[Cryptography] cryptography Digest, Vol 32, Issue 26

[CANNON NATHANIEL CIOTA]

>> Will I be twice as
>> safe with 2048-bit keys as
>> with 1024-bit keys?
> 
> 
> Nope.  Yous keys will be an extra 2^1024 bits harder to crack, but that
> doesn't mean you'll be safer because the weak links are all elsewhere 
> in
> typical systems.  2048 is the current recommended length, but bear in
> mind that although we've recommended 2048, it is one of 10000 details
> which we have no recommendations over.... "Trust us, we're doctors"
> 
> 
> 
> iang
> 
> 


In asymmetric cryptology the bit length of the key is not equivalent to 
the equivalent bits of security it offers.
As the key length doubles in bits it offers better protection and more 
bits of security. Though the security strength does not double like the 
keysize. As for recommended length, 4096 is better than 2048 if the 
option is there (not saying that 2048 is insecure). With RSA the key 
size is huge compared to the bit length of security that key offers. 
With elliptic curve cryptography (ECC) the key size is so much smaller 
and compact when compared to RSA for the same amount of security it 
offers. This is why I am excited about ECC support coming to Gnupg.

-- 
Cannon N. Ciota
Email: cannon at cannon-ciota.info
PGP Fingerprint: E7FB 0605 1BD4 8B88 B7BC 91A4 7DF7 76C7 25A6 AEE2