3 ms·
> what are the tests that can prove or disprove stronger guarantees for cryptographic security of a PRG than diehard? None. The difference between a good CSPRN
by blattimwind 7y ago
> what are the tests that can prove or disprove stronger guarantees for cryptographic security of a PRG than diehard?
None. The difference between a good CSPRNG and a broken one might not even be in the construction at all, but in who knows the seed. For example, a keystream generated using Chacha20 or AES-CTR makes for a good CSPRNG... except if the attacker knows the key.