Closed marcins closed 7 years ago
Performance on my previous benchmark suite (see comments on #21) is slightly faster than Immutable, uses slightly less memory than Icepick, and doesn't suffer from the same weird spike half way through the run. Really though all the options are very close together.
Hard to see from the overlapping graph but actual runtimes were:
I need to:
I'm going to run some quick smoke tests in the browser on these changes before merging - want to make sure I haven't accidentally introduced something that works in the Node tests but not in a browser.
This fixes #21 - replacing the use of ImmutableJS with native JavaScript, while maintaining immutability.
Note that Immutable is still a devDep, because I kept the test that had an Immutable reducer to ensure backwards compatibility in that regard.