Answers to your questions:
ritdav: "That was really interesting.First question coming up.Emptiers were they mainly badly written code or were they always deliberate.Reading the above they would have to be beyond the compensators to work .What were the legendry ones and were people fired over them.If you can't talk too much about them thats fair enough."
Almost always mistakes. Never intentional apart from one guy at Ace who decide to try and make a few quid by putting an emptier in. He got caught in a pub, fired and prosecuted. I think he got 2 years for fraud. That was on an old SP-Ace club machine.
Were people fired - generally no, all software has bugs and fruit machines are no exception. The famous ones, like Donkey Kong, were just oversights (nudge debugging code left in by mistake).
We had an emptier on one of our games, where changing stake should have cancelled holds, but if you were quick enough you could hammer the holds and change stake at the same time. Oops!
When you get a report of an emptier, all hell breaks loose. I did 18+ hr days just playing the game, trying to work out what was going on based on (generally useless) info from the site. You also pore over the code and look for anything awry.
Guitar: "An actual question for you though, where do you get the random values from on a machine with no RNG fitted?"
Good question! Believe it or not this was on JPM's standard interview question for software engineers. So we had a mixed prime congruential RNG but I suspect most people now use a mersenne-twister RNG. This was in the core libraries for games so we used that. But to answer your question; if for some obscure reason you had no RNG, then you write one, or you use inputs to generate random numbers, maybe timers of button presses, or system timers. I can't think of any library I've ever see that didn't have an RNG though. They're pretty fundamental to gambling!
evo1: "On an offtopic question when you played with tokens did the machine play on a different cycle/% cause again never really got any return on token play"
Yes, token payouts were typically in the 50% to 60% region.
fuzion: "A lot of people I used to play with referred to compensators as pots, even a lo-tech game like Golden Game had quite a lot of separate pots to play around with."
Yes indeed - those lo-techs had about 20 compensators. My first ever bit of fruit machine code (aged 17!) was doing the attract mode for Golden Oldie. Attracts were written in a sub-language to make them quicker when writing lamp sequences (all done by hand...time consuming!). If you watch the "ripple" on the button lamps, you can see that I missed out the Start button further on in the sequence. That was also in every other Mazooma lo-tech - makes me laugh when I see them in a seaside arcade even now!
wizard: "I think a lot of the emptiers arose due to a "Free Win", my understanding of this is that a Free Win didn't update the compensators when it should have done, leaving the game in the same happy state."
Correct - the win wasn't removed from the comps, or in some cases the win was added to the bank, and then some visuals happen, before the win is removed from the comp. So, obviously if you switch the power off during those bank pay visuals, the win never got taken from the comp and hence a "free win". We never had this problem as the line of code that adds the money to the bank is always immediately after the win has been removed from the comps, and also we separated logic and visuals entirely. You'd be amazed the number of companies that didn't - and were always the ones with problems.
fruitman69: "Not really, anti force code normally will monitor certain variables, if forcing is detected then it would set a flag"
Actually, in my experience, what people termed anti-force code actually wasn't. It was anti-stats testing code! Manufacturers used to buy other manufacturers machines and play them, log every spin, and work out what the game was doing. I remember visiting Maygay in 1996 and seeing a Barcrest 10 out of 10 being stats played. So what everyone did, was monitor the last, say 200 coins to go in - if they were all £1 coins then you mangle how the game plays. Still hits % but plays differently. Obviously in a pub you'd get other coins inserted, so this would never happen.
wayne123: "Thanks for the info - love to read how the fruit machines have ripped people off - strange question but is it possible to look at an actual programme for a fruit machine ?"
Yes, I will be releasing the code for a game I wrote soon (when I can find the backups, probably in my attic!). It's a Hungarian AWP but still very similar to a UK game (dual lapper board, etc).
shaun2097: "also, a while back some red gaming fruit machines used to put the first £3 or £4 you inserted stright to the bottom cash box"
Correct! We spotted this whilst stats playing one in the office one day. We thought like you did - was just to suggest to the players the machine was "backing". Found no evidence of anything else.