Announcement

Collapse
No announcement yet.

Wouldn't It Be Nice If MD5 Was A Two Way Street?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • IDN
    replied
    Sorry i got the terminology wrong

    Leave a comment:


  • okrogius
    replied
    Originally posted by IDN
    What would be the point of MD5 encryption, if you can decrypt it?
    It's NOT encryption, it's a hashing algorithm. And the point is an easy way to verify that data hasn't been altered with.

    Leave a comment:


  • Xenon
    replied
    Originally posted by filburt1
    BTW, it's actually 1 in 16^32 because it's a 32-char hex string, so my mistake, it's a less ungodly 1 in 18,446,744,073,709,551,616.
    if you would just have looked into my post, you could have seen that value

    Leave a comment:


  • IDN
    replied
    What would be the point of MD5 encryption, if you can decrypt it?

    Leave a comment:


  • Chen
    replied
    Originally posted by Mark Hewitt
    Not all. At the very simlest, Ceasar Cypher has exactly the same length. Or text which is run length encoded can be less. etc.
    With all due respect to Caesar... I'm talking about "real" encryption methods.

    Leave a comment:


  • Floris
    replied
    Originally posted by Chen
    Wouldn't it be nice if I could have a dollar every time someone asked how to "decrypt" MD5'd strings?
    You could earn 5$ for every successfull brute-force at least!

    Leave a comment:


  • Mark Hewitt
    replied
    Originally posted by Chen
    The problem is that most (all?) encryption algorithms inflate the data size...
    Not all. At the very simlest, Ceasar Cypher has exactly the same length. Or text which is run length encoded can be less. etc.

    Leave a comment:


  • filburt1
    replied
    BTW, it's actually 1 in 16^32 because it's a 32-char hex string, so my mistake, it's a less ungodly 1 in 18,446,744,073,709,551,616.

    Leave a comment:


  • NetherChris
    replied
    I'd say that's highly unlikely... hahahaha. I wonder if any have ever been discovered. That would be truly interesting.

    Leave a comment:


  • filburt1
    replied
    Originally posted by NetherChris
    Not to mention it is theoretically possible (yet HIGHLY unlikely) to get the same MD5 on two entirely different data items.
    1 in 6.3340286662973277706162286946812e+49, to be specific.

    Leave a comment:


  • DirectPixel
    replied
    Originally posted by xCRYINGoutLOUDx
    Bah.. use mcrypt_encrypt and mcrypt_decrypt.... a little more work, but so much better then just using md5.

    Plus it's real encryption
    Why not have a TRUELY secure system? Beam messages to people in photons.

    Leave a comment:


  • NetherChris
    replied
    Not to mention it is theoretically possible (yet HIGHLY unlikely) to get the same MD5 on two entirely different data items.

    Leave a comment:


  • Xenon
    replied
    right it would be really nice, because of the compression rate, but mathematically impossible

    a 32 char string (or more exactly a 32 number hex value) can just differ between 3.4 * 10^38 different texts, but there are much more possible combinations on a 14 page text

    so a nice dream, but not realistic

    Leave a comment:


  • CeleronXT
    replied
    Originally posted by Grumble
    When you md5() something you arn't "compressing" it or "encrypting" it. Md5 is a hashing algorithm, more akin to a fingerprint of the given input than any converted version of it.

    That is why md5 cannot be a two-way-street. Becuase it doesn't actually contain the original input.
    Eh? Well... uh.. same thing.

    Leave a comment:


  • Chen
    replied
    Originally posted by xCRYINGoutLOUDx
    Bah.. use mcrypt_encrypt and mcrypt_decrypt.... a little more work, but so much better then just using md5.

    Plus it's real encryption
    The problem is that most (all?) encryption algorithms inflate the data size...

    Leave a comment:

widgetinstance 262 (Related Topics) skipped due to lack of content & hide_module_if_empty option.
Working...
X