Skip to content

Comment on Project Euler is partly back

Comments

The part about salting confuses me. Shouldn't salting remain effective against rainbow tables even when the salting mechanism is known?

crypto101 says that because of how fast GPUs are at computing hashes, even using long per-user salt is now broken and you must use pbkdf2 or better yet, scrypt.

It should, and they just said "the integrity of weak passwords cannot be guaranteed" -- even the best salting cannot prevent attackers performing brute force attacks; it will be slow, but weak passwords may be cracked in reasonable time.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.