Timing test of GnuPG and old PGP

Hunter, Bryan Bryan_Hunter at archway.com
Tue Nov 11 15:51:24 CET 2003

Thank you all for your help and interest.  The PGP was using IDEA and the
GnuPG was using Triple-DES.  Retesting with CAST5 yielded similar timings.

This brings up an interesting question (at least to me).  PGP carries the
symmetric cipher type with the public key.  Switching from PGP to GnuPG
while maintaining transparency to the recipient would seemingly require IDEA
for those who have specified it.  Being in the USA, this would presumably
require a license?  Are licenses available?

Please excuse the delay in reporting back on my original query.
Bryan Hunter

-----Original Message-----
From: Ryan Malayter [mailto:rmalayter at bai.org]
Sent: Thursday, October 30, 2003 6:36 PM
To: Hunter, Bryan; gnupg-users at gnupg.org
Subject: RE: Timing test of GnuPG and old PGP

> From: Hunter, Bryan
> Subject: Timing test of GnuPG and old PGP

> A timing of a simple encryption of a large (327MB) file 
> showed that GnuPG
> took 2 1/2 times longer than an old PGP program.  Is this an expected
> typical result?

Maybe. What version was the "old" PGP? What symmetric cipher did you use
with each program? Perhaps GnuPG used Triple-DES, which is much slower
than the IDEA and CAST algorithms that were the default in older PGP
programs. You can change your kep preferences to use AES, CAST, or some
other speedy cipher with GnuPG.

I assume you eliminated the obvious possibilities for the slowness:
other system activity, file system fragmentation, etc.


Gnupg-users mailing list
Gnupg-users at gnupg.org

More information about the Gnupg-users mailing list