Explain why we deviate slightly from the PBKDF2 standard.

This explanation is based off of Ted's site.  Also, fix a comment from
the SHA-1 version.
ok tedu@

src/lib/libutil/bcrypt_pbkdf.3

@@ -1,4 +1,4 @@
-.\"	$OpenBSD: bcrypt_pbkdf.3,v 1.4 2013/06/05 04:01:53 tedu Exp $
+.\"	$OpenBSD: bcrypt_pbkdf.3,v 1.5 2014/09/27 06:40:07 doug Exp $
 .\"
 .\" Copyright (c) 2012 Ted Unangst <tedu@openbsd.org>
 .\"
@@ -14,7 +14,7 @@
 .\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 .\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 .\"
-.Dd $Mdocdate: June 5 2013 $
+.Dd $Mdocdate: September 27 2014 $
 .Dt BCRYPT_PBKDF 3
 .Os
 .Sh NAME
@@ -37,7 +37,7 @@ The salt value should be randomly generated beforehand.
 The repeated hashing is designed to thwart discovery of the key via
 password guessing attacks.
 The higher the number of rounds, the slower each attempt will be.
-.\" A minimum value of at least 1000 is recommended.
+.\" A minimum value of at least 4 is recommended.
 .Sh RETURN VALUES
 The
 .Fn bcrypt_pbkdf
@@ -64,4 +64,6 @@ function returns 0 to indicate success and \-1 for failure.
 .Sh CAVEATS
 This implementation deviates slightly from the PBKDF2 standard by mixing
 output key bits nonlinearly.
+By mixing the output bytes together, we require an attacker to perform
+all of the work without taking any shortcuts.
 .\" .Sh BUGS