Anyone into code breaking i.e. decryption?

[FiatKiller]

Several years ago, I had made an encryption program using several passes of different methods, one after the other, and been looking for someone to try to break it since. Anyone interested to trying to decrypt a simple paragraph? I think how well it works is dependant on what keys you choose, so it may vary greatly how effective it is.

[Bush]

Send me a PM I will check it out.

[FiatKiller]

If anyone wants to try, this is a simple paragraph with normal words without tons of Qs & Zs, etc.  lol

https://dl.dropboxusercontent.com/u/40920348/072076072022072075089072.DOC

I tried to think outside the box, and not just do a simple XOR pass etc.  But since I am not a math major, perhaps I missed some simple facts/principles about encryption.

[freigeist]

I have few questions too but about rar files security.

1) Does somebody knows if is it possible to unlock a locked rar file
and change for example a file name or directory name in it without
corrupting it ?

2) Is it possible to crack a password protected (encrypted) rar file if
let say 20 to 30 characters  random password was used to secure it?
If yes how long could the cracking process last on a computer grid?

[mprep]

I have few questions too but about rar files security.

1) Does somebody knows if is it possible to unlock a locked rar file
and change for example a file name or directory name in it without
corrupting it ?

2) Is it possible to crack a password protected (encrypted) rar file if
let say 20 to 30 characters  random password was used to secure it?
If yes how long could the cracking process last on a computer grid?

1) Yes. You just gotta brute force the password with the specific software.
2) Maybe, depend on your luck as the password can be any type. Depends on the characters though. A password with 20 random letters would take a desktop PC about 157 billion years to crack it (to go through every possible value). This was tested amaturely by me on the site howsecureismypassword.net by simply typing 20 random letters. Don't know if it's true but you can test it for yourself if you want.

[FiatKiller]

So no one cracked my file yet? Maybe I've come up with something.  :-D

[freigeist]

I have few questions too but about rar files security.

1) Does somebody knows if is it possible to unlock a locked rar file
and change for example a file name or directory name in it without
corrupting it ?

2) Is it possible to crack a password protected (encrypted) rar file if
let say 20 to 30 characters  random password was used to secure it?
If yes how long could the cracking process last on a computer grid?

1) Yes. You just gotta brute force the password with the specific software.
2) Maybe, depend on your luck as the password can be any type. Depends on the characters though. A password with 20 random letters would take a desktop PC about 157 billion years to crack it (to go through every possible value). This was tested amaturely by me on the site howsecureismypassword.net by simply typing 20 random letters. Don't know if it's true but you can test it for yourself if you want.

Thanks.

For question under point

1) for locked file I thought of an unencrypted file
without password protection that was locked by using rar locking option only which
prevents the user from changing the content or the rar archive.
By using only this option the content is still readable or extractable without the password
but can't be altered by the user in the rar archive itself.
I tried to alter a file name in the archive by opening the archive in text editor and changing the file name
and then saving the file again.
After I have opened it again in the winrar I receved an error that the file is corrupted.

2) so actually there is no threat from a normal pc user but could it be from a corporation or government
which could have a lot of computing power?

[pand70]

If anyone wants to try, this is a simple paragraph with normal words without tons of Qs & Zs, etc.  lol

What's the deal with Qs and Zs? 

[mprep]

I have few questions too but about rar files security.

1) Does somebody knows if is it possible to unlock a locked rar file
and change for example a file name or directory name in it without
corrupting it ?

2) Is it possible to crack a password protected (encrypted) rar file if
let say 20 to 30 characters  random password was used to secure it?
If yes how long could the cracking process last on a computer grid?

1) Yes. You just gotta brute force the password with the specific software.
2) Maybe, depend on your luck as the password can be any type. Depends on the characters though. A password with 20 random letters would take a desktop PC about 157 billion years to crack it (to go through every possible value). This was tested amaturely by me on the site howsecureismypassword.net by simply typing 20 random letters. Don't know if it's true but you can test it for yourself if you want.

Thanks.

For question under point

1) for locked file I thought of an unencrypted file
without password protection that was locked by using rar locking option only which
prevents the user from changing the content or the rar archive.
By using only this option the content is still readable or extractable without the password
but can't be altered by the user in the rar archive itself.
I tried to alter a file name in the archive by opening the archive in text editor and changing the file name
and then saving the file again.
After I have opened it again in the winrar I receved an error that the file is corrupted.

2) so actually there is no threat from a normal pc user but could it be from a corporation or government
which could have a lot of computing power?

I they used a supercomputer which most governments have under control I bet then they would definetly crack much much MUCH faster. It would just require enough computing power which as I mentioned they probably have.

[FiatKiller]

If anyone wants to try, this is a simple paragraph with normal words without tons of Qs & Zs, etc.  lol

What's the deal with Qs and Zs? 

In other words, less common letters that make codebreaking more difficult. The most common letters in english are ETAONRISH (in order). My test file definitely has alot of those.

[b!z]

If anyone wants to try, this is a simple paragraph with normal words without tons of Qs & Zs, etc.  lol

What's the deal with Qs and Zs? 

Uncommon english letters.

[OnkelPaul]

Caveat: I've not looked at your sample data yet, these are just some general observations regarding non-expert cryptography.

You should not conclude that just because a random selection of forum members can't decode your sample, your encryption algorithm is secure.
In the crypto scene, an algorithm is considered secure if the security of messages ONLY depends on secrecy of the encryption keys.
In addition, the algorithm must be practically usable (so no one-time pads which would otherwise provide perfect security) and secure even in the presence of multiple similar messages or known plaintexts (that was what enabled the decryption of german Enigma messages during WWII).

Even if you feel your algorithm is secure it is very likely that cryptography experts could show you its weaknesses if they bothered.

Onkel Paul

[FiatKiller]

Caveat: I've not looked at your sample data yet, these are just some general observations regarding non-expert cryptography.

You should not conclude that just because a random selection of forum members can't decode your sample, your encryption algorithm is secure.
In the crypto scene, an algorithm is considered secure if the security of messages ONLY depends on secrecy of the encryption keys.
In addition, the algorithm must be practically usable (so no one-time pads which would otherwise provide perfect security) and secure even in the presence of multiple similar messages or known plaintexts (that was what enabled the decryption of german Enigma messages during WWII).

Even if you feel your algorithm is secure it is very likely that cryptography experts could show you its weaknesses if they bothered.

Onkel Paul

Well that's what I'd like is for experts to have fun with it. Not looking to market it especially with the laws about exporting cryptographic tools. I used to break codes for fun when I was a kid and even knowing the algorithm I used... I don't think *I* could break it. I even sent it to a Navy guy and never heard back.

I not sure what you mean by the security depending only on the keys. You would need my program or at least the methods I'm using.