`Float64Array` sums about as fast as a plain array (6% gap)

Typed arrays are sold on memory efficiency, but a straight summing loop shows only a 6% speed difference.

Result

Run	plain array	`Float64Array`	Ratio
Run 1	2.047	2.247	0.91×
Run 2	2.067	2.182	0.95×
Run 3	2.119	2.197	0.96×

Median spread: plain array CV 1.5% / `Float64Array` CV 1.3% / Ratio CV 2.4% (computed from the 3 runs)

Conditions

Summing 1,000,000 elements, five times.

Where this number does not apply

Larger inputs, write-heavy workloads, or operations like sort may flip it. We do not treat this 6% as a selling point.

Reproduce

cd bench
node --expose-gc micro-fixed.mjs --out ../data/run.json --samples 21

Run it three times on the same machine; if the spread of median stays under 10%, you should land on the same conclusion. If your numbers disagree, tell us and we will retract.