Re: Virtual Matrix Encryption

From: John M. Dlugosz (john_at_dlugosz.com)
Date: 05/22/03 

Date: 22 May 2003 12:37:22 -0700

"Victor" <victor2000@Hotbox.ru> wrote in message news:<1053614517.840079@fobos.ldc.net>...

> Now I see... Right - it is snake-oil, simply because their algorithm is not
> opened. This is fear and logical enough considering the rules of the modern
> cryptographical world.

No, not at all. They don't know what they are talking about. The
algorithm is known, as it's been reverse-engineered from demos. Not
only has it been lampooned as a silly toy, but their code is pretty
bad, too!

Read the past discussions. It's been broken.

> What concerns me the most is the security in the real-world situation. E.g.
> AES is the standard now, and I have no doubts folks from NSA (or whatever
> other govn. agencies) will analyze or attempt to brake Rijndael until they
> succeed, die or simply another algorithm will be used in the AES. Consider a
> situation when they got your data encrypted with AES and by that time they
> already have broken it.

There was a recient discussion of that here. That is, does the
government have a secret trick to break AES? If you think that they
must, then use double-encryption with AES and GOST. If the US can
read the former and the USSR can read the latter, they would still
have to talk to each other to read your message.

Or, use one of the other AES candidates. They can't =all= have been
hoxes, as opposed to submissions from independant developers not
assocated with the government. Try using Twofish with a "family key"
so even if they have a billion dollar hardware cracker it won't work
with a non-standard "family".

> If you encrypt your data by some of the snake-oil
> products, I guess they would have much more troubles, because:
>
> 1.. They never pay any attention to the underlying snake-oil algorithm,
> and probably won't ever do.
> 2.. Even if they decide to brake it, they would have to get the original
> algorithm first, which leads to the problems with disassembly.
> 3.. Probably after all they would brake this snake-oil, but if it is
> designed with the comparable security strength to e.g. Serpent and/or
> Blowfish, they would spent the same amount of time and resources,

The flaw is in the 3rd item. The things labeled as "snake oil" are
not difficult to break at all. They are thrown together by hucksters
or idiots. To exploit the points you make, use a =good= algorithm
that is not AES. A short-list of "good" ones would be the finalists
for AES. They are probably all just as strong, but have different
tradeoffs with memory usage, speed, key setup time, etc.

> What if snake-oil product is not that bad, and has the same level of
> security then other opened algorithms?

Then it would not be called "snake oil" and would be be "in the
doghouse". There =are= products and low-level code available for
"other" algorithms, and not all of them are snake oil.

> The obvious advantage of such
> algorithms is that they hadn't studied by the whole crypto community and
> especially by the NSA for the long time, and thus it would require much more
> efforts and time to break them.

Not at all: by not being studied by the community, the bugs have not
been shaken out yet and it will be rather easy to break by a
professional. The good stuff goes through lots of iteration in the
planning stage with discussion among other experts, before it even
gets published; then flaws are found by the public. So, if it is
getting serious attention from experts to even look for bugs (rather
than turning up their noses and laughing) you can be assured that
there are no remaining obvious problems or stupid mistakes.

You had the wisdom to ask if this product was any good. It's not. So
don't be mad and think that we don't like it for reasons that don't
matter to your application. We don't like it because it's a pile of
junk. So keep shopping. Feel free to pop in again when you see
something else you like, for an expert opinion. There was an article
in CryptoGram (look around at counterpane.com) on how to spot
snake-oil that might be useful to you.

--John

Relevant Pages

  • Re: Virtual Matrix Encryption
    ... die or simply another algorithm will be used in the AES. ... >> algorithms is that they hadn't studied by the whole crypto community and ... by not being studied by the community, the bugs have not ...
    (sci.crypt)
  • Re: Virtual Matrix Encryption
    ... it's considered "snake oil" because it is promoted like snake oil. ... die or simply another algorithm will be used in the AES. ... If you encrypt your data by some of the snake-oil ...
    (sci.crypt)
  • Re: New Encryption Idea
    ... performing the 5 reads necessary in the example algorithm results in a delay ... Panama at 400MB/sec, or RC4 at about 90MB/sec, or AES in CTR mode at ... and the speed failings of your design become very clear. ... > Manansala Encryption and Authentication System ...
    (sci.crypt)
  • Re: FUD about CGD and GBDE
    ... >government has approved the use of AES with 256 bit keys for very ... that stress on the algorithm and maintain 256 bits of margin. ... I don't seriously think that either of CGD or GBDE will be broken ... The first reason is that it adds complexity. ...
    (freebsd-hackers)
  • Re: Quadruple Algorithms
    ... occurring" (a fatal flaw being found in AES, ... If you really want secure crypto use various layers of encryption ... with the output of one cipher feeding ...
    (sci.crypt)