I'm writing a program that needs fast encryption and decryption, but keys are kept for a long time, so key generation time doesn't matter much.
Also, one key is very important while the other keys are just a deterrent to make it harder to do petty hacking, so variable key size may be necessary.
By the way, it'd help lots if it was available as a Lua library.
AES, use libsodium
>2017
>AES
He's right about libsodium though. Don't ever hack together your own crypto.
Oh, I just remembered, it has to be public-private key encryption.
Because I'm using it to tamper-proof data, instead of hide data.
>>61788154
>tamper-proof data
digital signatures etc.
>>61788192
Obviously, but that still needs public-private key encryption.
Bcrypt?
>>61788116
>>61788323
>posts literally no details about it
How fast does it encrypt and decrypt?
Does it support variable key strength?
Can I make digital signature with it?
>>61788103
Just use gpg
>>61788386
You have no idea what you're talking about. What exactly are you doing? Why asymmetric? Are you sure ENCRYPTION is the solution to your problem?
>>61788463
I know what I'm doing, and it absolutely needs a digital signature.
The only question is which algorithm.
>>61788103
Asymmetrical encryption is pretty much 1000 times slower, than symmetrical and rarely supported by hardware.
Usually you use asymmetric encryption to encrypt symmetrical key, which does main encryption.
>>61788513
What, 1000 times slower at generating, or encrypting?
Anyway, it doesn't matter how bad asymmetrical encryption is, unless you can figure out another way to tamper-proof data as it spreads through a peer-to-peer network
>>61788507
If you knew what you were doing, you wouldn't have created this exceedingly stupid thread.
>>61788386
>Wants people to literally do his work for him