First-Hand:Building a New Generation of Slot Machines: Silicon Valley Meets Las Vegas

Submitted by Allan Alcorn

How I First Got Involved

Around 1993 I was finishing up work at Digital FX when I got a call from an “angel,” Dave Morse. "Hey Al,” Dave said, “let's start a company that makes slot machines for Vegas." He explained that he had a partner who was very close to the Las Vegas industry. This partner, who was a non-technical person, realized that the bulk of the revenue in Las Vegas comes from slot machines that were still technically very primitive. They still used physical reels spinning behind glass. Steve Wynn had shown that Las Vegas is not about gambling. It is about entertainment. Dave’s partner wondered if we could make the slot machines more entertaining. Slot machines that were more entertaining would make more money, and we'd have a product. That was the basic concept.

At the time, it struck me as a really an off-the-wall idea. Back in the old days at Atari, we had looked at it. But we shied away from the industry when we realized how difficult it was to get a new product into this industry. But when Dave Morse proposed the idea, I said, “Let’s do it.” After all Dave was a serious angel. He had previously started Amiga and then the 3DO company. It was something I didn't know how to do and it sounded like a good challenge. The project also sounded like a lot of fun. Then Dave said, "Why don't we go to the trade show starting up in a few weeks in Las Vegas?" Each year every the gambling industry has a trade show; one of the more fun trade shows to go to because you had smoking and pretty ladies, and the whole thing. But it's also a really good trade show.  I went and looked at all the offerings from all the companies making slot machines, and was just struck by how primitive they were. “Gosh,” I wondered, “why re they are so primitive? Why they don’t put more entertainment in these machines?”

At first I thought to myself, “What am I missing?” The idea seemed like picking apples off the ground. It just seems too easy. So I got a little paranoid. While in Las Vegas, I spoke to the head of the Nevada Gaming Control Board Technical Branch about some of the technical issues that one must address in order to make a slot machine comply with the Nevada gaming rules. It was also a very interesting experience. They were very helpful. I had thought that they would be standoffish. But in fact they were enthusiastic about having some Silicon Valley company make new slot machines. "If we make the slot machines more fun and better and they make more money,” they explained, “then we get more tax revenue. So it's in our interest to have better machines. It's simple. It's honest. So, it’s okay.” At that point, I was concerned about security. People like to cheat and steal from these machines and that's a big issue. So I asked them about their experience with modern cryptography – public-key cryptography, RSA, and stuff like that. They had never heard of it. Then I saw a real a opportunity here.

We had about a million dollars of start-up money from Dave Morse, the angel. But we had to get additional money. The next step, as with any start-up, was to build a prototype that would be good enough to convince investors to put money in. I had to put together a team of engineers to build the prototype and then the final machine.

Building the Team

In putting a team together it is always very, very critical that you build a new organization from the top down. You try to get the “A” players in at the top and they then they will hire the “A” players at the lower levels. So one of my first hires was a friend, Harry Jenkins, with whom I had worked with at Atari years earlier. He had been one of the best product designers that Atari had and had just become available. I gave him the tasks of building concept, the visuals, the product, the packaging. Then I continued to assemble the key players of engineering team. The computer-based machine that we were going to build was going to be a very powerful system with beautiful graphics like no one had ever seen before. Our concept was aimed at people who did not understand or care about computers, like classically blue-haired little old ladies that are drunk. The machine had to be very easy to play, very identifiable, enjoyable, and entertaining. So my next big hire was Adam Leventhal, one of the best graphics guys that I hired away from Apple Computer.

One of the key engineering decisions that we made very early on was to not use a custom electronic digital board, as was done on all the other video games and slot machines in 1993. That was the wrong way to go for the early 1990s. Instead we decided to use a standard Intel motherboard. It had far more power than anything we could design, and we could buy it finished, all debugged and tested with infinite amount of development tools and accessories, for $125.

Due to severe stability requirements we chose not to use Microsoft/Windows software. A slot machine cannot crash. It had to be extremely stable. If you're making a slot machine, the “blue screen of death" has a more literal meaning. I needed to hire people with top-notch, real-time, operating system expertise. There were a few of them around. I wanted experts, experts who had used these operating systems before. They were engineers who had designed things for the remote telephone answering machines that were all built on the little individual circuit cards, with a real-time operating system running on an x86 microprocessor. These remote answering machines could not go down. That was the design philosophy that I wanted. So I did not want the average PC type of designer. I wanted that kind of designer who understood hard, stable, product design. We had a design team with experience interfacing from the high-level stuff down to the hardware level.

It was also crucial that we hire someone who understood the casino slot business and its customers. Grade A nerd engineers generally don’t gamble. They know that they're going to lose so they don’t do it. But we were going to make a product that would appeal to someone who did gamble. We need to get that kind of understanding into the company, both by training our existing engineers and by hiring people from that industry.

In regards to knowledge of the slot machine industry, our key hire was a gentleman named Andrew Pascal, who at the time was running the slot operations at the Mirage Casino and happened to be Steve Wynn's nephew. He was a young man who wasn't too set in his ideas, and was enthused about the idea of coming to Silicon Valley to do a Silicon Valley startup. He was excited about bringing his experience and understanding of the players’ wants to our start-up, and then, hopefully, create the right content for our new machines.

Mr. Wynn was not happy when we hired Andrew. Mr. Wynn had daughters but no sons. He was grooming his nephew to succeed him in the business. For Andrew to leave and work for us was a shock for Mr. Wynn. We did not realize at the time that Mr. Wynn’s displeasure could work against us. We thought the hire was a great idea.

At the heart of the project was cryptography. Well, when you hire A-player digital engineers, they're all going to know about cryptography. We had wonderful meetings with all these very, very bright people, as we got into the architecture and the design. We had all seen the movie Casino that had just come out. And I pointed out to the group that the movie, even though it was supposed to be fiction, was based on fact. That sort of stuff really happens. Some of the guys on the design team had previous experience in video game design. Often in that world, engineers will put in Easter eggs – little surprises – in the software that they can show to their friends. If they hit the right combination of keys it'll say their name or something. You don’t want to do this in the casino slot business!

So first step was to get the core team of engineers to appreciate what we had to do. We explicitly stated that our goal was an architecture – a system – such that it would be impossible for any engineer to put a GAFT application, or cheating application, on the system no matter what he wanted to do. We had to create a security system that was so tight that it would not accept any foreign software. The solution had to deal with the entire architecture and not just the operating system. All programs that ran our slot machine system had to be approved and authenticated by company’s management. We proved mathematically that that was indeed the case.

Getting this security to work was a lot of fun. I mean it was really exciting.  The first patent, fortunately named the Alcorn Patent, was really the result of a team effort. It was one of my best experiences because a team of six or seven engineers were able to come up with something better than any one of us could have come up with alone.

The First Prototype

The canonical process in high-tech start-ups is you get your angel money – get your million bucks from the angel, put a team together, and that lets you live for a few months – six months, maybe. But during that period of time you've got to hustle and build a prototype compelling enough to attract your next round of investors. In this case it was going to be the B round of real funding. We were looking for several million dollars – eight, ten million, something like that. We had to come up with the prototype in a very short time. Unable to create the hardware from scratch, we did what everybody does and bought the finest graphics systems that you could buy at the time: a Silicon Graphics machine. We modified it to drive a special display that would allow us to show how our system would work and why it was different than any other systems around. For the prototype we had a wooden cabinet holding a couple of the design features: 16" by 9" cathode ray tube, the first and only such tube made. It was in portrait mode. The tube had been specially imported from Europe for us. We modified the Silicon Graphics machine to interface with the tube. We had a slot machine crank handle bolted to the side of it.

Showing our crude looking prototype to actual slot bosses and potential casino customers was an amusing process. We'd fly them out to our office in Saratoga, California, to see it. Several of them came away thinking, "They can't make something out of wood!" They were unable to abstract beyond the model. It is an interesting ability. Where a venture capitalist and an angel can see something and abstract what it could be, most people don’t see it that way. They have a different reaction. "Is that it? That's not going to work. You have to have a metal cabinet. You can't have wood." We eventually got these potential customers over this conceptual gap.

Some of the concepts we had that were very different that didn't stick that I think were fascinating because most of the time a slot machine sits passively on the floor. Most of the time, any given slot machine is not being played, Even though during its lifetime, in a year, it may take a million dollars, it just sits there idle. We had the idea of using a group of slot machines in a casino as a video wall to show advertising that could attract crowds. We had all the components needed. There was a big graphic display screen that could display information to a crowd. All the slot machines were to be hooked up on Ethernet, which was not in use at that time there in Las Vegas. Our machines could talk to each other. We could do all kinds of interesting things.

For example, in those days the slot machines worked with coins and dollar bills and could jam. The standard IGT machine would lock up and show on the LED display, F7G1. When it locks up no one can play. The customer may sit there waiting for someone to service it. What we did was have a video come up and a Professor Irwin Corey-type of character would come on and say, "Well, we've had some problems with the machine and we've got somebody coming over." Then the machine would start showing the customer card tricks or advertise products. We had graphics. Though these features were not gambling, they were entertainment. We showed that sort of stuff and people said, "Wow.” In the end, though, this aspect of our machine wasn't as popular as we thought it would be.

The Second Round of Investors

We went to the usual suspects down Sand Hill Road in Palo Alto. It was Dave Morse’s responsibility to get the investor crowd to come down. And so they came down to our beautiful little offices in Saratoga, and we would give them the dog and pony show. Andrew Pascal would talk about his part of it and I'd talk about my part of it. We had pretty much all the VC firms coming through. It was a blur. The most memorable potential investor for me was Larry Ellison. When he came by he announced, "I got 15 minutes." He had his limo waiting, and we had a 45-minute presentation. But as we into it, Ellison started asking all these technical questions about the graphics subsystem. We were building a beautiful, if not the best, graphic subsystem. So I said, "Hang on a second." I went and got Adam Leventhal, our hardware engineer, who was busy doing other stuff, and they just did the Vulcan mind meld. It was amazing. Larry said, "Wow. This is very much like the CAPS system that Disney used for animation." Adam said, "Yeah. I worked on that system." Then Ellison said, "Oh, wow." They spoke for about an hour and a half. Although Ellison was thrilled with the whole thing, he never invested. "Well, it's gambling,” Ellison said and then added, “I don’t want to invest in gambling." But we did get the guys from Kleiner Perkins, the number one VC firm in Silicon Valley, interested. Because we were a gambling investment, which was unheard of, we had to give the demo to every partner at Kleiner Perkins, about 50 partners, and half of them are in medical or biotech. So we had to sell it to a lot of people. Eventually they did invest. If you go to Kleiner Perkins in your second round and they say no, then you're pretty much doomed because nobody will want to invest if KP says no. On the other hand, if they say yes, then everyone wants to join in. It reminded me of a scene from a Laurel and Hardy movie where they're all trying to get through the doorway at the same time. It was amusing. We got the amount of investment needed to build the production model, but it was an unusual experience to say no to some of the top VC firms.

The Nevada Gaming Control Board

Our next challenge was to get approved by the Nevada Gaming Control Board. To sell a slot machine – a real gambling slot machine in Nevada - the corporation has to be licensed. To get this corporate license every executive or board member has to be investigated. Investigation by their standards means giving the Nevada Gaming Control Board $50,000 down payment on the investigation for each person that's going to be investigated. This investigation takes six months to a year. Every deal you were ever in gets unraveled. They want to keep out bad people. These investigations would take a very long time because our investors were all private. The Board explained to us that it would be a lot easier if we were a publicly traded company. In their eyes, a public company was vetted by the investors on Wall Street. But it's hard to go public if you don’t have a product to sell, and you can't sell a product unless you're public, which allows you to get a license, et cetera. So we had to do a concept Initial Public Offering (IPO), which was a very interesting process.

Normally, the standard way to fund a company is to get some investor money – private investor money – and you build a product and you start selling them, and once you have sales and you have earnings and profitability, then you can take a company public. But a concept IPO is one where you haven't actually shipped anything yet. You have a concept and that they'll give you millions of dollars for the team, the concept, the market, and the idea. We did one of the first concept IPOs.

The Nevada Gaming Control board wanted all the partners at Kleiner Perkins to be investigated. I told them that it would be impossible. The Board replied, "Well, Al, you don’t understand. Before you guys, the main source of venture funding in Nevada was the Teamsters Central States pension fund." And I said, "Well, Kleiner Perkins isn't like that and they're much nicer and if you lose a million dollars they don’t really mind so much." It was a real culture clash to have Silicon Valley meeting Las Vegas. We were considered these crazies from outer space. There are scoundrels in Silicon Valley, but they're a different kind. I said they're tough, but they're not like the Teamsters. Before we did our public offering, we had our final private investment round: the mezzanine round. We made it available to the casino owners in Las Vegas. We wanted them be part of the company. Our hope was that they would give us the money, at a better price than what Sand Hill Road would offer, and at the same time we would have involved customers.

For some bizarre reason the casino owners did not rush to take us up on the offer. But we got enough and we got our mezzanine round at a good price and then did the public offering, and then we got licensed. After we got our license we had to submit a prototype machine to the Nevada Gaming Control Board, which they had to go through meticulously, understand every bit of it – software, hardware, everything. Once they were convinced that it was a safe machine, an honest machine, then we got a conditional approval. We then put three machines on a trial run in the back of the Bally Grand Hotel.

When we first talked to the Nevada Gaming Control Board, we asked what their understanding was of modern cryptography: public-key cryptography and RSA. They really weren't aware of it. They had no experience with it. I remember asking, "Do you mean to tell me that the progressive slots that do million dollar jackpots – payouts – are based on single keys and keeping secrets?" Their reply was, "Yes.” We were going to change that. The Board was very supportive and enthusiastic. They had two techs who understood C. C, not C++, but C! So we made a point of using C, not C++. We had to educate the Board on modern software architecture. We contracted RSA Corporation to give seminars in Nevada to the Nevada Gaming Control Board.

We had to get the rules changed. In 1993, for security reasons, the existing slot machines, which had three reels spinning behind glass, were all microprocessor driven. Even though you see reels behind glass, the reels are just display mechanisms. The outcome is determined by the computer and the reels turn to show you what the thing is. It's not random in that sense. The rules required the program to be stored in a read-only memory chip. I found that an odd way to promote security, but they felt that there was security in having the program sitting on a ROM chip. If anyone wins the big jackpot and there's any question about it, they'll take the ROM chip out of the machine and put it in a comparator to make sure it hasn't been changed. That's how they verified integrity. We wanted something much, much better, so we invented this system that was far superior. But we had to convince Nevada Gaming Control Board to change the law – to rewrite the regs to allow this. And that's what we succeeded in doing.

The gaming industry in Nevada was very protective. They had used these regulatory laws to keep out new competitors, and we were pushing our way in. I recall a meeting in Carson City to announce the rule changes. The VPs of engineering from all the different slot manufacturers around the world attended. We were also sitting in the audience. The gaming Board was explaining public-key cryptography to the assembled engineers when an older engineer from one of the old-line slot companies gets up and he says, "Hey, can we delay the introduction of the rules until we catch up?" The head of the Nevada Gaming Control Board said, "Look, guys. Things improve, things change. Here's the new regs and here are the papers on RSA and how to do all this stuff." We were kind of smug; we had actually patented the whole process.

The important difference in our new generation of electronic slot machines from the existing ones using microprocessors was the graphic interface. The issue was one of virtual reels. On a three-reel slot machine with twenty spots on each reel, that's twenty to the third power of possibilities, which is 8,000 different outcomes. In other words the biggest jackpot is one in 8,000. With these odds, one cannot have a million dollar jackpot on a dollar machine. Instead, the machine actually had 50 stops for each reel in software. The software spun the reels with a random number generator that in turn produced the various spots, like a cherry, on the reels display. The physical reels spun as if they were random. With a 50 stop reel, the odds of winning are quite low, large payouts are possible, even though the physical reel you see have only 20 stops. We were essentially doing the same thing. We were spinning the reels in software and displaying, but we were using the cathode-ray tube and beautiful graphics to do it.

Talking to the people in the industry about the graphic interface was just amazing. I said, "Why haven't you done this?" "Oh, well, back in the '80s and the late '70s we tried. We made video slots and they didn't work. So we concluded that video slots don't work." Well, in fact I was around when that happened. They even came to Atari and asked us to help, but we declined and they did a terrible job. It was a horrible machine. So there was the attitude of “Forget video slots, they can't work.” It was clear that there was no interest among the traditional slot manufacturers to do a video machine. But our machine was more than just a slot machine. It also played poker, Keno, blackjack, and any number of other games of chance. It was all in the software. The player had the freedom to call up any of these different games, and our patented security system made it all tamper proof.

When creating a new architecture, new kind of machine, for an existing industry, you're going to be measured by the old stuff. We insisted that out slot machine have a handle on it. Oddly enough, the casino operators didn't want the slot machine handle any more. I told them, "Your players love it. That's what they want." They replied, "No, if we don’t have the handle we can save two inches and we can pack more machines in." But we insisted on a handle. We wanted the user to have a physical interaction, and the players’ responses proved that they wanted it also. The system would still have a huge touch screen. For the card games, only the touch screen was needed. The handle was a retro thing - we wanted it to look like a slot machine, be identifiable as a slot machine, but be a lot more fun than a traditional slot machine. Entertainment was the idea.

Difference between prototype and production model

To satisfy the Gaming Control Board and the operators, the production machine had to be completely reliable. It could not steal your quarter. It couldn’t just stop working. We used a real time operating system, on an off-the-shelf Intel motherboard. But we had to remove the flash ROM chip that came on the motherboard. The easiest way to cheat a machine is to bury something in flash inside ROM. Instead, we ran the operating system off a plug-in card on one of the PCI slots. Our machine had to survive all the standard attacks that the cheats can do to these machines.

People will risk their lives to steal 25 cents from a slot machine! They'll risk going to prison! Through the years, there have been quite imaginative and colorful schemes to cheat the slot machines. In the old days with the mechanical slot machines there the old lady with the five-pound purse. There was a magnet inside the purse. And what they would do is they'd wait until they got a cherry on the first reel, and then put that thing up there and now run the machine, and they could actually use it to move the reels to get the winning outcome they wanted. Another scam was to throw sparks on the machine to give free plays. They would use an automotive ignition coil, which could produce quite an energetic spark. The cheater could almost kill himself in the process. Another scheme involved putting the coin on a string and making the mahine think you put in multiple coins. The other favorite one involved the hopper that paid out the coins for a win. It's a standard mechanism that spits the coins out under motor control. There is an optical sensor that counts every coin that goes by to ensure an exact payout. So what they will do is take a grain of wheat light bulb – this is before LEDs and – and put it on a twisted pair of solid wires and a battery and a switch. They would then feed it up the coin slot somehow and get it to over-load the photo detector , so instead of paying out two quarters it'll pay out three or four. It was an old trick.

We built a file system using public-key cryptography. When any game module was loaded into the machine, the BIOS and R-loader in our real time operating system authenticated the signature. It would take that whole program, do a fingerprint on that, and authenticate it with the public-key technology. So even if one snuck in a GAFT – a cheated program – in there, it would never load unless it was signed by corporate at Silicon Gaming.

We had a big SCSI connector right inside the machine door. One of the potential customers said, "Well, there's a SCSI connector. Someone could just load anything in the machine." We had to explain, "Sure, you can go put whatever you want in the machine. It just won't run. It'll never run." And they never thought of it that way. Instead of trying to keep the machine from being hermetically sealed, which we couldn't do, we turn it the other way. One of the key innovations in our machine was that a gambling game would not run unless it was validated. Every time the program loaded, it checked to make sure it was authentic. This innovation was the key element in the patents that we filed related to our new gaming product.

Showing Our Product at the Trade Show

When Silicon Valley met Las Vegas, Nevada, it was a very unusual thing for the gaming industry. In Las Vegas they are all into secrets, not showing any of the secret technologies. Silicon Valley is wide open, usually, and people share their stuff. That's so hard for people outside of the Valley to understand. Our attitude was, well, I'm smarter than that and I could do something even better. Whereas other people say, "I'll just steal that idea."

We had a booth within nothing it but a sign that pointed to our suite, where we gave tours on how our machine worked. My job, as the VP of engineering, was to personally give tours of our machine to all our competing VPs of engineering. And they would ask incredulously, "You're going to show it?" My reply was, "come on in. Here's how it works. What questions do you have?" It was so weird for them because they felt like this can't happen like this. And it was almost arrogance on our part. We realized these guys didn't even know what to look for in our machines. We were the fair-haired boys with these magic toys.

We had the unique idea of a networked Ethernet gaming machine. Las Vegas didn't have that. They just used a very simple polling RS-232 that kept track of who was playing the machine for merchandising purposes . We wanted to put a real Ethernet network on the casino floor to do real time communication between the host and our machines. The machines could be aware of each other. If machines in the same row were not being played their displays could be turned into video walls. These features and other features were completely foreign to the gaming machine industry. I remember talking to the guy who owned the company, perhaps the most forward-thinking networking company that did networks for the slot industry He was a young guy. "Why would you want to do Ethernet out there?", asked. He just didn't understand. And so we explained it to him. We didn't even bother to patent the idea because it just seemed so obvious to all of us around here. Three months later he runs out and patents that very thing, and two years later tries to hit us with infringement on his patent. Unfortunately for him, he had signed a non-disclosure when he talked to us. I had done the demo for him. When the president heard about this lawsuit he asked me, "Alcorn, did you get him to sign a non-disclosure? I hope so.” “Yes," I replied. And that ended the lawsuit. We were just gushing with these innovative Silicon Valley ideas but our open behavior was so non-standard in the gaming industry.

The Challenge from IGT

Our enemy, or frenemy, was IGT in Reno, the biggest slot machine manufacturer. IGT posed biggest threat to our entry in Las Vegas. They could make it hard for you to get approved. We felt that we had to work with them. So we did a partnership with IGT. We even tried to get the chairman of their board to be on our board. IGT became one our second round investors, which gave them insight into what we were doing,. But we didn’t care that they knew. We did not think that we would be to take advantage from what they saw. IGT’s chairman made a videotape in which he stated that our stuff was the future of the business, which helped us get our public offering through.

The value proposition that we pitched to the casinos, which we thought was great, was our Achilles tendon. Slot machines account for the bulk of the revenue on the casinos. We claimed that our machine could get ten percent better earnings, a quarter million dollars a year per machine. The machines that IGT was selling at the time, old fashioned spinning reels behind glass, were about $5,000 apiece. They lasted a minimum of five years, maybe ten years. But let's just say five years. If we could sell a machine for $15,000, but it made 5 or 10% more it would easily justify the increased initial cost. All the investors thought so. But in reality the way it works in these big casinos is that they buy new slot machines from a separate budget that was fixed. The casino’s purchasing logic was to maximize the budget (get the most machines) and not the total return on investment. And if they bought from IGT they put more machines on the floor.

To sell the machines, IGT would put twenty new machines on the floor of a casino in a group for three months, at no cost to the casino and they get to keep all the money. Well, the funny part is that during that three-month time period those machines would earn twice as much money as it took to buy them. What casino would say, “we don’t want them”. It was an odd way of doing business. The numbers are so staggering and they don’t make rational business decisions. So that was an interesting and big surprise to us. We would explain that each of our machines would earn more money. The casino operators replied. "Yeah, but in my casino you're not increasing a net revenue. You're just stealing revenue from those other machines on my floor." And we said, "Well, yeah. Fill your casino full of our machines and you'll steal revenue from this casino down the street." The problem was that they did want to buy all those new machines.

On the opening day of the New York-New York Casino the slot operations guy – a big, high-level guy – lost the master key to all the slot machines, the physical key! He was fired on the spot, his career was ruined, and they had to change all the keys on all the machines. And to me it's kind of sad because you're telling me that security depends on keeping a key secret? That's ridiculous. But that's the way it was. They're that paranoid. So that's one of the reasons. It's not a fun business because there's so much money that it breeds a lot of paranoia.

Then we had some corporate problems of our own. There were some decisions made about how to run a company and how to grow a company.. Being a guy who did startups and new companies and new products, I felt that when you're doing something new you have to feel your way into the market and listen and learn quickly. There are going to be plenty of mistakes and the company must be able to recover quickly and evolve the product. It's like making waffles. You throw the first one away. The president did not share my views. He was more from an old line type of business where you just tooled the machines up and spent all the money on refining the product. I have to say that I'm not sure that we came up with exactly the right content that would produce a killer product on the first product.

As it turned out, the winning machine at that time was a machine called Wheel of Gold. It had a big six-wheel on top of a slot machine and was made by Anchor Gaming. It’s hard to know what the players want. You have to build a machine and get it licensed to prove out your machine,. So it's tough. The engineers who design electronic games of chance for a casino are not the same people who design the video games for the consumer who play games at home. It's a different mindset. Today it's starting to meld because now we got a lot of gambling being done on the internet and on handheld devices. So that's starting to kind of get blurred. But in casinos at the time it was different world.

Fate of the Company

Eventually our company went under and the remains were purchased by IGT basically for the patents. We changed the storage medium from a half a megabyte ROM chip to a terabyte hard disk. Because of our cryptographic security we could show that the hard disk could actually safer, more secure, than a ROM chip. One could tamper with the hard disk all you wanted and it didn’t matter. It would just never load in the machine and play. Because these were fundamental patents, IGT uses them today to police the market.

The Nevada Gaming Commission thought it cool that, because of our security measures, one could, for administrative and regulatory purposes, you could connect to these machines through the internet if we put them on the internet. But we knew that this was a very dangerous idea since it would open our machines to many attackers.

From a Better Gaming Machine to a Better Voting Machine

I always thought that there could have been other applications beyond their use inside casinos. My most perverse idea, it would make a great voting machine. Far more secure than the voting machines that are out there, and you could put advertising on it, and if you voted for the right guy it could spit money out! I mean how good could it be! But seriously, after the debacle with the Florida chads I got seriously interested in electronic voting machines. I got annoyed when people argued, “use electronic machines to act as voting terminals and that way it'll stop all the errors." That's crazy. A pencil and paper ballots are the best and it's insane to think otherwise.

I went around giving talks about this because I think I know quite a bit, in a technical sense, about making a public kiosk. A slot machine is a kiosk – electronic kiosk for the public to interface with, as is a voting machine. And we had far better security thanks to the Nevada Gaming Control Board regulations. Why shouldn’t a company that makes a voting machine have to be vetted first for corruption? Why shouldn't the machine be inspected? How dare they say, "Oh, I've copyrighted my software. You can't see it." And Nevada Gaming Control Board inspects all aspects of a machine and respects your privacy. We will look at it but we're not going to share it with anybody else. And why can't other people do that and inspect voting machines and then try them out on the public to see if it works before – as a trial before you actually commission it?

Additional First-Hand Histories from this Author

My Development as an Engineer in the Years Before Atari

The Development of Pong: Early Days of Atari and the Video Game Industry

Video Game and Computer Technology Interaction