From:	CSBVAX::CSBVAX::MRGATE::"SMTP::PYRITE.RUTGERS.EDU::SECURITY"  8-MAR-1989 10:40
To:	MRGATE::"ARISIA::EVERHART"
Subj:	Re: Simplex locks


Sender: security@pyrite.rutgers.edu
Date: 19 Feb 89 02:18:34 GMT
From: nichols@cbnewsc.att.com (robert.k.nichols)
Subject: Re: Simplex locks
To: misc-security@att.att.com

Some time ago I computed 1152 (as I recall) as the number of possible
combinations -- not exactly what is implied by the "thousands of
possible combinations" claimed by the Simplex folks (recently I saw
some blurb that claimed "more than 2000" combinations).

Incidentally, if (through psychoanalysis of the user) you suspect a
"complex" combination involving 3 buttons simultaneously, your guessing
job is greatly reduced.  There are only 10 such 3-button sets, each with
only 2 other buttons to consider.

There IS a way to generate some additional combinations, although it
violates the instructions of the manufacturer to "remove your fingers
from the buttons before turning the knob," risking damage to the lock if
you're not careful when opening it.  At the end of the sequence, an
otherwise unused button can be pressed lightly, just to the point of
increased resistance, and HELD THERE as the knob is turned.  Such a
combination is distinct both from pressing the button all the way and
from not pressing the button at all.  The basic principles for
setting/changing such combinations still apply: set the combination
using the same actions you're going to use to open the lock; to change
the combination, press the present combination and, while holding the
final button(s) activate the "change" slide.

It is possible to detect such combinations through manipulation and
determine the buttons involved, but such combinations would confound
the unaware.

It never ceases to amaze me the way these very useful locks are
mis-applied (largely due to mis-information supplied by the
manufacturer).  The locks are primarily suited (at least in my mind) to
preventing casual entry to a restricted area, provided that the location
of the lock is such that anyone attempting manipulation would be
immediately observed and challenged.

	-*-		-*-		-*-

Now that I have located the papers with my old calculations, I find that
the actual number of combinations for a Simplex lock is 1082 (even worse
than the 1152 I remembered).

Just for kicks, I calculated the effect of the "trick" I mentioned
(lightly pressing some of the unused buttons and holding them while
turning the knob).  There are a total of 2163 combinations if you allow
this.  Perhaps this is what the manufacturer had in mind when claiming
"greater than 2000" combinations, but the operating instructions
expressly forbid doing this.

What puzzles me is why the lock wasn't made with 6 buttons instead of 5.
The number of combinations then rises to over a million (without
"tricks") and security against manipulation is also improved, at least
for combinations that use all the buttons.

[Moderator add-on: The above-mentioned "trick" is Simplex's "high-security-
mode", which in theory they only point out to their high-muckymuck military
customers.  Years ago at a locksmith show I had one of the Simplex reps set
"some random combo" on one of their demos to see if I could open it; he used
one of these and I couldn't get it open at the time, although I rolled out of
bed with the answer the next morning...    _H*]