- Speed, compile time and code bloat tests from tinyformat.
int-benchmark: decimal integer to string conversion benchmark from Boost Karmaitoa-benchmark: decimal integer to string conversion benchmark by Milo Yip. See src/itoa-benchmark/readme.md.
Building and running int-benchmark:
cmake . make ./int-benchmark
Alternatively (requires ninja):
cmake -G Ninja . ninja int-benchmark ./int-benchmark
Sample results on macOS with clang and libc++:
2020-06-25 18:35:12 Running ./int-benchmark Run on (8 X 2800 MHz CPU s) CPU Caches: L1 Data 32K (x4) L1 Instruction 32K (x4) L2 Unified 262K (x4) L3 Unified 8388K (x1) Load Average: 3.05, 2.27, 2.05 -------------------------------------------------------------------------------- Benchmark Time CPU Iterations UserCounters... -------------------------------------------------------------------------------- sprintf 62015560 ns 61939300 ns 10 items_per_second=16.1448M/s std_ostringstream 153347073 ns 153167000 ns 4 items_per_second=6.52882M/s std_to_string 15058924 ns 15052045 ns 44 items_per_second=66.4362M/s std_to_chars 10995660 ns 10991238 ns 63 items_per_second=90.9816M/s fmt_to_string 11909200 ns 11905632 ns 57 items_per_second=83.9939M/s fmt_format_runtime 17071781 ns 17062878 ns 41 items_per_second=58.6068M/s fmt_format_compile 11897635 ns 11893517 ns 58 items_per_second=84.0794M/s fmt_format_to_runtime 13539784 ns 13534137 ns 51 items_per_second=73.8872M/s fmt_format_to_compile 8941199 ns 8937675 ns 77 items_per_second=111.886M/s fmt_format_int 8721323 ns 8718253 ns 79 items_per_second=114.702M/s boost_lexical_cast 29685237 ns 29668455 ns 22 items_per_second=33.7058M/s boost_format 315875676 ns 315739000 ns 2 items_per_second=3.16717M/s boost_karma_generate 12138668 ns 12134518 ns 56 items_per_second=82.4095M/s voigt_itostr 20177849 ns 19994243 ns 37 items_per_second=50.0144M/s u2985907 9445288 ns 9427174 ns 69 items_per_second=106.076M/s decimal_from 13009338 ns 12952232 ns 56 items_per_second=77.2068M/s stout_ltoa 39336210 ns 39325000 ns 18 items_per_second=25.4291M/s