🚀 go-pugleaf

In sci.crypt, strnlght@netcom.com (David Sternlight) writes:
>Finally, there seem to be two sets of codes. One, of up to six digits,
>is for the standard start times (on the half-hour) and lengths (multiples
>of 30 minutes). A second, of more digits, seems to be for the exception
>cases.

I'm one of the people the "broke" the codes.  It isn't that simple.

The algorithm starts out being quite simple for single digit codes, and
gets more and more complex for each digit you add.  The VCR Plus+
algorithm is desinged so that common shows map to shorter numbers -
which reduces the average number of digits you have to type.  The shorter
the code, the more "typical" the show time.

The algorithm gets more complex every time you add another digit.  There
is no second algorithm for 7 and 8 digit codes.  We just lost interest
in the project and stopped working on it before solving the 7 and 8 digit
codes.

>The first set has now been broken and a convenient pair of
>"c" routines is making its way around the world.

Those routines (as far as I know - I haven't looked at them that closely
yet) are an implementation of our algorithm - the one that was published
in Cryptologia.

>The second has not
>yet been broken, as far as I can tell, but is mostly of interest only to
>make sure the recordings of cable movies with odd ending times are
>correctly stopped to avoid wasting tape.

As far as I know, no one has made in progress on the 7 and 8 digit codes.
However, like all the digits before, it's most likely just an extension
of the algorithm used on the shorter codes.  The 6 digit algorithm stops
in the middle of a range of shows.  I'm sure the 7 digit algorithm picks
up where the 6 digit algorithm stops.

However, the change in the algorthim from the 6 to 7 digit codes seems
more complex than the change from the 5 to 6 digit codes.

>Either the primary code is a table look-up in the chip, or it has been
>broken by trial and error--the c routines work by table look-up.

Not really.  Most of it is is algorithmic.  However, there are tables
used in the algorithm.  The day and channel are calculated without the
use of any tables, and most of the start times and durations are also
calculated without the use of tables.  And I wouldn't be surprised if
the one table that is used could be replaced with an algorithm.

The algorithm was created by many, many hours of research.  We never
looked at the ROMs in the VCR Plus+ device (assuming it has ROMs), so I
have no idea what you could learn (or would find) by doing that.

>Since there aren't as many published canonical instances of the
>secondary code, it may be a while before it is broken.

You don't have to collect published codes, you can just sit down with the
VCR Plus+ device and start punching in numbers.  But there are 9,999,999
7 digit codes (and each one requires 9 button presses to test), so it can
take a long time to get enough numbers.

This is at least part of why we stopped working on the codes.  It was
taking too much time.

/-----------------------------------------------------------------------------\
| Curt Welch - KCW Consulting           By 1995, there will be just two kinds |
| curt@kcwc.com - NeXTmail please!      of professionals:  Those who use NeXT,|
| (703) 938-4152  FAX (703) 938-5756    and those without Jobs.               |
\-----------------------------------------------------------------------------/