PDA

View Full Version : Cryptography ( As posted by EtErNaL_L0ser)


Woodmann
October 18th, 2001, 18:30
Why dont we have a forum dedicated to Cryptography?To discuss DES MD5 El-Gamal RSA PGP and others....

Bengaly
October 19th, 2001, 06:47
Sound cool, but 80% of the ppl won't understand
The topic ;-) unless it's interested in some way or
some ppl actually learn it :-)
but i'd say..GO FOR IT :-)
bye

EtErNaL_L0ser
October 19th, 2001, 10:38
I am reading Applied Cryptography and its software implementation seems to be fun and challanging

Kythen
October 19th, 2001, 17:28
I'm all for it! I'd also speculate you'd get a few more advanced RE ppl to crawl out of the shadows for such a forum, as it's one of the primary areas of research lately. It's good to have someplace for the more advanced RE enthusiasts to play. There's a spot for newbies, and one for general RE topics, but not much for advanced things... and especially crypto.

Cheers!

DakienDX
October 20th, 2001, 05:39
Hello everybody !

Sounds good

But we must define what belongs to the forum and what not.
I think we wouldn't like posts like "how does asprotect crypto work?"

We must define if only known cryptography belongs into the forum or if we can discuss algorithms which seem to be real cryptography, but are only XOR/ROR/ROL/ADD/SUB algorithms in the end. (I know real cryptography uses this too, but in an other way)

We must also prevent posts like "Can please somebody bruteforce RSA-xxx in program yyy for me?"

If all these things are checked, then why not a crypto board?

EtErNaL_L0ser
October 20th, 2001, 11:28
I am totally agree with u Dakien No Lame Post! And starting with XOR would be really cool

DakienDX
October 20th, 2001, 16:33
Hello EtErNaL_L0ser !

I would like to start with
Code:

@@1:
Lodsb
Xor AL,1
Stosb
Loop @@1
Is this o.k. for the beginning?
Or should e start with something more difficult like adding one ROR?



Hello tsehp !
Hello woodmann !

Who's going to be the moderator/administrator for the crypto board if we get one?

Woodmann
October 20th, 2001, 17:33
A moderator for the Crypto forum huh?

Lets figure out the direction the forum shall take first.
If you are happy with what you have discussed then I will start the forum.

As always, use common sense, (I know that you all will),
and everything will be wonderful.

Oh, post some gudelines to include on the front page.

Peace, Woodmann

EtErNaL_L0ser
October 21st, 2001, 04:34
Dakien I am real newbie in software implementation f cryptography even XOR lol I better watch the forum and learn we can start with reading in www.counterpane.com . Really good essays


And who will be the moderator? Woodman +tsehp or another +HCU strainer would be ok I think

Out of the topic I wonder who voted for No! lol

DakienDX
October 23rd, 2001, 11:55
Hello woodmann !

I agree with you. We must have everything discussed clearly before starting a new forum.

I see 19 "yes" and one "no" in the poll. But only 9 replies (three from me, one from you). So I think we should hear more voices on the topic before deciding about if (or if not) we should create a new forum.

A reply from someone being familiar with cryptography would be a nice addition to this poll.

Final question: What can we use the forum for?
I think no one can explain why RSA works or why SHA was converted to SHA-1.
All other crypto algorithms are well documented and commented by cryptographical specialists.

So the question: Not "if" or "if not", just WHY

Woodmann
October 23rd, 2001, 13:55
DakienDX,
You raise a very good point.

I shall leave it too you all, what reason do we start this forum?

I don't see any restrictions. You can discuss anything related to crypto.
Of course I would like to see how it works and how, if possible, it can be reversed/defeated.

Peace, Woodmann

McCodEMaN
October 23rd, 2001, 15:52
Greetings!

I belive that this is an excellent idea!
Over the years I've practised RCE the relationship with cryptography has grown, and I belive that you all feel the same.
Today the knowledge and understanding of cryptography tend to have a major impact on both RCE and computer science in general.

So why should we start this forum when they already exist elseware?

The answer is simple and I think we can come up with many good reasons. I for one hope that that this forum (if created) will help ppl. to exchange ideas on how to apply the knowledge on RCE tasks, along with cryptography in general.

Best Regards
\\McCodEMaN [TRES2000]

EtErNaL_L0ser
October 26th, 2001, 02:57
Although I am the owner of this suggestion I think we should close this topic because topic has more than 200 views but only 22 votes.This s**.Even +tsehp didnt make a little comment on subject .So we better close this topic

Best Regards

CARTMAN aka DaWsOn aka EtErNaL_L0ser

DakienDX
October 26th, 2001, 13:18
Hello everybody !

I can't vote again, but here's my vote no. 2: "NO"

I think EtErNaL_L0ser is right, we should close this topic.

270 views and 22 votes is nothing unusual, and tsehp may be busy at the moment (I can only find one post in the last 11 days from him), but this discussion arrived a dead point.

Cryptography may be discussed as before on the RCE General, because is has to do with general reversing (most of the time )

EtErNaL_L0ser
October 27th, 2001, 16:59
Great to hear that!

DakienDX
October 28th, 2001, 02:23
Hello everybody !

Well, ok, start a new forum. If it's blessed by tsehp, it shoud become true. Here's my bless. (as if anyone woud care about that )

Can someone correct the "\n" in the "Mini Project Area" 's forum description and the "with a 500k limit" in "Reverse Software Engineering General" 's forum description on the way to the new forum?

DakienDX
October 28th, 2001, 11:25
Hello ArthaXerXes !

Thank you.

It was nothing special, but it was anoying when looking at it every time going to the board.

tsehp
October 28th, 2001, 13:07
1- I'm not tired ;-) (artha likes to bite me here, because he's 10 years younger than me, but even ten years couldn't maybe corrupt a great amount of grey matter I built since those long and tiring rce years (3 in total,whaou !)

A cryptology forum is badly needed, this feature is really not well covered on our archives and really deserves some comments,
me thinks that we should directly start it without moderator at first, then we can naturally choose the guy that will participate the most.

So it's just created, just tell me if the name deserves to be changed.

LONG LIFE TO THE Rce cryptographics !!!

DakienDX
October 28th, 2001, 13:43
Hello tsehp !

LONG LIFE TO THE Rce cryptographics !!!

You should now change all forum descriptions to "flames, crack, serial and bruteforce requests will be removed"

EtErNaL_L0ser
October 28th, 2001, 15:05
Yeah no lame post! But some resources

McCodEMaN
October 28th, 2001, 19:55
Greetings everyone!

I'm glad to see that a forum finally was created for such a important and thrilling subject, as I belive that this is!
I hope to see many interesting posts in the future , and perhaps even share some of my ideas.

Regards
McCodEMaN

mike
October 29th, 2001, 13:23
Quote:
Originally posted by DakienDX
...
I think no one can explain why RSA works or why SHA was converted to SHA-1.
...


I can do both.

First is Fermat's Little Theorem:

a^(p-1) mod p = 1, where p is prime, ^ is exponentiation, and a is any number less than p. So

a^b mod p = a^(b mod p-1) mod p.

That implies that for n=p1 * p2 * ... * pk, the product of kdistinct primes, that a^(p1-1)(p2-1)...(pk-1) = 1. This product in the exponent is usually called phi(n).

So a^b mod n = a^(b mod phi(n)) mod n.

So pick two large primes p and q, and multiply them to get n. Then phi(n) = (p-1)(q-1).

Now, you want a two step process, encrypt then decrypt, so that you get the same number back. Well, a^1=a, obviously, so you need two exponents whose product equals 1 mod phi(n).

Pick e at random, and since you know phi(n), you can calculate

d = e^-1 mod n

Publish n and e as your public key. In order to break the code, you need to find d. You could calculate it if you knew (p-1)(q-1), but the only way anyone knows how to get that from n is to factor it. This is very hard.

Then to encrypt your message M, you have

C=M^e mod n

and to decrypt you have

C^d mod n = M^ed mod n = M^(ed mod phi(n)) mod n
= M^1 mod n = M.

Voila!

SHA was probably changed to SHA-1 to avoid an attack that works somewhat like differential cryptanalysis. I say probably because a guy later found an attack that works on SHA-0 but not on SHA-1. If anyone is interested, I can post the reference.

EtErNaL_L0ser
October 29th, 2001, 14:29
Leonhard Euler and Pierre de Fermat both prime-number guys! They got great theorems

One mus know them all!

Sphinx
November 12th, 2001, 08:06
hello you all

well i would say get this topic of the ground, i`m very in too this
i sure want to learn more on this. I have been searching on the web and collection info but as said before you have to have a background to fall back on becuz it`s very very hard. The math involved ain`t easy.

also rv it is difficult becuz lot of cyphers work on the idea big key so hard to brute force. It takes time sometimes lots of it. math is involved to calc the time needed to crypt and decrypt it.


i`m checking this thread often now

Sphinx

Sphinx
November 12th, 2001, 08:15
maybe you can look here h**p://www.claymath.org/prizeproblems/index.htm
these are some big math problems unsolved (yes you can win money) but my point is some off these problems are used in cryptography cuz No one is yet able to use a function to get to an answer in stead you`ll have to go by all possibilties takes time a lot of time just think of prime numbers we can generate them but we can`t say

function p(x) = prime (p is a function not yet discoverd)
say i want the 100th prime number fill in x=100 and you get your 100th prime number

the p versus np problem is something like that too