From: "David Neal" <dneal@electrotex.com>
Date: Fri, 26 Jan 1996 08:51:57 +0800
To: "Erik E. Fair"  (Time Keeper) <cypherpunks@toad.com
Subject: Re: another thought about random numbers
Message-ID: <199601252102.PAA26495@etex.electrotex.com>
MIME-Version: 1.0
Content-Type: text/plain


> Date:          Wed, 24 Jan 1996 18:58:27 -0800
> To:            cypherpunks@toad.com
> From:          "Erik E. Fair"  (Time Keeper) <fair@clock.org>
> Subject:       another thought about random numbers

> While musing over a roulette table, and noticing the preponderence of
> electronic games in the various Casinos in Stateline, NV, a thought
> occurred: does anyone know what sorts of random number generators those
> electronic games use, and how (if at all) they are measured and regulated by
> the Nevada Gaming Commission? They might have something to teach us.
> 
> Erik Fair

I cannot speak for what algorithms are used in the devices, but someone made
a low-tech analysis and hack for these machines a while back.  He noticed
that a certain brand of keno machine reseeded its random number generated
with a constant each time power to the machine was lost.  Unfortunately he
hit the same casino three times (the stainless steel rat says always be
prepared to walk away no matter how much is at stake to steal again
another day) and he was forced to divulge his method.  The machine were
consequently fixed.  I doubt those machines use crypto strength RNGs because
who is going to spend the time, energy and money to hack a .25c slot machine?
The payout is too small.   For an excellent real-world example of this try
'The Eudaemonicus Pie.'  Briefly, its a book about people who successfully
hacked roulette machines but at a net loss because of all the time and
energy spent developing their technique.


> 
>