Fascinating mate I wonder if anything like this could have been done on the original real machine when programming was being installed and if so how we could get it on the alpha numeric display to see how the mpu was thinking about what percentage to pay out.
I’m sure it was but I’d be equally sure the original programmers achieved their chosen percentage payout differently. I’ve no doubt that the originals paid attention to the current payout in order to keep it within an acceptable range. I’d be very keen to have a look at the original ROMs but I don’t know how to read the physical ROMs and emulators don’t seem to have the option to let you look at the source code itself. The emulator programmers probably didn’t think anyone would be interested in a long series of 0s and 1s but they equally didn’t think there was anybody as weird as me around either! 😊
Superb insight. Was wondering if you won a jackpot or 2 in close succession would the right hand number be lower on each spin until the perctage targets of the machine went back to the required levels. The reality is,the machine seems to offer out a max win regardless of if it was won or not. Interest stuff though mate.
Thanks, Chris. I came up with several ways of fiddling the percentage payout - this was just one and the easiest to explain. The number is picked completely at random each turn, irrespective of how much has been recently won. It’s between 0 and 11 rather than nine as that increases the number of high-value wins it will pay out. The program itself never considers the current payout percentage. I tried various combinations of parameters and let the program play itself all night. It will play 3,000 credits in 10 hours and I’ve found that the parameters it has now seem to give around 135% consistently over that time.
Two very interesting videos. Was the 'Treendous' at the end of the first video because the progra knew you were well over target or wasitbeing funny because it gavesuch a lousy win onthe last go?
Thank you, Richard. It’s programmed to give a ‘Tremendous’ if it pays at least £5. If anyone downloads it to play with their own emulator, it plays in blocks of 25 credits so £5 represents a 200% payout - very feasible if it pays a jackpot during those credits. I don’t modify the trigger payout so, when I video 100 credits or so, it would be a disaster if it didn’t give a ‘Tremendous’ at the end 😊
And I'm sure I remember soe years ago it was proved on the eulator that numbers or codes were used to pre determine whether or not you would be successful on a hi/lo gamble and there were ties where the achine would pre determine that a winning gablewould be impossible on the high low reel.
I’m sure you’re right. I’d love to see the source code for any of the machines I’ve simulated just to see how the professionals did it. I’ve not found anyone with an emulator that will actually display the source code. It also adds an extra layer of complexity. For example, consider Chris’ Monopoly machine. If it gives him a 1 on the feature which leads to ‘+3 nudges’, it has to be certain that it’s happy to give it twice as he won’t lose if he gambles. The nudge ladder is not linear at higher numbers of nudges either!
I wondered what would have happened if you had exchanged that £1 from the Cherry Earner win to super series? Would ithave awarded 10p for each spin or does the super series feature play differently?
The Super Series play differently. If there’s only one, it picks a random win (heavily biased away from the higher wins) and rounds up 10p or 20p to 30p. If there are more than one, it selects five wins without rounding any of them up subject to the following criteria: No more than one 10p win, No more than two 20p wins, and The total win value for all of the Super Series does not exceed £3. If the total win exceeds £1.50 it tells the payout routine to pay every win in tokens. Hence, before you press the ‘start’ button for the first spin, it already knows what it will give you. In this respect, it’s similar to the fact that it already knows what will appear on the reels when you exchange cash for nudges before you’ve even got the cash win! Since the cherry earners take precedence over the numbers displayed in the video, this is why the earner succeeded despite the ‘highest acceptable win’ being 10p and it’s also why £1 was available for two nudges.
I guess the rea machines in the day probably had siular ways of stopping you fro continously winning but probably more coplex as I'm sure they had to keep to target so would have had to be aware when it had lost sufficient to start a streak. I reeber JPMs in the early days would give extra long spins to force a 20p win.
Yes, I think so. I suspect that they had to avoid the total payout dropping below the legal limit (I think it was 78%) as well as avoiding paying too much. I often wonder if that’s why Chris’ machines occasionally have a jackpot dropping straight in and holding.
Fascinating mate I wonder if anything like this could have been done on the original real machine when programming was being installed and if so how we could get it on the alpha numeric display to see how the mpu was thinking about what percentage to pay out.
I’m sure it was but I’d be equally sure the original programmers achieved their chosen percentage payout differently. I’ve no doubt that the originals paid attention to the current payout in order to keep it within an acceptable range. I’d be very keen to have a look at the original ROMs but I don’t know how to read the physical ROMs and emulators don’t seem to have the option to let you look at the source code itself. The emulator programmers probably didn’t think anyone would be interested in a long series of 0s and 1s but they equally didn’t think there was anybody as weird as me around either! 😊
Superb insight.
Was wondering if you won a jackpot or 2 in close succession would the right hand number be lower on each spin until the perctage targets of the machine went back to the required levels.
The reality is,the machine seems to offer out a max win regardless of if it was won or not.
Interest stuff though mate.
Thanks, Chris. I came up with several ways of fiddling the percentage payout - this was just one and the easiest to explain.
The number is picked completely at random each turn, irrespective of how much has been recently won. It’s between 0 and 11 rather than nine as that increases the number of high-value wins it will pay out.
The program itself never considers the current payout percentage. I tried various combinations of parameters and let the program play itself all night. It will play 3,000 credits in 10 hours and I’ve found that the parameters it has now seem to give around 135% consistently over that time.
Two very interesting videos. Was the 'Treendous' at the end of the first video because the progra knew you were well over target or wasitbeing funny because it gavesuch a lousy win onthe last go?
Thank you, Richard. It’s programmed to give a ‘Tremendous’ if it pays at least £5. If anyone downloads it to play with their own emulator, it plays in blocks of 25 credits so £5 represents a 200% payout - very feasible if it pays a jackpot during those credits. I don’t modify the trigger payout so, when I video 100 credits or so, it would be a disaster if it didn’t give a ‘Tremendous’ at the end 😊
And I'm sure I remember soe years ago it was proved on the eulator that numbers or codes were used to pre determine whether or not you would be successful on a hi/lo gamble and there were ties where the achine would pre determine that a winning gablewould be impossible on the high low reel.
I’m sure you’re right. I’d love to see the source code for any of the machines I’ve simulated just to see how the professionals did it. I’ve not found anyone with an emulator that will actually display the source code.
It also adds an extra layer of complexity. For example, consider Chris’ Monopoly machine. If it gives him a 1 on the feature which leads to ‘+3 nudges’, it has to be certain that it’s happy to give it twice as he won’t lose if he gambles. The nudge ladder is not linear at higher numbers of nudges either!
I wondered what would have happened if you had exchanged that £1 from the Cherry Earner win to super series? Would ithave awarded 10p for each spin or does the super series feature play differently?
The Super Series play differently. If there’s only one, it picks a random win (heavily biased away from the higher wins) and rounds up 10p or 20p to 30p.
If there are more than one, it selects five wins without rounding any of them up subject to the following criteria:
No more than one 10p win,
No more than two 20p wins, and
The total win value for all of the Super Series does not exceed £3.
If the total win exceeds £1.50 it tells the payout routine to pay every win in tokens. Hence, before you press the ‘start’ button for the first spin, it already knows what it will give you. In this respect, it’s similar to the fact that it already knows what will appear on the reels when you exchange cash for nudges before you’ve even got the cash win! Since the cherry earners take precedence over the numbers displayed in the video, this is why the earner succeeded despite the ‘highest acceptable win’ being 10p and it’s also why £1 was available for two nudges.
I guess the rea machines in the day probably had siular ways of stopping you fro continously winning but probably more coplex as I'm sure they had to keep to target so would have had to be aware when it had lost sufficient to start a streak. I reeber JPMs in the early days would give extra long spins to force a 20p win.
Yes, I think so. I suspect that they had to avoid the total payout dropping below the legal limit (I think it was 78%) as well as avoiding paying too much. I often wonder if that’s why Chris’ machines occasionally have a jackpot dropping straight in and holding.