>>>>>>>>>>Why should the logic be one way ?
>>>>>>>>>
>>>>>>>>>Taxes are not supposed to be a game of chance.
>>>>>>>>>
>>>>>>>>>Slot machines are.
>>>>>>>>
>>>>>>>>Before we argue this further are we agreed that the $43m was an erroneous figure ?
>>>>>>>
>>>>>>>It depends. If the machine had signs on it, or it displayed *before* the wager was taken, that the maximum payout was $215K, then yes, it's an error.
>>>>>>>
>>>>>>>If there's no such display, the contract is, "I pay $X, I might win
a lot of money". If the amount is unspecified, then if the machine announces a $43M win, that is clearly not an error, as it's clearly
a lot of money.
>>>>>>>
>>>>>>>Bear in mind, it's in a casino's interest to tempt punters with higher potential payouts - it's more attractive to wager on a machine that might pay $10M than one that might pay $100K. Gaming commissions take a very dim view of any practices where punters are misled, even unintentionally.
>>>>>>
>>>>>>Doesn't the machine *have* to show (or at least make accessible) a pay table which would specify the maximum payout ?
>>>>>
>>>>>I don't know. That might depend on the gaming jurisdiction.
>>>>>
>>>>>>IAC, according to this link:http://www.huffingtonpost.com/2010/03/31/fortune-valley-casino-say_n_520182.html
>>>>>>"Colorado gaming authorities say the casino has no legal obligation to pay the $42.9 million. Don Burmania, a spokesman for the division, said the top prize of $251,000 was
clearly posted in the casino and that a software malfunction is to blame for the glitch."
>>>>>
>>>>>Hmm, has anyone seen a picture of the interior of that casino where the maximum payout is "clearly posted"? If it's posted in a EULA-esque fashion, that could be challenged.
>>>>
>>>>From Colorado Department of Revenue (Gaming Division) Rule 12 47.1-1233 (2):
>>>>"The paytable for a slot machine game program must be displayable prior to making a wager and must include an explanation of any special features and the amount of the awards for all winning combinations."
>>>>
>>>>So the maximum amount paid out is stated by the operator. Obviously a player could not claim that the maximum amount should not apply simply because they had not bothered to read the paytable.
>>>>
>>>>>Assuming the $43M is wrong, then what's not in doubt is that there was a software error. That, in turn, makes it clear that the machine was not programmed properly to generate its expected payout.
>>>>>
>>>>>So, not only do we not know how much money was supposed to be paid out in this particular instance, there's no way to know whether any previous punters were ripped off or overpaid. IOW it's a rogue machine.
>>>>
>>>>The document referenced above also lists in detail the required software specification including error checks and acceptable error rates. The error could have been caused by a stray quark for all we know
>>>
>>>Exceedingly unlikely. Suppose it was a cosmic ray - the chances of such an event leaving the machine operational except for displaying an anomalous payout is basically nil.
>>>
>>>>and, given the specs, it's extremely unlikely that any previous error on that machine would not have been caught.
>>>
>>>I totally disagree here. The machine could have underpaid, or overpaid on any or all wins as long as they were within the payout range, or simply not paid out anything when it should have. This could have happened with literally *all* prior wagers and there would be no way for any punter to detect a problem. It might have been possible for the casino operator to spot a problem by comparing the machine statistically with its peers, but even that isn't guaranteed.
>>>
>>>If I had to guess to save my life over what actually happened, it would likely be something like this:
>>>
>>>- Gaming electronics are difficult, expensive and time-consuming to develop and get certified
>>>
>>>- One set of certified electronics is likely used in a wide range of machines, with a wide range of payouts - and/or any given machine can be set for a wide range of payouts
>>>
>>>- For security, it's possible payout tables are burned into an EEPROM or similar
>>>
>>>- *Maybe* an EEPROM from a high-payout machine was re-used, without being fully erased, and the punter hit an obsolete lookup value
>>
>>In answer to most of that I'll just refer you to the relevant regulations on software auditing: There's a link from here:
>>
http://www.colorado.gov/cs/Satellite?c=Document_C&cid=1214386480925&pagename=Rev-Gaming%2FDocument_C%2FRGMAddLink>>Start at about 47.1-1222.....
>>A lot of the wording there seems to be the same as that for the Nevada Gaming Control Board - but if anything the Nevada version seems more detailed and onerous
>
>Thanks for the link - that gibes with what I remember; that the process makes a moon shot look simple.
>
>But, despite all the requirements, testing, and auditing, an anomaly still occurred. One would think with all the audit trail in place, figuring out what went wrong should be simple.
>
"How hard can that be?" ;-)
