test: using initialized array and for..of index

[facugon]

No description provided.

[facugon]

Hi ,

Thank you for your work! It is exactly what I was looking for.

I was playing around with your benchmarks and I made a few changes in the map benchmarks in a way I think are fair comparison and I found out that using an initialized array in the for .. of gives very different results and I wanted to share the results of my discovery with you.

Perhaps using indexed arrays with a for .. of / for .. in add more lines of code and is kind of weird but the results are relevant.

I was unaware of the huge differences in performance of using initialized arrays against not using it.

My setup:

Ubuntu 20.04
Node 16.18.1

Regards

[facugon]

initialized array

Without

Benchmarking map:
Array.map x 33.02 ops/sec ±1.05% (57 runs sampled)
Array.map, destructuring x 33.27 ops/sec ±0.84% (58 runs sampled)
for of x 38.05 ops/sec ±10.96% (51 runs sampled)
for of, destructuring x 38.10 ops/sec ±11.71% (50 runs sampled)
for, init array x 70.51 ops/sec ±0.90% (72 runs sampled)
for, init array is faster

With

Benchmarking map:
Array.map x 33.43 ops/sec ±0.79% (58 runs sampled)
Array.map, destructuring x 32.79 ops/sec ±0.75% (57 runs sampled)
for of x 106 ops/sec ±0.48% (76 runs sampled)
for of, destructuring x 106 ops/sec ±0.56% (76 runs sampled)
for, init array x 68.66 ops/sec ±3.21% (71 runs sampled)
for of,for of, destructuring is faster