5 ms·

From a software/hardware engineering perspective, how difficult is it to get into the slot machine game? I imagine it must have quite a rigorous set of guidelin

From a software/hardware engineering perspective, how difficult is it to get into the slot machine game? I imagine it must have quite a rigorous set of guidelines and testing procedures that must be followed diligently to avoid running afoul of regulations set out by gambling states. Also, I assume it must require a lot of trade show visits to try and sell major casinos on your machine, since the number of customers for your machines is severely limited. Is it an upfront cost business (aka, they pay you $XXX for each machine) or is it more of a partnership with the makers where the casinos share profits on the machines with the developer/designers? Off the top of my head, I'd think the latter would be more feasible as I'm sure popularity of machines must be somewhat short lived, as there always seems to be different machines all the time. Anyway, I'm curious if anyone with more experience in this area could give some brief highlights on what it's like working in that arena. I've always been curious :)

joezydeco 12y ago
I worked in the business for a number of years and still have friends/colleagues that are there. From an engineering perspective slot machine design isn't very hard at all. It's pretty much video game design. There's a relatively recent CPU putting down images, playing back video, doing some light/sound choreography, and maybe doing some mechanical motion on some games. Let's put it as slightly more complex than Candy Crush but way less complex than writing Halo. There's also a mechanical component with artists and designers making cabinets, art glass, and wacky add-ons that sit on top or around the machine. The hard work in slot machine design is in the math of the slot machine. You're trying to design a game that balances something interesting on-screen with enough volatility that keeps players hooked. If you decimate their bank in the first 100 spins that player will never come back. One typical way to manipulate this is through the secondary games or bonus games that pop up on certain reel combinations. At first this was a way to work around certain patents (the most famous being IGT's Telnaes patent) and now players just come to expect various levels of extra action when certain slot reels are hit. Every player likes something different, of course. IGT's Double Diamonds reel-spinning slot is over 25 years old and players still look for it. The math also has to jive with the gaming board. Certain states will mandate minimum payout percentages, or a mixture of these that average a certain payout. (This is why dollar slots in Vegas pay way better than penny slots. As long as all the action in the casino averages to the legal minimum, they're okay). Your typical slot manufacturer has a number of math/stat MS or PhDs working out the numbers on paper as well as a testing department that tries to verify the math empirically. I'll say "tries" because some designs are so complex and non-deterministic or dependent on player choice that it can't be verified entirely on paper anymore. Almost all American jurisdictions employ independent gaming labs that inspect the code and verify the math as well before the slot is approved to release in a given state. If you can get approved in Nevada, you're usually good everywhere as a rule of thumb. And then yes, the theme of the slot has also become incredibly important. Licensing pop culture items (movies, TV shows, dead musicians, celebrities, etc etc etc) is a constant fight between the manufacturers to capture and leverage into a usable slot idea. Some think it is creatively lazy to grab the next hot movie and turn it into a slot. Some think that's what you need to do to attract people's attention in a visually crammed environment. C'est la vie. A slot machine designer worth his salt can mix and meld all of these things into an interesting game. New and different game mechanics are hard to come by, though. When something does hit, you'll see that model replicated dozens of times with slightly different math or new themes slapped on top of it. You throw a lot of shit at the wall to see what sticks. Business wise, up until a few decades ago, slots were sold by manufacturers and entirely owned by the casinos. In the last 20 years or so this has transitioned into a participation relationship where the manufacturers own, maintain, and operate the games themselves and share a percentage of the handle with the casino. This has made the business a lot more cutthroat because a streak of lousy games can get you kicked off a property (or corporation) entirely. This is the downside of throwing shit at the wall, so you test and test continuously at smaller properties. Participation also allows interesting game mechanics like multi-casino slots (like Megabucks) where the manufacturer is holding the jackpot in an interest-bearing account in their reserves. Think about that when the jackpot is in 9 figures and nobody has hit it for a year. Anyway, if you're really interested, the big companies are always hiring. These days it's IGT in Reno, NV and WMS in Chicago, IL. There are smaller companies here and there (Konami, Aristocrat, Mikohn) and then a constellation of independent design groups that consult and design games on spec for the big dogs. Most designers work their way up from programming games or designing art for slots but if you have any video game or casual game experience there's always room for new talent.
- jedberg 12y ago
  > IGT's Double Diamonds reel-spinning slot is over 25 years old and players still look for it. It's the only slot I'll play because it has the best pay table in the casino. It's the only one that pays 1600 coins for a 2 coin bet with sufficient payouts otherwise.
  - toong 12y ago
    Euhm, I don't know anything about slot machines, but isn't the (statistical) expected value not always about the same ? The payout-distribution could be different, maybe that's what you like about it ? (anecdotal: I calculated the expected value for a few games on a gambling site a few years ago and they were all about the same, for every game, for every size of bet: expected value was around .96 dollar for every 1$ bet. YMMV.)
    - lewispollard 12y ago
      That expected value is known in the industry as RTP - return to player - and is generally in the 85-98% range. Once a game is certified as having that RTP, it can't change - BUT the standard deviation can vary wildly from machine to machine, with some paying out little and often, and others having a more 'lumpy' distribution of not paying out much and then paying out large prizes every so often. High rollers will often go for these lumpier machines, staking large amounts and hoping for the elusive x1000 multipliers.
      - joezydeco 12y ago
        Very good point. But players should also know that a given machine can be released in multiple payout levels. You can have a row of identical games but the game in the middle could be set at 95% and the others around it set to 87%. And the casino doesn't have to identify which is which. Sometimes you can glean this from the payout tables, you can almost always glean it from video poker games. There's a voodoo science around how casinos are laid out and how games are arranged to maximize cashflow. You can get a degree in this at UNLV.
downandout 12y ago
The big issue isn't engineering, it's licensing. In the words of the attorney that helped me with a Nevada gaming license application, "They really want to make sure you're clean before they give you a license to steal". For example, if you went to high school in Hawaii, they will send two gaming agents to Hawaii - on your dime - to interview your high school teachers. The process is long and expensive. And that's just Nevada. You have to be licensed in each individual jurisdiction where your games will be shipped. This, of course, is all before you even broach the subject of putting your game on casino floors with slot/table game managers. As far as revenue models, both of those you mentioned are used. This also goes back to licensing - if you are only licensed as a manufacturer (a slightly less onerous process) then you cannot get a percentage of profits. You must have an operating license to obtain a percentage. This is why many table game manufacturers (3 Card Poker etc) get a flat fee per table per day instead of a percentage. The machine market is dominated by IGT and WMS, and good luck penetrating it. You may, however, be able to develop a fun game and license it to one of the big slot firms. Probably the easiest way.
- joezydeco 12y ago
  If you hang out at the back of the Global Gaming Expo (aka G2E), there are dozens of inventors in little booths trying to show off their new game and get you to try it. If you don't have a connection to a big company, this is one way to get discovered, but by judging what I've seen in some of those booths, you really have to have a compelling, easy, addicting game to get noticed. And most of those little guys have none of the above.
  - downandout 12y ago
    Anyone can demo any game they have developed, but before those games go into a casino, they must go through everything I said above. Many of those guys go to G2E to either get noticed by a larger manufacturer, or to find investors to help them go through the licensing process (or convince existing investors to do the same after a good response at the show).

joezydeco 12y ago

I worked in the business for a number of years and still have friends/colleagues that are there. From an engineering perspective slot machine design isn't very hard at all. It's pretty much video game design. There's a relatively recent CPU putting down images, playing back video, doing some light/sound choreography, and maybe doing some mechanical motion on some games. Let's put it as slightly more complex than Candy Crush but way less complex than writing Halo. There's also a mechanical component with artists and designers making cabinets, art glass, and wacky add-ons that sit on top or around the machine. The hard work in slot machine design is in the math of the slot machine. You're trying to design a game that balances something interesting on-screen with enough volatility that keeps players hooked. If you decimate their bank in the first 100 spins that player will never come back. One typical way to manipulate this is through the secondary games or bonus games that pop up on certain reel combinations. At first this was a way to work around certain patents (the most famous being IGT's Telnaes patent) and now players just come to expect various levels of extra action when certain slot reels are hit. Every player likes something different, of course. IGT's Double Diamonds reel-spinning slot is over 25 years old and players still look for it. The math also has to jive with the gaming board. Certain states will mandate minimum payout percentages, or a mixture of these that average a certain payout. (This is why dollar slots in Vegas pay way better than penny slots. As long as all the action in the casino averages to the legal minimum, they're okay). Your typical slot manufacturer has a number of math/stat MS or PhDs working out the numbers on paper as well as a testing department that tries to verify the math empirically. I'll say "tries" because some designs are so complex and non-deterministic or dependent on player choice that it can't be verified entirely on paper anymore. Almost all American jurisdictions employ independent gaming labs that inspect the code and verify the math as well before the slot is approved to release in a given state. If you can get approved in Nevada, you're usually good everywhere as a rule of thumb. And then yes, the theme of the slot has also become incredibly important. Licensing pop culture items (movies, TV shows, dead musicians, celebrities, etc etc etc) is a constant fight between the manufacturers to capture and leverage into a usable slot idea. Some think it is creatively lazy to grab the next hot movie and turn it into a slot. Some think that's what you need to do to attract people's attention in a visually crammed environment. C'est la vie. A slot machine designer worth his salt can mix and meld all of these things into an interesting game. New and different game mechanics are hard to come by, though. When something does hit, you'll see that model replicated dozens of times with slightly different math or new themes slapped on top of it. You throw a lot of shit at the wall to see what sticks. Business wise, up until a few decades ago, slots were sold by manufacturers and entirely owned by the casinos. In the last 20 years or so this has transitioned into a participation relationship where the manufacturers own, maintain, and operate the games themselves and share a percentage of the handle with the casino. This has made the business a lot more cutthroat because a streak of lousy games can get you kicked off a property (or corporation) entirely. This is the downside of throwing shit at the wall, so you test and test continuously at smaller properties. Participation also allows interesting game mechanics like multi-casino slots (like Megabucks) where the manufacturer is holding the jackpot in an interest-bearing account in their reserves. Think about that when the jackpot is in 9 figures and nobody has hit it for a year. Anyway, if you're really interested, the big companies are always hiring. These days it's IGT in Reno, NV and WMS in Chicago, IL. There are smaller companies here and there (Konami, Aristocrat, Mikohn) and then a constellation of independent design groups that consult and design games on spec for the big dogs. Most designers work their way up from programming games or designing art for slots but if you have any video game or casual game experience there's always room for new talent.

jedberg 12y ago
> IGT's Double Diamonds reel-spinning slot is over 25 years old and players still look for it. It's the only slot I'll play because it has the best pay table in the casino. It's the only one that pays 1600 coins for a 2 coin bet with sufficient payouts otherwise.
- toong 12y ago
  Euhm, I don't know anything about slot machines, but isn't the (statistical) expected value not always about the same ? The payout-distribution could be different, maybe that's what you like about it ? (anecdotal: I calculated the expected value for a few games on a gambling site a few years ago and they were all about the same, for every game, for every size of bet: expected value was around .96 dollar for every 1$ bet. YMMV.)
  - lewispollard 12y ago
    That expected value is known in the industry as RTP - return to player - and is generally in the 85-98% range. Once a game is certified as having that RTP, it can't change - BUT the standard deviation can vary wildly from machine to machine, with some paying out little and often, and others having a more 'lumpy' distribution of not paying out much and then paying out large prizes every so often. High rollers will often go for these lumpier machines, staking large amounts and hoping for the elusive x1000 multipliers.
    - joezydeco 12y ago
      Very good point. But players should also know that a given machine can be released in multiple payout levels. You can have a row of identical games but the game in the middle could be set at 95% and the others around it set to 87%. And the casino doesn't have to identify which is which. Sometimes you can glean this from the payout tables, you can almost always glean it from video poker games. There's a voodoo science around how casinos are laid out and how games are arranged to maximize cashflow. You can get a degree in this at UNLV.

downandout 12y ago

The big issue isn't engineering, it's licensing. In the words of the attorney that helped me with a Nevada gaming license application, "They really want to make sure you're clean before they give you a license to steal". For example, if you went to high school in Hawaii, they will send two gaming agents to Hawaii - on your dime - to interview your high school teachers. The process is long and expensive. And that's just Nevada. You have to be licensed in each individual jurisdiction where your games will be shipped. This, of course, is all before you even broach the subject of putting your game on casino floors with slot/table game managers. As far as revenue models, both of those you mentioned are used. This also goes back to licensing - if you are only licensed as a manufacturer (a slightly less onerous process) then you cannot get a percentage of profits. You must have an operating license to obtain a percentage. This is why many table game manufacturers (3 Card Poker etc) get a flat fee per table per day instead of a percentage. The machine market is dominated by IGT and WMS, and good luck penetrating it. You may, however, be able to develop a fun game and license it to one of the big slot firms. Probably the easiest way.

joezydeco 12y ago
If you hang out at the back of the Global Gaming Expo (aka G2E), there are dozens of inventors in little booths trying to show off their new game and get you to try it. If you don't have a connection to a big company, this is one way to get discovered, but by judging what I've seen in some of those booths, you really have to have a compelling, easy, addicting game to get noticed. And most of those little guys have none of the above.
- downandout 12y ago
  Anyone can demo any game they have developed, but before those games go into a casino, they must go through everything I said above. Many of those guys go to G2E to either get noticed by a larger manufacturer, or to find investors to help them go through the licensing process (or convince existing investors to do the same after a good response at the show).