Re: advice sought on key/data histogram analysis of rijndael/128 and serpent

From: Bryan Olson (fakeaddress_at_nowhere.org)
Date: 10/24/05 

Date: Mon, 24 Oct 2005 17:29:34 GMT

In his second reply to a post of mine, lkcl wrote:
>>Then your data is way-far out from the expected.
>
> _tell_ me about it!!!!

About 1 in 100 trials should have p <= 0.01, and you are scoring
over 5000 of 204800 at that level. That is over 65 standard
deviations above the mean.

> now demonstrate to me where the flaw is,
> in the source code or in my reasoning.

Been there; done that.

Your code is buggy and unreadable, your graphs are ill-defined
to the point of nonsense, and by your own account, your lack
mathematical rigor.

> this should _not_ be happening.
>
> it should _not_ be possible in any encryption
> algorithm to simply count the 1s and 0s of
> an encryption output and be able to draw
> conclusions about its input.

Which is probably why you've done no such thing. 

-- 
--Bryan

Relevant Pages

  • REPOST: Re: advice sought on key/data histogram analysis of rijndael/128
    ... In his second reply to a post of mine, lkcl wrote: ... That is over 65 standard ... deviations above the mean. ... > in the source code or in my reasoning. ...
    (sci.crypt)
  • Re: CFAN (was: A small wildcard matching algorithm)
    ... of source code into a single file and run it like it's a bunch of files ... We have a standard for files, ... first file, a virtual file access table, and which then let you access ... value to doing it independent of the OSes' file systems, ...
    (comp.lang.forth)
  • Re: Forth 200X and the Hayes tester
    ... compilation can work with minimal changes to source code. ... the assumption that programmers could be confused. ... standard names for some routines) to handle it. ...
    (comp.lang.forth)
  • [OT] Re: linker errors in vc6
    ... Your problem isn't related to your source code. ... newsgroup where you are likely to get more specialist help; ... often described as Standard C, ...
    (alt.comp.lang.learn.c-cpp)
  • Re: "NUL" device is missing
    ... > required for a specified formating string. ... I spent quite a bit of time last night reviewing the source code for the ... It is encouraging to hear than an updated ANSI C standard does finally ...
    (microsoft.public.windowsxp.embedded)