[Cryptography] The best TRNG architecture, comming soon?

[Ryan Carboni]

Low entropy RNGs are interesting because many attacks assume low entropy and small state, while low entropy RNGs provide low entropy but large state (much as the effect of adding an initialization vector to a protocol).

Regardless, given that the world uses always-on devices, fast boot times shouldn’t be prioritized too much (although the timer entropy daemon, video entropy daemon, and the audio entropy daemon prove there are other faster sources of entropy) .

As long as the amount of entropy exceeds the bruteforce cost of other forms of hacking, the cost-benefit ratio will never lean towards directly breaking a cipher. Unless the state is insufficiently mixed.

Sent from ProtonMail Mobile
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.metzdowd.com/pipermail/cryptography/attachments/20190826/53604d7e/attachment.htm>