Re: Online poker and RNG...

Gerry wrote:
So cheater has his program work like this: First, just before the deal,
the software knows, by seeing the cheater's position and the number of
other players at the table, which 5 points in the sequence it will be
able to use for analysis. Second, as soon as the hole cards are dealt,
and during the subsequent, say, 90 seconds until the flop is displayed,
the software can use those two cards to begin to narrow down the
possibilities of deck organizations. Then, finally, once the flop is
displayed, those missing three cards can be "plugged in," in order to
further refine -- as per your BOTE calculation -- the prediction to
full accuracy!

Sounds plausible. Another possibility is that, if the seed is only 32 bits,
then it may be possible to do all the computation in advance. One might be
able to precompute a "dictionary" which lists, for each possible combination
of those five cards, what seeds are possible. This can be done at your
leisure before you join the table. Then when you are dealt a hand and see
the five cards, you can look up in your dictionary what the value of the seed
is (this lookup can be done very quickly) and determine how the rest of the
deck is likely to be organized, and apply the rest of your strategy.

And there seems to be a set of horns of the dilemma, here: either a
site publishes its algorithm, in which case the missing ingredient is
the seed (which can be determined, yes?),

Sounds right.

or the site refuses to do so,
likely because its PRNG relies on the "secrecy" of the algorithm,
which, from what I understand, makes the whole thing weak!

Well, if the site refuses to make the algorithm public, that's usually a
bad sign. It is commonly viewed as a possible (though not authoritative)
indicator that either the designers don't know enough about crypto to
do a good job, or have something to hide.

Also, it is difficult to keep these algorithms totally secret --
for instance, insiders are likely to have access to the secret; if
their development machines ever get hacked, then someone can steal the
secret; etc. You can look up "security through obscurity" on Wikipedia
and elsewhere to see more about this argument.

Let me put it this way -- I would never bet real money at a site that
advertised the use of a 32-bit seed, or that refused to publish how its
random numbers were generated and refused to have the PRNG audited by
competent experts.

I should say that I have absolutely no familiarity with the specific site
you mentioned, and I'm speaking only in hypotheticals, assuming the facts
are as you have given them.
.

Relevant Pages

  • Re: Card dealing and random repetition
    ... the PRNG stores more state than it shows the user in its output. ... No, the length can be no more than the number of seeds; ... >>> in the order they landed on the table, then shuffle them as usual. ... the important issue is how many cards are visible (or ...
    (comp.programming)
  • Re: famous literature
    ... Before I describe the algorithm, ... there is an important mathematical function - xor - that has an ... will generate random seeds. ... the run length random number generator. ...
    (sci.crypt)
  • Re: Online poker RNG...
    ... Many sites use an RNG algorithm based on 2^32. ... even find one that accurately matched the cards dealt during several ... games of poker. ... each player gets two cards and three face-up cards ...
    (sci.math)
  • Re: suspicious or paranoid ??
    ... no other indications at the table, a majority of players at a given table can ... the site's clock that was used to shuffle and deal ... ... Another flaw in a different algorithm ... ... knowing your hole cards allowed the enterprising computer expert ... ...
    (rec.gambling.poker)
  • Re: Atlantic Coast Regionals
    ... algorithm that troubles me is that the seedings it provides are not ... seeds would be unstable because the algorithm would reverse the order ... even though ACB has two local instabilities. ...
    (rec.sport.disc)
Loading