diff options
Diffstat (limited to 'benchmarks.md')
| -rw-r--r-- | benchmarks.md | 13 |
1 files changed, 10 insertions, 3 deletions
diff --git a/benchmarks.md b/benchmarks.md index b515c73..848c54a 100644 --- a/benchmarks.md +++ b/benchmarks.md @@ -14,8 +14,8 @@ Some benchmarks, run on an ancient Thinkpad running Intel Core 2 Duo at 2GHz. | marked | 1.94 | | **commonmark.js** | 1.93 | | discount | 1.86 | -| **cmark** | 0.36 | -| sundown | 0.34 | +| sundown | 0.33 | +| **cmark** | 0.33 | To run these benchmarks, use `make bench PROG=/path/to/program`. @@ -28,6 +28,13 @@ Markdown sources of all the localizations of the first edition of time is the *difference* between the time to run the program with the benchmark input and the time to run it with no input. (This procedure ensures that implementations in dynamic languages are -not penalized by startup time.) Amedian of ten runs is taken. The +not penalized by startup time.) A median of ten runs is taken. The process is reniced to a high priority so that the system doesn't interrupt runs. + +Note that these benchmarks were done on a 32-bit machine. On a 64-bit +machines, sundown is significantly faster than cmark (0.146s vs 0.237s +on Intel i5/OSX with Clang, 0.130s vs 0.191s on a 64-bit Debian VPS +with GCC). I do not know why the performance difference shows up on +the 64-bit architecture and not the 32-bit, but that is something that +might be investigated. |