Is Ply going to be obsolete?

[JustinWick]

I've been using Ply to great effect on an ultra-low-latency proof of concept, and it's been a joy to use. As of RFC FS-1072 native CE support for task creation is now an official part of F#, albeit on an experimental basis.

Do you believe that this will obsolete F#? Are there any benchmarks that compare Ply performance and the official F# support?

[pbiggar]

I made some benchmarks comparing Ply and F# 6. The results are a little hard to read (and some of the results, eg vtask and uvtask aren't run in F# 6), but the summary is this:

• F# 6 tasks got 15.7k req/s
• The uply builder got 22.6k req/s

So uply is 43% faster. I believe it's because the F# 6 Task has an allocation, and uply doesn't.

[dsyme]

Yup, uply is a masterpiece and this is about what we expect.

My understanding the perf difference lies on the fundamentals of the UPly and Task types, the latter not being under F#'s control but defined in the BCL.

That said uply has subtleties which task doesn't have, summarised on the home page https://github.com/crowded/ply/

@NinoFloris Do you think resumable code will bring any benefits to uply or is it optimal?