Backing up Private Keys

Peter Lebbing peter at
Mon Apr 15 22:14:04 CEST 2013

On 15/04/13 13:33, Ashley Holman wrote:
> But, if I account for moore's law, I should assume that the attempts per 
> second will increase each year.  If I assume 50% increase in computing power 
> per year

Just a very quick answer: this doesn't account for thermodynamics. Other
calculations for instance relate to the energy required to destroy/change a
certain amount of information. There's an information-theoretic limit to that,
and if the models hold it is not possible to flip a bit with less energy expended.

Furthermore, you are now indeed in the realm of almost brute force[1]. Security
of a cipher algorithm is no longer significant. What matters is the amount of
calculations needed for one guess. Off the top of my head I don't think this
depends significantly on the cipher. It depends on the hash algorithm (not it's
strength, it's amount of computation), and --s2k-count.

Also, calculating it depends on a good measure of the entropy of your
passphrase; the 10^25 you mention.

> Does this mean that people in the 22nd / 23rd century might be able to crack
> old passphrased keys really easily?

Dunno. What would a 19th century academic say of my Core 2 Duo PC with a video
card that can spit out nearly 50 milliard textures each second?

Other than that: do you care what people by then do with your stuff?



[1] You're still making educated guesses about the password, right?

I use the GNU Privacy Guard (GnuPG) in combination with Enigmail.
You can send me encrypted mail if you want some privacy.
My key is available at <>

More information about the Gnupg-users mailing list