I was curious, so I compared the two one-liners (s/// and sprintf), as follows:

#!perl
use strict;
use Benchmark qw( timethese cmpthese ) ;
my $fixlen = 5;
my @words = qw( s so som some somew somewo somewor someword somewords)
+;
my $b = timethese( -5, {
  srchrep => sub{foreach my $word (@words) {
                   $word =~ s/^(.{1,$fixlen}).*$/pack "A$fixlen", $1/e
+;
                }},
  xprintf => sub{foreach my $word (@words) {
                   $word = sprintf "%-$fixlen.${fixlen}s", $word;
                }},
});
cmpthese $b;
__END__
Benchmark: running srchrep, xprintf for at least 5 CPU seconds...
   srchrep:  5 wallclock secs ( 5.54 usr +  0.00 sys =  5.54 CPU) @ 54
+48.01/s (n=30182)
   xprintf:  6 wallclock secs ( 6.76 usr +  0.00 sys =  6.76 CPU) @ 20
+874.11/s (n=141109)
           Rate srchrep xprintf
srchrep  5448/s      --    -74%
xprintf 20874/s    283%      --
[download]

Looks like sprintf wins the speed test, by a large margin.