Author Topic: RNG Selection of Numbers?  (Read 2375 times)

0 Members and 2 Guests are viewing this topic.

Reyth

RNG Selection of Numbers?
« on: March 20, 2018, 07:18:03 PM »
variance within the bet (amount and type) is key to any RNG

Hey Heat!  What do you mean specifically?
 

heatmap

Re: RNG Selection of Numbers?
« Reply #1 on: March 20, 2018, 07:40:23 PM »
welpppp i rarely use any type of system any more if i do play roulette but i mean specifically changing your bet in general. i play very specific bets which maximize the skips a ball can possibly do. which is why ive seen such interesting stuff on the wheels i play on. i try to corner the ball, etc, yada yada, ive said all of this before.

i have found that if i play in between big banking people and walk away before they start to win i can be successful.

^ this is how i used to think ;-)

now i play big sections but i play AROUND people and do what i consider to be randomizing my bets. i still have to walk away before the opposite person starts to win though. its a very touchy time and if i dont win immediately i walk away after my first bet. ive lost too much money SITTING at the table and now its more of a skill thing slash fun because ive put too much money and time into believing i could beat a\ physical, electromechanical wheel, with RNG generator meant to feel like a real wheel.
 

heatmap

Re: RNG Selection of Numbers?
« Reply #2 on: March 20, 2018, 07:43:57 PM »
im going to go as far as saying that ive seen this happen almost every time i bet.

like i said im notorious for betting outside columns or dozens. i know them like the back of my hand.

whenever i leave the game a repeat happens.
 

Reyth

Re: RNG Selection of Numbers?
« Reply #3 on: March 20, 2018, 08:14:01 PM »
Ok!  I just wondered if you might have thoughts about the structure of bets and the way that an RNG might choose numbers. :)
 

heatmap

Re: RNG Selection of Numbers?
« Reply #4 on: March 20, 2018, 08:56:18 PM »
i believe that, when a seed is created, they are not necessarily trying to be RANDOM but UNIQUE.

So you want to generate a random number and dont want anyone to be able to GUESS how you created it?

You need a unique seed obviously, or one that has not ever been placed on a list to eventually be used in brute forcing anything.

so you combine properties that involves the user and what they are doing at the moment such as bet amount, maybe a time stamp, and something the player cant even predict. Its definitely based on what bet you are doing, how much you are betting, how much you have won, how much you have lost, and the more variables they can combine into a seed, the more unpredictable (or unique) the outcome becomes.
 
The following users thanked this post: kav

Reyth

Re: RNG Selection of Numbers?
« Reply #5 on: March 20, 2018, 09:15:24 PM »
VERY interesting concept.  Using the player's parameters as the seed.  Why wouldn't this be caught by auditing?
 

heatmap

Re: RNG Selection of Numbers?
« Reply #6 on: March 20, 2018, 09:22:30 PM »
auditing doesnt look at software per-se. simply the last N number of spins. auditing around where i live is done on the fly as well. they have a requirement that all transactions and what not get sent directly to a server owned by the state. i gaurentee the person doing an audit doesnt look at code. software is usually locked down by intellectual property laws and usually people privy to see code are gagged by nondisclosure agreements.
 

heatmap

Re: RNG Selection of Numbers?
« Reply #7 on: March 20, 2018, 09:28:52 PM »
the closest you or anyone in the industry will ever get to verifying a specific code is by the hash number created when the binary is created. that hash number will NEVER change unless someone manipulates the program and every single time the machine starts i bet anything that hash is checked against a server for a change in the number. when an auditor checks the machines they have a pre made program that they have no clue what it does, but i bet that program does some kind of hash checking
 

heatmap

Re: RNG Selection of Numbers?
« Reply #8 on: March 20, 2018, 09:44:07 PM »
i also want to say that i think repeats happen because people are betting the same exact bets that they have been. i see it on tables where the same amount of people all keep betting the same amount and bets that were bet on in the last round. and when someone tries to take advantage of the repeating, the outcome changes because the outcome is seeded by everyones bets at the table partially
 

Reyth

Re: RNG Selection of Numbers?
« Reply #9 on: March 20, 2018, 11:38:52 PM »
Well what I meant by auditing was, true random is where every possible item in a list has an equal chance of coming out and if this is being "manipulated" by the actions of the players, why wouldn't THAT be caught by auditing because it should be clear that there isn't a natural random flow?
 

heatmap

Re: RNG Selection of Numbers?
« Reply #10 on: March 20, 2018, 11:58:24 PM »
@ryeth - oh it is audited in real time and can be stopped immediately. ive only seen one game stopped for what i think was cheating or non random results, but the table was vacated after the last person and inspected til it was reopened. like i said i am not sure as to why they shut it down because from a very vague memory i think i asked them and they said it was security purposes and weren't specific
 

heatmap

Re: RNG Selection of Numbers?
« Reply #11 on: March 21, 2018, 12:04:02 AM »
(pennsylvania) ยง 524.6. Integrated live Roulette wheels used on fully automated electronic gaming tables.
 (a) A fully automated electronic gaming table with an integrated live Roulette wheel must randomize the method by which the outcome is determined. This includes, but is not limited to the speed at which the ball is ejected onto the wheel and the speed that the wheel rotates.
(b) A fully automated electronic gaming table with an integrated live Roulette wheel must be capable of determining if the wheel meets a 95% confidence limit using a standard chi-squared test for goodness of fit. The calculation shall be made based on the following criteria:
(1) 10,000 outcomes have been generated.
 (2) A new calculation must be made for each 10,000 subsequent outcomes.
(3) The calculation must consider only the most recent 10,000 outcomes.
(c) A fully automated electronic gaming table with an integrated live Roulette wheel must be capable of displaying a visual notification, clearable by an attendant, if at any time the live Roulette wheel has failed the chi-squared test for goodness of fit under subsection (b).
(d) A fully automated electronic gaming table with an integrated live Roulette wheel must be capable of disabling play in the event that the wheel has failed to meet the 95% confidence limit for goodness of fit test required under subsection (b) for two consecutive testing periods. Attendant interaction shall be required before enabling the table for play.
(e) A fully automated electronic gaming table with an integrated live Roulette wheel must be capable of providing a report that shows the results of the last chi-squared test as well as the previous 9 chi-squared tests. The report must contain the following:
(1) The time and date the test was performed.
(2) The table ID or any comparable identifier.
(3) The number of games used to perform the test.
(4) The outcome of the test.
 

Reyth

Re: RNG Selection of Numbers?
« Reply #12 on: March 21, 2018, 12:08:54 AM »
Exactly, so if the last 10,000 bets was altered by player attributes then it would fail the test, right?
 

heatmap

Re: RNG Selection of Numbers?
« Reply #13 on: March 21, 2018, 12:17:36 AM »
@ryeth - yes i would say so - because one of the attributes you speak about is not the player but the croupier and considering one of the specified random variables is supposed to be the speed of the ball it is directly dependent on them and can be stopped theoretically if the dealer was found to be bias or cheating or whatever you want to call it
 
The following users thanked this post: Reyth

scepticus

Re: RNG Selection of Numbers?
« Reply #14 on: March 21, 2018, 01:26:13 AM »
and 10,000 trials seem to the accepted number of trials to establish a " confidence level " ?