"Robert J. Hansen" <r...@sixdemonbag.org> writes: > On Tue, 22 Mar 2011 18:11:37 +0000, Jerome Baum <jer...@jeromebaum.com> > wrote: >> Okay so let's try again. Correct me if I'm wrong on this one, but it >> does make your key weaker _compared with using an algorithm that >> supports 512 bits of hash, all else being equal_, right? > > If such an algorithm existed in GnuPG, then yes. You'd need about RSA-16K > to get your money's worth out of SHA512, though.
Ah, see that's what I was hoping for. So, there is indeed no reason not to use DSA-1024 with SHA-512. Just as there is no reason not to use RSA-4096 with SHA-512. But the OP was talking about RSA-2048 (with any hash), and there is a reason not to use that. I was assuming that the mention of DSA-1024 with SHA-512 was meant as an analogue to RSA-2048. Apparently it wasn't. -- PGP: A0E4 B2D4 94E6 20EE 85BA E45B 63E4 2BD8 C58C 753A PGP: 2C23 EBFF DF1A 840D 2351 F5F5 F25B A03F 2152 36DA
pgp00eLvQMy91.pgp
Description: PGP signature
_______________________________________________ Gnupg-users mailing list Gnupg-users@gnupg.org http://lists.gnupg.org/mailman/listinfo/gnupg-users
