AA Performance in Benchmarks

[Laeeth Isharc via Digitalmars-d]

On Thursday, 23 April 2015 at 15:29:37 UTC, Steven Schveighoffer 
wrote:
> On 4/23/15 10:40 AM, Shammah Chancellor wrote:
>
>> 2) I saw a thread awhile ago where someone had posted an 
>> updated
>> implementation of the AA with benchmarks and showed that it was
>> substantially faster in all cases.  I can't find the thread 
>> again --
>> what happened with this work?
>
> There is a PR being discussed right now. Please take a look and 
> chime in:
>
> https://github.com/D-Programming-Language/druntime/pull/1229
>
> It would be interesting to know how the new AA performs in this 
> test.
>
> -Steve

At a slight tangent: is there a way to reserve capacity for an 
associative array?  The obvious way does not seem to work.

I noticed also in a talk at nativecode (possibly by Herb Sutter) 
that C++ gives you the ability to tune almost any parameter of 
the associative array implementation in the standard library.  
(I'm not a C++ guy).  Is this entirely spurious, or would it be 
helpful for D?


Laeeth.