Re: [Boost-users] Boost or Standard when there is the choice?

Subject: Re: [Boost-users] Boost or Standard when there is the choice?
From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2014-02-16 12:09:55

On 16 Feb 2014 at 13:04, Daniel James wrote:

> On 15 February 2014 21:11, gast128 <gast128_at_[hidden]> wrote:
> >> my experience with Visual Studio where the Dinkumware implementation
> >> almost always outperforms or is similar to a Boost implementation
> >
> > I had one big bad experience with std::unordered_map in the past, see topic
> > 'Performance difference between std::unordered_map and boost::unordered_map'
> > (https://groups.google.com/forum/#!topic/comp.lang.c++.moderated/pFBjDiW6mW0).
> 
> According to Joaquin's recent benchmarks the recent standard
> implementations are generally faster than the boost implementation,
> which IMO is how it should be. I'm looking forward to
> Boost.Unordered's obsolescence.
> 
> http://bannalia.blogspot.co.uk/

It seemed to me that blog says libstdc++ is equally much faster and 
much slower than Boost depending on the test with a huge variance 
(sometimes 4x). libc++ is always slower, sometimes much slower. 
Dinkumware is usually faster than Boost, but not always.

Besides, unordered_map<> while important is not the end of all of a 
STL quality of implementation. A 4x performance difference one can 
usually live with, while a 400x performance less so.

Also, performance isn't everything, memory footprint is also very 
important. It's unfortunate that blog didn't show differences in 
unordered_map<> memory consumption. I'd warrant that Dinkumware's 
unordered_map<> is very, very good on memory footprint. For the other 
STLs, I wouldn't like to guess.

Niall

-- 
Currently unemployed and looking for work in Ireland.
Work Portfolio: http://careers.stackoverflow.com/nialldouglas/