29-08-2026, 11:12 AM
I know it's a little hard to understand, given this length-dependence.
As an Example let’s take “aiin”—a word whose overrepresentation has already driven generations of VMS cryptologists to despair: 10 percent of tokens contain “aiin.” (which, in terms of frequency, matches the ending “en” in German perfectly...
In that regard, the “aiin” here is just an example meant to illustrate what this is all about.)
Now imagine this: “aiin” as a 4-glyph word (i.e., standing alone) means something different than “aiin” in a 5-glyph word, such as “daiin” or “kaiin.” And “daiin,” in turn, means something different in a 6-glyph word.
At that point, the frequency of “aiin” would break down into small clusters. Here’s a table illustrating this.
From 0.12 to 3,42 percent—that’s much more manageable than 10 percent.
If you design this cipher this way, you can be sure that it will drive any poor cryptologist to the brink of madness!
And now you have to apply that to the other families...
Suddenly, the VMS is much less repetitive
As an Example let’s take “aiin”—a word whose overrepresentation has already driven generations of VMS cryptologists to despair: 10 percent of tokens contain “aiin.” (which, in terms of frequency, matches the ending “en” in German perfectly...
In that regard, the “aiin” here is just an example meant to illustrate what this is all about.)
Now imagine this: “aiin” as a 4-glyph word (i.e., standing alone) means something different than “aiin” in a 5-glyph word, such as “daiin” or “kaiin.” And “daiin,” in turn, means something different in a 6-glyph word.
At that point, the frequency of “aiin” would break down into small clusters. Here’s a table illustrating this.
From 0.12 to 3,42 percent—that’s much more manageable than 10 percent.
If you design this cipher this way, you can be sure that it will drive any poor cryptologist to the brink of madness!
And now you have to apply that to the other families...
Suddenly, the VMS is much less repetitive