Debunking Slot Myths: The Random Number Generator in Class III Gaming

[uniman]

Wow, how did I miss this thread. Good stuff and no dildo's, sex toys, etc.
Well there is the blinking VCR.  

Saw on TV a few months back they had a show on this Spanish family that had traveled Europe using physics to determine roulette wheel bias. They made good money! They came to Vegas and didn't have as much luck with the 00 and 0 american wheels.
The casinos would change/move their wheels around to different tables to confuse them. Good show.

RNG's
What most folks don't seem to understand is that, at least with the Universal's, a number is generated for each reel. So a three reel game would have three numbers. They could be the same number or not. When I say "generated" it means like Stat said earlier, the number needs to fit the virtual stop map. So an RNG produces, say, .0140089, and that number is run through some math to produce a number between 0 and 255 for a 256 stop table. That final number is an address in the eprom where a reel position is stored.
A standard reel has 22 stop positions. Eleven are symbols and the other 11 are blanks. The eprom sees them as locations 0-21. Zero being the first position on the reel strip and 21 the last. A three reel game will have a virtual stop table for each reel. So my virtual stop table has 256 numbers consisting of zeros through 21's. If the jackpot symbol is the last symbol on the reel strip than it is position 21. Chances are there are few number 21's in my three tables as they are high payouts. There will be a lot of number 20's and zeros as these sit next to the 21 and are the blanks next to the jackpot symbol. So you will see that you just missed the jackpot symbol as they come up often!
The RNG and the math after it want to create an even balance of numbers between 0-255. It's the amount of zeros through 21's in the 256 number table that determine the chance of losing and winning.
Hope that helps.

[StatFreak]

Op-Bell and Uniman, thank you both for filling in some of the blanks.     
K+

... Someone mentioned Class 2 games a couple of pages back. There are two main kinds of these, the pari-mutuels and the so-called "finite" games.
...
- you could be playing out the last 1000 outcomes that have not a single winner at all. From a player's point of view it's just like a Class 3 slot in that you have no idea what's going to come up next, but the RNG did its work in the past - the next game outcome is already known and nothing the player does can change it.
...

Op-Bell, I have to slightly disagree with the way you phrased that. The RNG may be random in both cases, but the use of it's output is not the same, and so the player's expectation cannot be (or at least, should not be.) It's not a question of past vs. future selection.

In the class II, there are a fixed number of outcomes randomized into a true cycle (not a theoretical one), and although their order is scrambled, only one of each can occur until all have occurred. In class III machines, there is a fixed universe of outcomes with a predetermined weighting, but every outcome is possible in every trial. Class IIs are using dependent trials and class IIIs , independent trials, so someone who plays a class III machine can rest assured that (s)he has the same chance of hitting every single payout on every single spin.


As an exercise, a class III machine could be designed to select a long series of outcomes in advance so that the player could not change what was to come, as in your example. But if it did so using a reasonably unbiased RNG following class III requirements, all outcomes would still be available to a player according to the weighting of the symbols, which would not be the same as what happens in a class II machine.



From the operator's point of view, the class II machine provides the security of a guaranteed house hold in both the short and long term. The class III machine could end up holding much more or less, or even paying out over 100% for a time, and the operator must rely on having many machines and a high volume of play over time to provide the expected income.

I still consider the class II finite slot a fraud (that's a personal opinion, only). When the average Joe plays a scratch off ticket, they understand that some of the prizes may have already been won, especially after a particular game has been out for a while. Slot machines are viewed differently by the public.

[Op-Bell]

From the operator's point of view, the class II machine provides the security of a guaranteed house hold in both the short and long term.

I don't think the operator could care less. What a Class 2 machine gives you is a fig leaf that allows you to pretend it's not gambling, since they operate in places where gambling is illegal - except for state lotteries, which everyone knows are not gambling, otherwise they wouldn't be allowed. If there were Justice in the world the state lottery people would all be serving lengthy prison terms, since the odds they offer are considerably worse than mobster games like the numbers racket. But they go to church on Sunday and preach about the "evils" of legalized gambling. F*ing hypocrites.

I still consider the class II finite slot a fraud (that's a personal opinion, only). When the average Joe plays a scratch off ticket, they understand that some of the prizes may have already been won, especially after a particular game has been out for a while. Slot machines are viewed differently by the public.

Your personal opinion concurs with mine, including the bit about the public perception. I've watched people playing Tab Force machines, which are basically just sophisticated readout units for pulltab cards with a bar code on them. They get excited when they win and think the machine is hot, whereas in fact the outcomes of the games they play are encoded on the piece of cardboard they're holding in their hand They would get exactly the same results on any other machine in the place.

[StatFreak]

From the operator's point of view, the class II machine provides the security of a guaranteed house hold in both the short and long term.

I don't think the operator could care less.
...

I was thinking of small time operators who might not want to take the risk of short-term deflection, but you're probably right for most establishments.

If there were Justice in the world the state lottery people would all be serving lengthy prison terms, since the odds they offer are considerably worse than mobster games like the numbers racket. But they go to church on Sunday and preach about the "evils" of legalized gambling. F*ing hypocrites.

I couldn't agree more. The California State lottery pays back an abysmal 50%.  I'd take the mob's odds any day over that.  

...They get excited when they win and think the machine is hot, whereas in fact the outcomes of the games they play are encoded on the piece of cardboard they're holding in their hand They would get exactly the same results on any other machine in the place.

That's why I try to debunk these myths when I hear gamblers talking about them as if they were true. People deserve to have access to the truth so they can make informed spending decisions. Keeping gaming machines shrouded in mystery is no better than the deceptive practices of other businesses.

I've visited online casinos that provide the payback percentage of every slot game on the help screens. I would love to see that mandated in Nevada casinos. At least with table games, the odds can be calculated by anyone by observing the payouts and the rules, or by going online to a site like the Wizard of Odds.

One of the activities I enjoy when visiting casinos is to look for new table games and/or rules. I ask for one of their color brochures, write the payouts down if they're not included, and then adjourn to my room to calculate the payback. When it comes to the newer games, they usually have too much of a house edge for my taste, but every once in a while I am pleasantly surprised.

Truth be told, I generally won't place a single wager on a new table game or game modification until I've run the numbers. It's not that I'm looking to win, but to know that I'm going to get good value for my money, or at least to know how much I'm betting in to the teeth of the tiger if I choose to do so.

Of course, I usually stick to the one game I can beat...               

Stat