Benchmarks showing differences btwn this and `implicitly`?

[lloydmeta]

Hi there,

I just found out about this project and I find it very interesting; mostly because I didn't even know there was a runtime cost for using implicitly !

Do you have any benchmarks that you can share comparing imp with implicitly?

Thanks!

[non]

Hi @lloydmeta I do have benchmarks in various places but none here.

I'll leave this issue open and try to add an unpublished subproject with some benchmarks that show the cost.

[DarkDimius]

https://github.com/DarkDimius/imp-bench - here are the benchmarks that suggest that there's no difference.

[lloydmeta]

@DarkDimius Thanks for posting that. I modified your benchmark in an attempt to benchmark a slightly less trivial example of typeclass usage (branch here) and got the same results:

[info] Benchmark                  Mode  Cnt  Score   Error  Units
[info] ImplVsImplicitly.baseline  avgt   30  3.268 ± 0.025  ns/op
[info] ImplVsImplicitly.explicit  avgt   30  3.292 ± 0.036  ns/op
[info] ImplVsImplicitly.imply     avgt   30  3.291 ± 0.039  ns/op

[DarkDimius]

@lloydmeta would you mind if I incorporate your change?

[DarkDimius]

could you also indicate what platform\java-vm did you run on? I'd like to include your results. Thanks

[lloydmeta]

@DarkDimius sure thing, just sent a PR.

I'm on OSX (El Capitan), Oracle Java8 JDK 1.8.0_05 (I know it's s a bit old...)