nlepage
commented
5 years ago Improved the benchmark "framework":
- First a starting phase during a 1/100 of the maximum time, with only 1 iteration, then 2, then 4, ... gives a rough estimation of the mean time of 1 iteration
- Then the real measurement with runs of ~ 1/10 of the maximum time
New results:
| ES2015 destructuring | immutable 4.0.0-rc.12 | seamless-immutable 7.1.3 | immer 1.8.0 | qim 0.0.52 | immutad●t 2.0.0 | qim 0.0.52 curried | immutad●t 2.0.0 curried | |
|---|---|---|---|---|---|---|---|---|
| Set a property | 6265665ops/s (160ns/op) | 168348ops/s (5.94µs/op) | 65270ops/s (15.3µs/op) | 215294ops/s (4.64µs/op) | 781006ops/s (1.28µs/op) | 943753ops/s (1.06µs/op) | 754148ops/s (1.33µs/op) | 874432ops/s (1.14µs/op) |
| Set a deeply nested property | 958405ops/s (1.04µs/op) | 23181ops/s (43.1µs/op) | 5906ops/s (169µs/op) | 18939ops/s (52.8µs/op) | 120080ops/s (8.33µs/op) | 74560ops/s (13.4µs/op) | 118606ops/s (8.43µs/op) | 74787ops/s (13.4µs/op) |
| Update small todos list (1000 items) | 74076ops/s (13.5µs/op) | 1856ops/s (539µs/op) | 506ops/s (1.97ms/op) | 2825ops/s (354µs/op) | 15919ops/s (62.8µs/op) | 24051ops/s (41.6µs/op) | 16011ops/s (62.5µs/op) | 22800ops/s (43.9µs/op) |
| Update medium todos list (10000 items) | 7155ops/s (140µs/op) | 182ops/s (5.48ms/op) | 51ops/s (19.5ms/op) | 283ops/s (3.53ms/op) | 1698ops/s (589µs/op) | 3087ops/s (324µs/op) | 1706ops/s (586µs/op) | 3036ops/s (329µs/op) |
| Update large todos list (100000 items) | 266ops/s (3.77ms/op) | 18ops/s (55.6ms/op) | 5ops/s (204ms/op) | 27ops/s (37.7ms/op) | 127ops/s (7.87ms/op) | 177ops/s (5.66ms/op) | 128ops/s (7.84ms/op) | 189ops/s (5.30ms/op) |
| Final score | 100 | 3 | 1 | 5 | 24 | 33 | 24 | 33 |
Codecov Report
Continue to review full report at Codecov.