ryzen-threadripper-3960x_linux_gcc13.3_a10220f
Time per double (lower is better)
| Method | Time (ns) | Speedup |
|---|---|---|
| zmij | 16.24 | |
| xjb64 | 19.62 | |
| yy | 21.40 | |
| dragonbox | 24.76 | |
| uscale | 28.96 | |
| fmt | 31.55 | |
| schubfach | 34.49 | |
| ryu | 43.99 | |
| to_chars | 52.17 | |
| double-conversion | 83.20 | |
| sprintf | 517.68 | |
| ostringstream | 781.02 |
Click any row to use it as the speedup baseline.
Times include a fixed loop-overhead floor of 1.17 ns (measured with a no-op stand-in for dtoa).
ostringstream and sprintf omitted; they are an order of magnitude slower than the rest.
Time vs. digit count (log scale)
Hover or click a method to highlight its series.