Kangaroo Twelve Implementation in Go
github.com
Kangaroo Twelve Implementation in Go
1–10 of 30 posts
Re: Kangaroo Twelve Implementation in Go
#2A mention of the always-on culture of these times seemed appropriate. Easy access to information doesn't seem beneficial if said information is geared to attract clicks or foment fear. Or both.
Re: Kangaroo Twelve Implementation in Go
#3Isn't this SHA-3 less safe, too? Why not just go with BLAKE2?
Re: Kangaroo Twelve Implementation in Go
#4KangarooTwelve on PicoLisp: https://bitbucket.org/mihailp/tankfeeder/src/5a91a025d78eacf...
Re: Kangaroo Twelve Implementation in Go
#5Isn't this SHA-3 less safe, too? Why not just go with BLAKE2?
BLAKE2 is less safe than BLAKE as well.
>Isn't this SHA-3 less safe, too?
Yes, both K12 and M14 are less safe than SHA3. Instead of 12 + 2ℓ rounds (where ℓ is 25 * 2^ℓ = r + c [where r + c = 1600 in both sha-3 and K12]) it uses just 12 rounds (14 for M14). It also uses half the c (c = 512 in sha3-256, c = 256 in K12, where c is the sponge capacity). It provides the same amount of bits of security against collision attacks for both quantum and classical systems, however the preimage security is the ~85 bits for quantum systems (128 in sha3-256) and 128 bits in classical systems (256 in sha3-256) due to the changes in c.
Re: Kangaroo Twelve Implementation in Go
#6(from linked resources)
> Keccak won the SHA-3 competition, and became the FIPS 202 standard on August 5, 2015. All the other contenders lost.
I don't think many professionals would agree with this view of cryptographic competitions.
> Being too late to take part in the fight to become SHA-3, it just stood on its own awkwardly.
Again with the wording... SHA-3 is not an epic battle for blood and honour. Acting as if a NIST competition is a public fight is rather disconcerting.
> It is only in 2013 that SHA-2 joined Intel's set of instructions, along with SHA-1...
Technically true, but only now -- four years later -- desktop processors were released supporting the SHA extensions, and none of these are from Intel so far.
Re: Kangaroo Twelve Implementation in Go
#7Isn't this SHA-3 less safe, too? Why not just go with BLAKE2?
BLAKE2 is less safe than BLAKE as well. >Isn't this SHA-3 less safe, too? Yes, both K12 and M14 are less safe than SHA3. Instead of 12 + 2ℓ rounds (where ℓ is 25 * 2^ℓ = r + c [where r + c = 1600 in both sha-3 and K12]) it uses just 12 rounds (14 for M14). It also uses half the c (c = 512 in sha3-256, c = 256 in K12, where c is the sponge capacity). It provides the same amount of bits of security against collision at…
The word "safe" isn't doing us any favors here. What we're really talking about is the "security margin" of the different hashes. Every hash we're talking about on this thread has a margin far beyond any plausible or foreseeable attacks. None are unsafe.
Re: Kangaroo Twelve Implementation in Go
#8K12 is not a "faster SHA-3". K12 is K12, M14 is M14.
K12 is related to SHA-3, but if you're throwing standards out the door (and I think you should), you can use any of the hash core algorithms, whether or not they're related to Keccak.
The best all-purpose cryptographic hash is probably truncated SHA-2 512, and the hipster modern hash is probably still Blake2.
Re: Kangaroo Twelve Implementation in Go
#9K12 is not a "faster SHA-3". K12 is K12, M14 is M14. K12 is related to SHA-3, but if you're throwing standards out the door (and I think you should), you can use any of the hash core algorithms, whether or not they're related to Keccak. The best all-purpose cryptographic hash is probably truncated SHA-2 512, and the hipster modern hash is probably still Blake2.
> The best all-purpose cryptographic hash is probably truncated SHA-2 512
Could you elaborate on that please? Why would you consider it as the "best all-purpose cryptographic hash"?
Re: Kangaroo Twelve Implementation in Go
#10K12 is not a "faster SHA-3". K12 is K12, M14 is M14. K12 is related to SHA-3, but if you're throwing standards out the door (and I think you should), you can use any of the hash core algorithms, whether or not they're related to Keccak. The best all-purpose cryptographic hash is probably truncated SHA-2 512, and the hipster modern hash is probably still Blake2.
> The best all-purpose cryptographic hash is probably truncated SHA-2 512 Could you elaborate on that please? Why would you consider it as the "best all-purpose cryptographic hash"?
It is one most scrutinized hash function, for which no flaw is known. While no flaw is known for SHA-3 and many other hash functions and some might even say that likelihood of finding flaws in them is lesser than SHA-2, none has been analyzed by that many people for that long.