Causing a bit of a stir.

[Projectgilda]

Hi everyone, my name's Andrew. First time on the forum. I'm the chap who leaked the Union Jackpot exploit on Facebook and that has made a couple of YouTube videos from a guy off the forum. I was a games designer in the industry for 15 years and worked for JPM, Perfect Coin Machines and Barcrest. I also worked in London for an online, now defunct, company called Kismet Studios. I'll have a look around and contribute where I can. Peace out ✌

[thealteredemu]

Welcome aboard, look forward to hearing any insights and shenanigans :).

J

[johnparker007]

2 hours ago, Projectgilda said:

Hi everyone, my name's Andrew. First time on the forum. I'm the chap who leaked the Union Jackpot exploit on Facebook and that has made a couple of YouTube videos from a guy off the forum. I was a games designer in the industry for 15 years and worked for JPM, Perfect Coin Machines and Barcrest. I also worked in London for an online, now defunct, company called Kismet Studios. I'll have a look around and contribute where I can. Peace out ✌

Welcome Andrew    Your comment about the Union Jackpot exploit along with @Chopaholic's video did indeed cause quite the stir over here lol

 

Edited July 30 by johnparker007

[Projectgilda]

Well done in finding it in the code Proves I don't lie The developer in question lost his bottle and removed it from all other releases. You've found that too I see. It was put in for personal gain, not to be sold on. I lived with him at the time and he let it out after a few drinks. I kept my mouth shut as not to get anyone in trouble.

[redbags]

3 hours ago, Projectgilda said:

Hi everyone, my name's Andrew. First time on the forum. I'm the chap who leaked the Union Jackpot exploit on Facebook and that has made a couple of YouTube videos from a guy off the forum. I was a games designer in the industry for 15 years and worked for JPM, Perfect Coin Machines and Barcrest. I also worked in London for an online, now defunct, company called Kismet Studios. I'll have a look around and contribute where I can. Peace out ✌

Welcome aboard Andrew. I would personally be fascinated if you ever had any stories to share on the T7/random technology. It seems like such a leap in technology compared to the fruit machine stuff that I'd be so curious what it was like going through that transition.

Hope you enjoy your stay here.

Regards

[thealteredemu]

1 hour ago, Projectgilda said:

Well done in finding it in the code Proves I don't lie The developer in question lost his bottle and removed it from all other releases. You've found that too I see. It was put in for personal gain, not to be sold on. I lived with him at the time and he let it out after a few drinks. I kept my mouth shut as not to get anyone in trouble.

So speaking from experience, would you say most emptiers/bugs etc are mistakes or was the industry as corrupt as some people think.   You’re always going get some corruption is any industry when money is involved.  I honestly believe most are unintentional but it would be great to hear your experience and the ultimate price you would pay if you ever got caught with questionable code.

J

Edited July 30 by thealteredemu

[thealteredemu]

As a bit of a player late 80’s early 90’s it did seem like there was something on pretty much everything in that era, whether it was a complete emptier or some type of fraud, plugging/sparking etc.  Lots of machines seemed to susceptible to the old power down issue.

J

[Projectgilda]

Hi, 99.9% are unintentional from my experience and get re-prommed when they are discovered. The turn around of awp's is so fast they never get the testing they fully deserve. I worked on a 3/1 ratio. Meaning three games were developed in unison and then tested around the same time. If one was successful it's went instantly into production with no further testing. As the testing period for awp's, bingo et al was 8 weeks few games that had faults had them found during that period. All of a sudden there's 3,000 machines in the wild with an accidental exploit and an expressive re-prom required. Much of the time, unless requested by breweries for a change, they were left as is. The Union Jackpot exploit was intentional but only for personal gain by the person who coded it. It's the only experience I've had of 'dodgy' code going in a machine. I worked on hundreds of games, but so did many other manufacturers, and I can't speak for them. Hope that helps.

[thealteredemu]

What games did you work on say at Project Coin?

You don’t have to answer any questions by the way just interested.

J

Edited July 30 by thealteredemu

[Projectgilda]

Nothing you'd know. I was employed as an AWP designer as they weren't in the market at the time. We tried for a couple of years but nothing stuck. Hi-Lo to Heaven sold in small numbers and the rest failed testing. They only had one developer and he was overwhelmed by the complexity of my design compared to the arcade stuff he was used to. It was destined to fail tbh. Nice people though.

[thealteredemu]

Nice, I really wanted to design AWP when I was younger.  Well played.  Do you still work in the industry?

J

[Projectgilda]

No. I work in TV now. After Barcrest i'd had enough. It's a volatile industry. Worked in management for a few years then onto what I do now. It's quite surprising the level of impact my games have had. I've had lots of nice emails from various people who enjoyed them to bona fide collectors with huge private collections of my games. I never had an addiction issue, I played games all day and was paid well to travel the world playing other peoples games so I was never itching to play out of work, though I did, it was usually other manufacturers to see what they were offering up win or lose. Gaming should be fun first and foremost. I designed my games with the principal of AWP, Amusement WITHOUT prize. If you had fun playing it then the prizes (profit) were the icing on the cake. I designed high frequency games with lots of hidden features. The more you played the more your won. That was my principal.

Edited July 30 by Projectgilda

[Chopaholic]

Hi Andrew, welcome to the forums and thanks for signing up, hopefully you'll find this a friendly place and choose to stick around. I'm sure some folks here will have a lot of questions and comments for you (myself included) so please be prepared for that  Over the years it's been an extremely rare occurrence for anyone who worked in the industry - (and certainly someone who did so as extensively as you) - to be prepared to discuss much of anything, so thanks for taking the time to do so.

For clarity I'm the guy who made the two videos about Union Jackpot (Harry Headwound, formerly Degsy Degworth), having seen your post over on FB in the Cashman group about the Union Jackpot tell.

I saw the list of the machines you coded at JPM and there are some all time great AWPs in there, a few of them have videos on my channel which I'll link below, not least to show that I do have nice things to say about fruit machines! However I'm also pretty irked by all the nonsense that got out into the wild in the form of exploits and emptiers over the years, which go far beyond 'player skill' features, and left a lot of people essentially picking up the tab for the people 'in the know'.

As I said in the second Union Jackpot video, none of this about is about revenge or witch hunts for me, but I do want to be able to understand how so many problems were prevalent in the AWP (and clubber) market for so long. Stuff like plugging for example, which was a naughty action on the part of the player, but how on earth were Barcrest still putting out pluggable games like Cash Attack (the £35 era one) when plugging had been a core AWP weakness since the late 1980s? (Eventually of course they just gave up and put that 'overwatch' feature in, that causes a total lock on the machine if it goes too far over far percentage and insists that the operator contact Barcrest.)

Anyway, here are some videos saying nice things about some of the machines you made  (Although I'd draw your attention to Big Wheel, which will allow itself to get hugely behind percentage if played badly and simply never try to correct itself, something which JPM addressed later with the jackpot just spinning in when it needed to compensate back towards RTP.)

Oh yes that's reminded me, the WIN SERIES award on Indiana Jones on the Staff Of Ra feature, no one managed to achieve it, ever, in normal play. We did a hack thing with the ROM and it does have the code behind it to award it, but it just seems to be impossible to achieve without the hack.

 

 

 

[Projectgilda]

Thanks for your kind words. I think I covered the faults in machines in a previous post and how manufacturers were resistant to change the proms unless requested. Indiana Jones is a great game but suffered from a high token % which we were told needed to be addressed. We refused. Staff of Ra, as opposed to Leap of Faith (or Holly grail for that matter), was restricted, but can happen, just rarely (like Big Money on Rollercoaster). I don't remember the data, it's 30 years ago Your videos were very interesting and I was well aware I was posting publicly. So no issue from me. Going back to Holly Grail, you need the physical machine in order to achieve high wins on that (just look at the machine from an acute angle showing only the holy grails on the mirror flips) pushing it over percentage. It was on purpose, wasn't hidden, you needed to be savvy. Many weren't, so we never had an issue with percentage. Unless there was a genuine fault in JPM machines, however skillfull you were it's always clawed back that percentage by reducing the feature frequency. Just slightly, but enough. Killed the based game slightly too. 

[sjakie43]

Welcome Andrew, nice info indeed 

[spa]

Welcome!

Have moved your intro to here

[Chopaholic]

On 30/07/2025 at 19:53, Projectgilda said:

Hi, 99.9% are unintentional from my experience and get re-prommed when they are discovered. The turn around of awp's is so fast they never get the testing they fully deserve. I worked on a 3/1 ratio. Meaning three games were developed in unison and then tested around the same time. If one was successful it's went instantly into production with no further testing. As the testing period for awp's, bingo et al was 8 weeks few games that had faults had them found during that period. All of a sudden there's 3,000 machines in the wild with an accidental exploit and an expressive re-prom required. Much of the time, unless requested by breweries for a change, they were left as is. The Union Jackpot exploit was intentional but only for personal gain by the person who coded it. It's the only experience I've had of 'dodgy' code going in a machine. I worked on hundreds of games, but so did many other manufacturers, and I can't speak for them. Hope that helps.

So what was the deal with the 1/12 numbering (or later, the numbers that stepped to 1/12 on machines with steps), was that a deliberately coded in indicator of value or an accidental effect of something else? It was an absolute mainstay of the JPM/ACE codebase for years and years.

On a non-numbering Roller Coaster you're never landing on Test Your Strength or higher on the feature stack, for example, unless you've been seeing 1/12 in normal play. Indiana Jones was another with very clear numbering, and this was very useful, especially when it got as far as the £10 jackpot. (Which also totally messed the balance of the machine up IMO and made it far less fun to play.)

Roller Coaster was extremely unusual in that it went all the way from the £6 jackpot to the £15 jackpot, I had a late chip £15 Roller Coaster in an arcade here that would store value in a very unusual way for a JPM, but as long as it was showing 1/12 you knew it'd go eventually, and that could do some seriously mental results from the Roller Coaster feature itself, £90 or even a £105 in extremis.

Numbering wasn't an emptier or anything like that, but if you played £3-£4 through a machine and didn't see a 1/12, you knew just to walk away.

EDIT - Changed 'numbering' to 'non-numbering' above, my bad!

[MikeyMonster]

Welcome Andrew - long time fan of a lot of those JPM games you worked... some absolute belters and I presume you were a prize yourself in the industry.

[dondplayer]

On 30/07/2025 at 22:25, Chopaholic said:

How on earth were Barcrest still putting out pluggable games like Cash Attack (the £35 era one) when plugging had been a core AWP weakness since the late 1980s? (Eventually of course they just gave up and put that 'overwatch' feature in, that causes a total lock on the machine if it goes too far over far percentage and insists that the operator contact Barcrest.)

As I understand it the problem with the Cash Attack was when it powered up it reverted to one stake and therefore wrongly assumed it had been paying the MS from that stake and therefore value was still there in the actual stake - this is a different issue to many of the others vulnerable to plugging.

[slotsmagic]

Hi Andrew, pleasure to see you here! Hopefully you don't get too overwhelmed with it all, it's a pretty quiet community but I know plenty of people will be intrigued. It's a treat to have someone who has seen the industry from a different angle, as opposed to most of use who either played, owned, or maybe operated them!

You might be quite surprised at some of the efforts people have gone to with the emulator too, especially considering the author of MFME is no longer with us.

Before culling my collection early this year during my fresh start as a non-gambler, I had 3 emulation cabinets, one made out of an Interplay (31.5" portrait touchscreen) one made out of a Psycho Cash Beast (retrofitted with dual screens) and one made out of an Electrocoin Vegas Strip cabinet (stock dual monitors).

With the exception of one, they were worked out using shared experiences of others here, using logic (and maybe some trial and error).

Basically PCs, off the shelf I/O boards, coin validators, hoppers - giving the chance to have a proper modern multi-game machine, but filled with classic slots, real money in and out and lamped buttons and coin entries.

Edited July 31 by slotsmagic

[Projectgilda]

33 minutes ago, MikeyMonster said:

Welcome Andrew - long time fan of a lot of those JPM games you worked... some absolute belters and I presume you were a prize yourself in the industry.

I started off as a tester but swiftly moved to design. I started in the best prize to stake ratio (£3/£6) and my games were meant to be fun. The increase in jackpot made it difficult to retain that fun, but I tried nonetheless. 

[Projectgilda]

1 hour ago, Chopaholic said:

So what was the deal with the 1/12 numbering (or later, the numbers that stepped to 1/12 on machines with steps), was that a deliberately coded in indicator of value or an accidental effect of something else? It was an absolute mainstay of the JPM/ACE codebase for years and years.

On a numbering Roller Coaster you're never landing on Test Your Strength or higher on the feature stack, for example, unless you've been seeing 1/12 in normal play. Indiana Jones was another with very clear numbering, and this was very useful, especially when it got as far as the £10 jackpot. (Which also totally messed the balance of the machine up IMO and made it far less fun to play.)

Roller Coaster was extremely unusual in that it went all the way from the £6 jackpot to the £15 jackpot, I had a late chip £15 Roller Coaster in an arcade here that would store value in a very unusual way for a JPM, but as long as it was showing 1/12 you knew it'd go eventually, and that could do some seriously mental results from the Roller Coaster feature itself, £90 or even a £105 in extremis.

Numbering wasn't an emptier or anything like that, but if you played £3-£4 through a machine and didn't see a 1/12, you knew just to walk away.

Numbering in Rollercoaster is a myth. Let me explain. The base game is irreverent to the feature game. 1's and 12's all day gives you no idea of the feature compensator. But again 1's and 12's don't prove anything as they can spin you to death, add again to death, boost to death etc... there's clear control on jackpot features but 1's and 12's aren't it. The biggest hint in rollercoaster is 'roll up roll up'. That means the machine couldn't give a fuck what you do, if you're skillfull enough, it's yours. A 12 or 1 on tin can alley is rare. Roll up is the key to Rollercoasters generosity.

[Projectgilda]

29 minutes ago, slotsmagic said:

Hi Andrew, pleasure to see you here! Hopefully you don't get too overwhelmed with it all, it's a pretty quiet community but I know plenty of people will be intrigued. It's a treat to have someone who has seen the industry from a different angle, as opposed to most of use who either played, owned, or maybe operated them!

You might be quite surprised at some of the efforts people have gone to with the emulator too, especially considering the author of MFME is no longer with us.

Before culling my collection early this year during my fresh start as a non-gambler, I had 3 emulation cabinets, one made out of an Interplay (31.5" portrait touchscreen) one made out of a Psycho Cash Beast (retrofitted with dual screens) and one made out of an Electrocoin Vegas Strip cabinet (stock dual monitors).

With the exception of one, they were worked out using shared experiences of others here, using logic (and maybe some trial and error).

Basically PCs, off the shelf I/O boards, coin validators, hoppers - giving the chance to have a proper modern multi-game machine, but filled with classic slots, real money in and out and lamped buttons and coin entries.

I'm astonished by the work you guys do on this forum. Seriously. I was part of the 1990s era, but so were you guys and this forum just shares it's love for that time. I'm glad to have a part of it. 

[Chopaholic]

7 hours ago, Projectgilda said:

Numbering in Rollercoaster is a myth. Let me explain. The base game is irreverent to the feature game. 1's and 12's all day gives you no idea of the feature compensator. But again 1's and 12's don't prove anything as they can spin you to death, add again to death, boost to death etc... there's clear control on jackpot features but 1's and 12's aren't it. The biggest hint in rollercoaster is 'roll up roll up'. That means the machine couldn't give a fuck what you do, if you're skillfull enough, it's yours. A 12 or 1 on tin can alley is rare. Roll up is the key to Rollercoasters generosity.

Hi Andrew thanks for that, to be clear are we definitely talking about the same thing here when it comes to numbering? Because numbering is definitely a thing!

Basically you keep an eye on the hi/lo reel during normal (base game) play and see if it's spinning 1s or 12s or not. If you don't see any 1/12s then the machine is not paying a jackpot when you get onto the feature, end of story. Indiana Jones does this clear as day, you're not landing on Leap Of Faith or Pathway To Riches, for example, and you're certainly not getting to the end of the trail, nudge pots won't be anything good, it'll avoid the big cash values towards the end of the trail etc. (Its favourite out there is an ESCAPE off a mystery and going back to £1.60 or Runaway Cash or something.)

You can see this all very clearly in the emulator, which is running the exact same ROMs as the real machines did.

It was more apparent as the jackpots got bigger as the machine had to defend them more vigorously, but even on the smaller prizes it's absolutely there and can be seen clearly in the emulator, or on the real things in retro arcades where they're still sited.

[Mort]

The way to prove this, potentially, is to find the block of compensators in RAM and then put read breakpoints in MFME across this compensator range (to include lo-resolution ones as well, 1-6 etc.).

Play the base game as normal, and when number reel spins, if it's reading compensators each time, then it will break each time OR it may only read compensators every so often. In which case, just before a spin to 1 or 12, I would expect to see a break in code execution due to the read BPs you have set.

If the base game is separate to the feature game entirely, then it could just be reading compensators, to allow that initial gamble to be won in the base game (with no knowledge of the feature being JP ready). Which was always what I saw numbering to be. Is it true that when numbering you will always be guaranteed a jackpot win on the feature game  (obviously with JPM that jackpot win could be under any JP capable feature, due to spread value) or if not, could it just be a by product effectively of the compensator read from the base game being high enough to allow the JP in the feature game _most_ of the time ?

Interesting stuff.

Edited August 1 by Mort