Benchmarks are interesting:

use strict;
use warnings;
use Set::Scalar;
use Benchmark qw(cmpthese);

my @array = qw(a b c d e f g h i j k l m n o p q r s t u v w x y z);

cmpthese
  (-1,
    {
    'GF' => sub {my %hash; @hash{@array} = (1) x @array;},
    'map' => sub {my %hash = map {($_, 1)} @array;},
    'map++' => sub {my %hash; map {$hash{$_}++} @array;},
    'set' => sub {my $set = Set::Scalar->new(@array);},
    }
 );
[download]

Prints:

         Rate   set   map map++    GF
set    2203/s    --  -84%  -91%  -92%
map   13740/s  524%    --  -46%  -49%
map++ 25598/s 1062%   86%    --   -6%
GF    27113/s 1131%   97%    6%    --
[download]