Adds an optimized ListImpl for lists of doubles only.
Math operations happen directly on the data in double[] and no longer need to use Number wrappers or NumberMath. The result is much faster vectorized operations.
Additionally, many of the algorithms (slice, rotate, ect.) have been optimized.
Still working on some proper benchmarks but here are some examples:
I don't plan any additional work here for the time being so I'll go ahead and merge this into v0.4-dev and split any remaining tasks off into other issues/branches
Requires #78
Adds an optimized
ListImpl
for lists of doubles only. Math operations happen directly on the data indouble[]
and no longer need to useNumber
wrappers orNumberMath
. The result is much faster vectorized operations.Additionally, many of the algorithms (slice, rotate, ect.) have been optimized.
Still working on some proper benchmarks but here are some examples: