Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon. Entire thread

But you can't encrypt/decrypt with hash functions!

Name: Anonymous 2017-05-08 16:16

Oh yes I fucking can you anus-mouthed niggerian jew. I can trivially push this shit into counter mode and \(Ciphertext_{counter} = H(key \| nonce \| counter) \oplus Plaintext_{counter}\)

Use your favourite MAC construction instead of H!

Name: Anonymous 2017-05-08 16:24

LATEX Fail!
But seriously, why didn't it work?

Name: Anonymous 2017-05-08 16:57

>>2
Works for me but only when I am on the thread page and wait a while.

Name: Anonymous 2017-05-08 17:18

>>2
Because it's Javashit.

Name: Anonymous 2017-05-09 1:39

Jokes on you, I encrypt my data using little-endian.

Name: Anonymous 2017-05-11 12:31

oboNc ydr na daesihttub m'I yag .

Name: Anonymous 2017-05-12 5:12

>>1
When people talk "encrypting with a hash" they probably mean something retarded like reversing a hash back to source string(somehow avoiding collisions) not counter mode

Name: Anonymous 2017-05-12 5:17

Name: Anonymous 2017-05-13 1:58

>>8
PCLMULLQLQDQ

Name: Anonymous 2017-05-13 9:37

>>8
Oh neat, I didn't know GCM was also proven secure and in the concrete security model no less.

Name: Anonymous 2017-05-19 19:19

Who said you can't?

Name: Anonymous 2017-07-02 15:26

>>10
Most WC-style MACs (such as poly1305 and UMAC) are provably secure.

Name: Anonymous 2017-07-03 3:34

>>1
They mean it's difficult to recover some valid hash input given a specific hash output, for a well built hash

Name: Anonymous 2017-07-03 3:45

Does H(counter||key||nonce) have better properties?

Name: Anonymous 2017-07-03 3:56

some C.S.counter using key x nonce

Name: Anonymous 2017-07-03 15:29

>>14
Does \(H(key \oplus nonce \oplus counter)\) have better properties?

Name: Anonymous 2017-07-03 15:47

>>16
I don't know, you tell me

Name: Anonymous 2017-07-03 16:56

rcrypt uses hash function for key generation
https://www.reddit.com/r/frozenvoid/wiki/void2/random_h
(of course Crypteria11( https://www.reddit.com/r/frozenvoid/wiki/algorithms/encryption/crypteria11h ) is much stronger cryptographically, but is also much slower slower)

Name: Anonymous 2017-07-04 4:21

>>16
Is the counter little-endian?

Name: Anonymous 2017-07-04 6:14

>>18
hey FV I don't have time to cryptanalyze those functions right now but have you made sure that those are actually cryptographically strong? you know, you can use any RNG as a stream cipher but it will usually result in a shitty cipher (e.g. the attacker being able to recover key/RNG state/plaintext)

Name: Anonymous 2017-07-04 6:53

>>20
The only algorithm thats cryptographically strong is Crypteria.
rcrypt has 64bit keyspace and is built for max speed

Name: Anonymous 2017-07-04 7:18

>>21
what do you mean by 'cryptographically strong'? large keyspace is required but it isn't enough. just like checking my dubs[/spoiler]

Name: Anonymous 2017-07-04 8:55

>>22
FAILure in BBCODE

Name: Anonymous 2017-07-04 9:34

>>23
but success in dubs!

Name: Anonymous 2017-07-04 10:58

>>20
//spray bits from data into result array,gap_len 0-63 from xorshift

Name: Anonymous 2017-07-04 11:08

>>22
I think stegano

Name: Anonymous 2017-07-04 11:14

>>26
please elaborate (inb4 definition of stegano - I know what it is, but how is it related to FV's cipher being or not being strong?)

Name: Lain 2017-07-04 13:42

Crypteria as in Cyberia?

Name: Anonymous 2017-07-04 13:50

note:the algorithm is named after Krypteria(the symphonic metal band) not Crypteria(metal band).

Krypteria is a symphonic power metal/gothic metal band from Aachen, Germany.[1] Originating from a 2001 "pop musical theatre studio project" of the same name, the idea was to use different vocalists, though German-Korean singer Ji-In Cho assumed the role of lead vocalist in 2004. One self-titled studio album was released under this moniker (later edited and re-released as Liberatio).[2] The group went on hiatus in 2012 and later reformed as And Then She Came in 2016.[3][4]

Name: Anonymous 2017-07-04 13:50

>>28
It's possible that a weak cipher with strong stegano could be pretty secure

Name: Anonymous 2017-07-04 14:19

>>30
What does strong stegano even mean? If the enemy can't find the payload, why bother with crypto?

Name: Anonymous 2017-07-05 0:21

>>31
crypto can be used for both privacy and integrity, so stegano can provide the privacy but you still want integrity checks probably

Name: Anonymous 2017-07-05 0:34

encrypted data also tends to look random where plaintext doesn't, so you'd need to use similar data for camo

Name: Anonymous 2017-07-05 0:46

a permutation with words wolf decoys of doctor bad consider

Name: Anonymous 2017-07-05 0:52

has 3.6 million variations lol

Name: Anonymous 2017-07-05 1:03

don't forget your change blah keys to lol blah but

Name: Anonymous 2017-07-05 6:10

>>33
not really, modern stegano is usually about inserting any kind of data into a medium without the user being able to notice. strong stegano will do it in a way that isn't detectable through statistical analysis

Name: Anonymous 2017-07-05 13:16

>>32
crypto can be used for both privacy and integrity
Encryption is used only to guarantee privacy. If you care about integrity you would use a WC-style universal MAC or a hash in a mac mode (sha3(key||msg), kmac, hmac, etc).

Name: Anonymous 2017-07-05 13:23

>>38
what is authenticated block cipher mode of operation?

Name: Anonymous 2017-07-05 13:31

It could hide amongst junk-positives

Newer Posts
Don't change these.
Name: Email:
Entire Thread Thread List