sub f {
    my $l = '';
    my $u;
    for (sort split(//,$_[0])) {
        $u++,$l=$_ if ($_ ne $l);
    }
    return $u;
}
[download]

Update: I posted this because it was faster than a hash based solution but OM_Zen's idea with rob_au's adjustment is even faster. I'd bench merlyn's Inline::C version, but I don't have Inline::C installed...

---
print map { my ($m)=1<<hex($_)&11?' ':'';
$m.=substr('AHJPacehklnorstu',hex($_),1) }
split //,'2fde0abe76c36c914586c';
[download]