Compare ArraySeq (fka Spandex) with Vector using macrobenchmarks

[julienrf]

This might give us new insights about the general performance of ArraySeq

[julienrf]

Idea of macro benchmark: https://en.wikipedia.org/wiki/Quickhull

[Ichoran]

Do you mean that we should write a Quickhull algorithm and use its performance as a benchmark?

[julienrf]

Yes, that’s the idea. What do you think?

[odd]

I think it is a great idea. I have been working on an updated version of ArraySeq (f.k.a. Spandex) that is fully specialized. It can be thought of as a hybrid between ImmutableArray (which it behaves as for all non primitive types) and my previous work on Spandex (which it behaves as for all primitive types). I hope to have something ready for review in a couple of weeks (I’m currently working on making the views specialized as well). Hopefully the resulting collection will be performant for both primitive and non primitive types (with an edge towards the former). Naturally, any additional or improved benchmarks (especially complex/composite ones) would be quite beneficial during the evaluation phase. Greetings Odd Odd Möller odd.moller@gmail.com

…

On Mar 2, 2018, 09:12 +0100, Julien Richard-Foy ***@***.***>, wrote: Yes, that’s the idea. What do you think? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[szeiger]

Assigning to backlog to reconsider for 2.14

[NthPortal]

We need a new name for this collection since ArraySeq is taken now.