Cheating Fruit (Slot) Machines

ebbdr writes "Ever think that fruit machines cheat you? You would be right, at least in the UK. This article provides proof that fruit machine outcomes are predetermined and that the players inputs have little, if anything to do with it. And it lets you download the emulators and machine code required to test the hypothesis for yourself...."

Comments

This is a surprise?

[mrleemrlee]

I don't know how it is in Britain, but it's well-known that U.S. slot machines pay a fixed percentage that is set by the house. The symbols that come up on the reels aren't random, and aren't advertised as such. So I'm not entirely clear why this is news.

Maybe these kinds of machines are different in Britain, or maybe they're advertised differently ...

Re:This is a surprise?

[PurpleFloyd]

I'm not familiar with British gambling laws either, but they make one accusation that is particularly damning: some machines have a "double or nothing" feature after a big payout; one of the site's accusations is that the "double or nothing" games are rigged so the player never wins. Still, some of this does seem fishy: their ROMs seem to feature cabinet art and other machine decorations rather than just display outputs, and the tests have been statistical rather than based on the programming.

A really convincing argument would be based on the ROM's internal code rather than on statistical analyses of the emulator output. After all, as long as an emulator is used, there is no way to know whether the ROM really has "cheater code" built in, or whether it's an emulator bug that is causing the analyses to come up wrong.

In Soviet Russia, the machine cheats on you! Or..

[JaredOfEuropa]

Heh, I remember the gool ol' Random Runner, before it had its programming upgraded. It would offer you exactly the same gamble as the article mentions, ie. two flashing lights with 'win' and 'lose', you hit a button and one of the lights stays on. What you did on the Random Runner was keep the button depressed. If you lost, too bad, but if you won... the next level bet would start but you'd win automatically, and the next bet, and so on.

Random Runners were popular with proprietors as well, as it was easy to obtain ROMs for these machines that would drastically lower the payout. Seeig this kind of machine is like a red flag for inspectors; they'll be sure to inspect the ROM in the machine.

How else would they work?

[Rimbo]

I've thought about this myself. The whole key to a slot machine or a fruit machine is that you need to be able to set the "payout" percentage, typically something high like 98%. 98% means that the player gets back $0.98 on every $1, assuming he plays an infinite amount of times.

The only way to guarantee this is by determining what the payout is as soon as the money's in the slot. The "pick high or low" and all of these other things are just meant to help keep the player interested, so that the player keeps playing.

Slot machines use other tricks, too: You can play on multiple lines, or you can play multiple coins for higher bonuses. Obviously the bonuses are multiples of the number of coins you play, so they have zero effect on probability. Multiple lines increases the probability you'll win per spin, but it doesn't affect the probability per coin, which is what matters to the proprietor.

This isn't a scam or cheating or anything like that. It is the same principle behind coin-op arcade machines: You pay to play. On a machine that has 98% payout and takes quarters, that means you pay (theoretically) half a cent every time you spin. In reality, you spend more or less than that depending on random outcomes, but over millions of plays on thousands of machines it means a good twopence on the pound for the Brits and two cents on the dollar for Americans.

Companies, that is. Not for the players.

Re:where'd they get the rom from?

[JaredOfEuropa]

They aren't lying, the ROMs are likely to be originals, and the machines' owners are not breaking the law.

Confused?

The law (at least around here) does not state that the machine has to be truly random. The law merely specifies a minimum average payout. The machine probably has to cheat in order to meet the lawful minimum payout, without (obviously) exceeding it by too much.

Re:Do They Really Have Proof?

[neonstz]

I think this quote from the article demonstrates that kind of behaviour:

This RAM file demonstrates the "hold dilemma". On the second spin, two cherries will appear on the second and third reels, with the option to hold. If you elect NOT to hold the cherries, a cherry will spin in on Reel 1 on the next turn, leading the player to believe holding the cherries would have yielded a win. However, if you DO hold the cherries, a red 7 will spin in on the first reel instead.

The pseudo-code would probably look like this:

if( something_is_held )
{
....if( player_should_loose )
........spin_with_loosing_result();
....else
........spin_with_winning_result();
else
{
.... if( player_should_loose )
........spin_with_result_which_would_have_been_ win_if_the_player_had_held();
....else
........s pin_with_winning_result();
}

(How do I create indents? :)

Sometimes skill does make a difference

[Len]

A few years ago someone won over $600,000 from a machine at the Montreal Casino by analyzing patterns in the numbers that came up. The sequence repeated because the machine wasn't seeding the pseudo-random number generator properly. More info in Risks Digest.

Re:where'd they get the rom from?

[Zeinfeld]

Someone unlocked a machine and stole the board, and put the chip on a rom reader and dumped the code. Obviously, highly illegal.

Not if you own the board. Spare parts from slot machines are pretty easy to come by. There is a whole trade in used machines, cabinets, etc.

Slot machines are pretty common in the UK, most pubs have one. It is no more difficult to come across parts legally than it is to find parts for any other common appliance.

How (IGT, perhaps other) slot machines work

[fo0bar]

I've lived in Nevada for a little over a year now, and know several people who work at IGT, a gaming manufacturer. Some of the things I learned suprised me, some didn't. All I know is the odds are definitely not in your favor. Is it illegal? Hell, no.

First off, the article (yes, I read the article). The author's biggest peeve is that the outcome of the "double or nothing" option on the fruit machines is determined before the user even chooses. Big whoop. Whether the magic number is determined before or after you choose is meaningless; it does not affect the odds.

Second, a previous poster mentioned the RNG. In IGT slots (and I would imagine most modern ones), the RNG device is a super-sensitive measurement device that detects tiny vibrations in the chassis. This is a much better way of seeding a number generator than any software-based solution. No, banging on the chassis won't increase your odds, but it will cause the machine to tilt and will probably get the attention of a security guard. ;) Also, the machine uses this entropy to re-seed itself thousands of times per second, not just once in the beginning.

Lastly, there's the method for choosing if you win or loose. As soon as you press the "spin reels" button (or pull the handle on machine that still support that), the outcome is already known. Let me repeat: THE OUTCOME IS KNOWN before the reels start spinning. The actual spinning of the reels is just eye candy.

This part takes a bit more explaining: say each reel have three symbols on them (we'll call them A, B, and C; in reality, the reels have maybe a couple dozen). In this example, C is the most favorable; you get a jackpot if you get three C's. You would think that this would mean that you would have a 1 in 27 chance of hitting the jackpot (3^3). Nope. The internal mechanism works like so: Okay, you have 3 symbols on each wheel. Inside the program, there are 3 arrays of symbols, but the number of elements inside the array is much more than 3. Say these are the arrays:

• Reel 1: AAABBBBCCCCC
• Reel 2: AAAABBBBCCCC
• Reel 3: AAAAAAAABBBC

The machine picks a random element from each array. Do you see what's going on here? There are more Cs in the first reel array, making it very likely to hit a C on the first reel. Next is a slightly less chance to hit C again. The third time is nearly impossible. Yet it builds you hope up, thinking you're about to hit the jackpot.

Is this deceitful? Yes. Does it prey upon the stupid? Yes. Is it illegal? Nope. These methods produce a certain payout percentage, and the techniques for producing them are "public" knowledge, usually regulated by your state's gambling office.

In conclusion, stick to blackjack.

Re:Don't Trust Machines!!!!

[topham]

It was explained to me by my boss at one point in time that machines in the U.S. (not sure of the locale, sorry) may be required a payout a certain percentage of the time, BUT, the machine plays ITSELF when no-one is using it.

Any wins which occur at that time are LEGALLY counted as customer wins...

Re:Don't Trust Machines!!!!

[lbonser]

In Nevada, the source code for any legal game is reviewed by the state Gaming Control Board before approval. So it's pretty tough (but not impossible!) for a programmer to add such twists. And when a really big jackpot is hit, the casinos have internal audit personnel that tear the machine apart to check for any sort of tampering. At this time, they also review the chips to make sure they are approved chips. Many Nevada casinos advertize 98% payback, but by law, the payout can be as low as something like 78% (not sure the exact amount, but it's down in the 70's); which means on the average, the casino makes a lot more money than just a few pennies. The payback for some table games is even worse (some are better... in Blackjack, the odds constantly shift back and forth between the player and the casino). I work for a company that makes a computer system that interfaces to slot machines; I get to play slots everyday... A good job if you can get it ;-) I also have friends that work in casinos and for actual slot manufacturers. I'm not a mathmetics expert, nor even a gaming expert, but I do sorta know a bit about it. If anyone's interested, you can check out all the rules and regulations at: http://gaming.state.nv.us/

I have written slot machine software myself

[Namarrgon]

Though I doubt I am one of the friends of which you speak, I have written slot machine ("pokies") software for Tatts Victoria.

I can confirm that it's heavily regulated, and the RNG used is carefully analysed for randomness, with the the payoff tables (and to a lesser degree, the ordering of the symbols on "fruit machine" types) controlling the payout (which usually varies between 83% and 91%)

The results are only "pre-determined" at the time of the user starting the roll, but are completely random nonetheless. In other words, when the user pulls the arm (if the machine has an arm), the results of the roll (and any related results, e.g. from a "double-up") are randomly pre-selected, then the reels are spun to those positions.

What struck me most was the incredible security and redundancy the system has. In Victoria, the legal accounting requirements are very stringent, and the manufacturers themselves have a long list of attacks they have to be proof against (from long experience - everything from massive magnetic fields to electrical cattle prods have been used to defeat a slot machine's defenses).

For example, not only is the casing solid steel, locked and with mil-spec proofing against EMI, the CPU board and coin trays are both locked within separate steel compartments within the unit, and each requires a different key to unlock. All locks have failsafe mechanisms to record opening, and the cabinet door has a randomly-pulsed optical sensor as well.

Particularly, the win-loss game data is recorded into triple-battery-backed static RAM, in multiple CRC'd locations, with the same data being recorded simultaneously onto physical counters, printed in duplicate to a roll of paper (on some machines), and sent in real-time via encrypted LAN to a central host, which must verify all large payouts. Every coin and every game must be accounted for under any circumstance, particularly power failure in the middle of a game.

The coin sensors and payout mechanisms were equally sophisticated, and had to accurately deal with punters feeding large numbers of coins very rapidly into the machines, whilst still defeating "coin-on-a-string" style attacks.

It was an interesting project, but involved considerably more than I first expected. I can say that, after many all-nighters testing, I have come to truly dislike the sound of a slot machine :-/ (Ironically, for some years my next job required me to go to tradeshows in Las Vegas - from the very moment you step off the plane, you're assaulted by pokies on all sides)