The benchmarks can be kinda hard to read without formatting, so I added some formatting. I picked en-US since you were using dot for the decimals already, but that part's trivial to swap.
Now
"123456789.123456789 bytes" will read "123,456,789.12 bytes".
"skipped" will still read "skipped"
"13.6.11" will still read "13.6.11"
Caveats:
I didn't rerun any benchmarks. I just reran formatting (npm run table) and pasted the new format into the readme. However, some numbers were actually different. For example, automerge's B1.5 (time). I'm guessing that the readme just hasn't been updated since the benchmarks were last run and committed?
I didn't reformat any but the top benchmark table because I just used what comes out of npm run table, so the old tables are still tricky to read.
dmonad
commented
7 months ago
The benchmarks can be kinda hard to read without formatting, so I added some formatting. I picked en-US since you were using dot for the decimals already, but that part's trivial to swap.
Now
Caveats:
npm run table) and pasted the new format into the readme. However, some numbers were actually different. For example, automerge's B1.5 (time). I'm guessing that the readme just hasn't been updated since the benchmarks were last run and committed?npm run table, so the old tables are still tricky to read.