms | 3.50 | 1 | 8.23 | 2.95 | 1 | 1.11 |
| scheme-i | 320 | 843 ms | 1.24 | 1 | 2.06 | 1.33 | 1 | 1.45 |
| scheme2 | 434 | 194 ms | 4.78 | 1 | 35.37 | 1.49 | 1 | 1.06 |
| sort1 | 38 | 900 ms | 8.13 | 1.32 | 8.94 | 1.77 | 1 | 1 |
| tak | 171 | 5934 ms | 1.33 | 3.03 | 1 | 1.00 | 1.44 | 1 |
| takl | 208 | 4452 ms | 1.30 | 1.64 | 1 | 1 | 1.46 | 1.34 |
| takr | 533 | 2503 ms | 3.13 | 1.11 | 98.54 | 1 | 1 | 1.88 |
| takr2 | 425 | 2183 ms | 0 | - | 120.30 | 1 | 1 | 2.38 |
| triangle | 327 | 5017 ms | 1 | 2.65 | 1.24 | 1.29 | 1.19 | 1 |
____________________
Racket Users list:
http://lists.racket-lang.org/users
