a simple demo for bench float/double to decimal and string algorithm.
| algorithm | Description |
|---|---|
| schubfach | Raffaello Giulietti's algorithm.Origin source code from https://github.com/c4f7fcce9cb06515/Schubfach |
| schubfach_xjb | It is improved by schubfach and has the same output result. |
| ryu | Ulf Adams's Ryū algorithm. |
| dragonbox | Junekey Jeon's CPP implementation. |
| fmt | Victor Zverovich.https://github.com/fmtlib/fmt. version:12.1.0 |
| yy_double | yy's algorithm , origin link : https://github.com/ibireme/c_numconv_benchmark/blob/master/vendor/yy_double/yy_double.c |
| yy_json | yy's algorithm. https://github.com/ibireme/yyjson |
| teju | youtube link : https://www.youtube.com/watch?v=fPZ1ZdA7Iwc . An academic paper will be written to provide proof of correctness.https://github.com/cassioneri/teju_jagua |
| xjb | XiangJunBo's algorithm. https://github.com/xjb714/xjb |
(1) run :
makefile variable settings:
float/double to decimal : set variable BENCH_F_TO_STR = 0
float/double to string : set variable BENCH_F_TO_STR = 1
benchmark double : set variable BENCH_DOUBLE = 1
benchmark float : set variable BENCH_DOUBLE = 0
compile and exec :
gcc: make g or make
icpx:make i
clang:make c
(2) benchmark results :
CPU : AMD R7 7840H , OS : ubuntu 24.04
compiler : (1)gcc13.3 (2)icpx 2025.0.4 (3)clang 18.1.3
CPU : Apple M1 , OS : MacOS 26.1
compiler : apple clang 17.0.0
author : Xiang Jun Bo (xjb) , China , email : [email protected] date : 2025.5.13