Still vec can be used... differences are minimal:

update: included jmcnamara's code, which is really fast.

#!/usr/bin/perl
use Benchmark qw(cmpthese);

my $buf = chr(1) x 2e6;

cmpthese(-10,
    {
        unpack => sub { # BrowserUk
            join( chr(0), unpack '(A1)*', $buf ) .chr(0)
        },
        substr => sub { # salva
            $out = chr(0) x (length($buf) * 2);
            substr($out, $_ * 2, 1, substr($buf, $_, 1)) for 0..length
+($buf);
            $out;
        },
        vec    => sub { # shmem
            $out = chr(0) x (length($buf) * 2);
            vec($out,$_<<1,8) = vec($buf,$_,8)
                for 0..length($buf)-1;
            $out;
        },
        vec2   => sub { # jmcnamara
             my $out = pack 'v*', unpack 'C*', $buf;
        },
    }
);
__END__
       s/iter    vec unpack substr   vec2
vec      1.44     --    -1%    -3%   -81%
unpack   1.42     1%     --    -2%   -81%
substr   1.40     3%     2%     --   -80%
vec2    0.275   423%   417%   408%     --
[download]

But your unpack routine doesn't seem to work correctly.

update2: added jmcnamara's code to the stuff inside readmore tags

But your unpack routine doesn't seem to work correctly.

It does:

$s = 'a' x 10; 
print join( 'B', unpack '(A1)*', $s ) . 'B';;
aBaBaBaBaBaBaBaBaBaB
[download]

But not with nulls. The template would have to be '(a1)*' in that case.

---

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice.

"I'd rather go naked than blow up my ass"