Re: Top Secret Crypto 3.70

From: Mack (macckone_at_a_nospamjunk123_ol.com)
Date: 01/01/05


Date: Sat, 01 Jan 2005 19:45:47 GMT

On Fri, 31 Dec 2004 00:27:34 -0800, headcrash <headcrash@platter.com>
wrote:

First I have to say I agree with Tom St. Dennis on his assessment of
the poor code quality. And I agree with headcrash in general. This
is not a product that I would recommend.

[snip]
>>http://www.topsecretcrypto.com
>>mkp@topsecretcrypto.com
>>
>
>I can help with that. It's easy when you have this kind of BS on your
>site to describe the product in jingoistic, non-proven terms:
>
>This paragraph was taken verbatin from your website
>
>"Top Secret Crypto Gold's strength rests on three basic concepts:
>(1) a true source of random bits which is provided by the program
>(2) a very large key space for the pseudo random number generators
>(3) a simple, but elegant, encryption formula. We call this The
>Black-Hole Encryption System. Like a black hole in which nothing can
>escape from, not even light, data encrypted using our system cannot be
>decrypted and extracted without the correct key."
>
>
>OK, let's start with number 1: Bullsh*t - there is not a true random
>source of bits on a deterministic-by-nature PC. Anyone who claims
>differently is a snake oil salesman

This is not strictly true. The method used in the program is the
collection of the TSC or QueryPerformanceCounter. This has
been discussed somewhat in sci.crypt.random. The gist of it is
the random bits are collected from the interrupts and activity
(network, keyboard, mouse, hard drive activities) and put through a
chaos generator (the operating system). Using the low bits of these
counters is pretty effective based on chaos theory. Especially if
they are hashed after an accurate entropy estimate is determined.
So far no one has come up with a way to make a valid entropy
estimate.

The way the program in question uses them is another matter entirely.
The following code snippet is a perfect example.

        while(TRUE)
        {
                GetRandomBits(32,&dwTestNumber);
                if (dwTestNumber >= 100000001)
                {
                        break;
                }
        }

This shows a complete misunderstanding of what random means.
This specifically eliminates some values. Of course these bits
are further manipulated which prevents the output from looking
bad but the method is entirely questionable.

>
>Now on to number 2: Bullsh*t - very large keyspace for the pseudo
>random number generators? What kind of double-speak is that? And
>don't explain what keyspace means as everyone already knows it. A
>well-crafted cipher only needs 128-bits of security. Meritless claims
>of a zillion bits of keyspace are worthless, and the fodder of snake
>oil peddlers.

Agreed.

>
>Hey, we're already at number 3: Bullsh*t - I don't even know where to
>begin in this one, it stinks so much. Black-Hole Encryption System?
>WTF is that supposed to mean? How about your competitor's
>Supermassive Black Hole Encryption System? As everyone (with a bit of
>astro-physics) knows, supermassive black holes have the mass of over a
>billion black holes. Suppermassive black holes eat regular black
>holes. How puny your system looks now. Their system is over a
>billion times better and stronger than yours. Whatever.
>
>And the decription of "simple but elegant". Simple - possibly.
>Elegant - extremely highly unlikely. Everyone before you that has
>spewed the kind gobbledegook that can be found on your website
>describing your nimrod encryption product has turned out to have a
>most inelegant product.
>

Looking at the source code leads me to the conclusion that the
method may be simple but the source code is far from elegant.

>
>The obvious point here is that anyone who foregoes using an
>established algorithm like AES or 3DES or Blowfish or Twofish that are
>available FOR FREE in many reputable products like GNUPG in order to
>pay actual money for an unproven and most likely insecure product like
>yours is <explitive deleted> insane.
>

I agree completely with using standard ciphers.
However the product is free for personal use.

I would recommend against this product unless you believe in
security through obscurity. I was unable to decipher exactly
what the program is supposed to do thanks to the lack of
organization in the source code and odd mixing of assembly
with C

[valid ranting snipped]

>So, in closing, I think that when he said:
>
>;) C 3.70 is a bit more than it seems...
>
>
>He was being much nicer than I'm being, but the nessage was the same,
>which is your product is a bigger bag of snake oil than all get out.
>
>
>Again, the better product to use would be GNUPG
>
>www.GNUPG.com
>
>Free
>
>Known-good algorithms designed by some of the best in the non-black
>crypto-world.
>
>Compatible with PGP
>
>Open, well-tested source
>
>The implementation of GNUPG has been recommended by many of the top
>crypto people. They've looked at its model closely and see that it is
>correctly designed and uses proper security techniques.
>
>And GNUPG doesn't use the snake oil terms "true one time pad" or "true
>source of random bits" or "Black Hole" anywhere in their website or
>documentation.
>
>I can explain it even further for you if this was not sufficient.
>

Leslie 'Mack' McBride
remove text between _ marks to respond via e-mail



Relevant Pages

  • Re: Top Secret Crypto 3.70
    ... >a very large key space for the pseudo random number generators ... >Black-Hole Encryption System. ... >billion black holes. ... Looking at the source code leads me to the conclusion that the ...
    (sci.crypt)
  • Re: Top Secret Crypto 3.70
    ... >a very large key space for the pseudo random number generators ... >Black-Hole Encryption System. ... >billion black holes. ... Looking at the source code leads me to the conclusion that the ...
    (linux.redhat)