This week marks a monumental accomplishment for one indie game developer in California - Jason Rohrer, is the end of a three year long process to release his DS game, Diamond Trust of London. After multiple redesigns, publishers, and various other setbacks, the two-player diamond smuggling strategy game is finally on its way into the hands of consumers as of yesterday when Rohrer shipped out 861 US-bound packages. Thanks to his final push of support via Kickstarter, the journey of the first player-funded Nintendo DS game is finally beginning to come to an end.
For someone who is used to creating games experimental indie games on his own, Rohrer's first foray into dealing with a publisher and creating a game for the Nintendo DS didn't go as smoothly as he expected. When asked about the high points and low points of the development of Diamond Trust of London Rohrer called out the Kickstarter support as a high at the end of a road of lows. "It’s been a process of things that seem like they are going to materialize, and are going to work: Oh, here’s a way this game can come out, and then it actually can’t come out this way. Oh, actually this idea isn’t going to work. Oh, actually, these people are pulling out." With more than a few bumps, fans are fortunate to finally be able to experience the final product after all this time.
At first glance, this may come across as indie developer naïveté, and the hope that the games industry values creativity over profitability, but Rohrer, who has released eight other games on a variety of platforms ranging from the PC to iPhone, recognizes how unique Diamond Trust is, and that for a lot of people it might not be the most savvy business investment,
"I’ve gone through two publishers, and it’s a risky thing to do. It’s not really a sensible business decision to throw so much money at such a strange niche game. I have an audience of fans who seem to appreciate my work, who are dedicated, and will probably buy my next game, but that audience isn’t gigantic. It’s not like 100,000 people; maybe it’s 10,000 people. It’s a big risky thing to do. It’s very hard to get someone with money to want to follow through all the way to the very end once they realize how scary it's going to be to throw money at this. I wouldn't have embarked on this project if I hadn’t been talked into it by a publisher at the very beginning."
So, how did he end up here? Rohrer has been a big name in the indie games scene for a few years now, making a name for himself by crafting games well outside of the box. He first garnered some mainstream attention with his third original game, Passage, back in 2007. More of an experience than a game, Passage dabbles with the constraints of time, direction, and partnership in a way unlike any other game. (It's still available as a free download on PC, Mac, and Linux, as well as for purchase as DSiWare and App Store downloads.) As he continued to make more free thinking games on his own he eventually garnered the attention of Majesco Entertainment, who seemed to be willing to dabble in indie dev inspired games like Blast Works, a title based off of Kenta Cho's TUMIKI Fighters, and Diamond Trust was born.
"Three years ago Majesco approached me and said, 'Hey do you want to make a DS game?' I said, 'Okay, it sounds like a reasonable thing for me to do.' With A small screen and relatively limited hardware, this was something one guy could probably do by himself, and make something that's pretty good. So I pitched a couple of different ideas to them, and they rejected a couple, and finally accepted this idea about deception and corporate espionage that took place within the world of diamond trading."
However, things didn't end up going so smoothly from there, and Majesco dropped the game as Rohrer was just about finishing up the development. "I got the DS game pretty much done, and then lost my publishing agreement, and started trying to figure out how to save that work, and how to get it out to the people who have been waiting for it." The game eventually made its way into the hands of Zoo Games and their IndiePub imprint, and they signed on as the game's second publisher. However, things didn't go exactly as planned there either.
"After another year of me finishing the game, and negotiating with Nintendo, and going through lock check and all these other things to get the game approved and ready to come out my second publisher was like, actually we don’t want to throw all this money at this game. There’s a huge minimum order from Nintendo that they would have to put money front up for. There’s a really big financial risk, If you don’t sell enough cartridges after the fact, you won’t ever recoup that money, and it’s just like money thrown down a hole. It was just a huge risk that they just couldn’t justify any more. So then I’m sitting there again with this game that’s totally done, the packaging art is done, approved by Nintendo and it’s just sitting there, unable to come out."
But that's just the story of how this interesting little title game about trading diamonds came to be, but, in reality, all that doesn't matter if the game isn’t any good. While Rohrer's reputation precedes him, there's still the question of what the game is all about.
"I wanted to make a game about deception where two players have different views of the information because they have two separate DS screens over wifi. Therefore one player can have secrets that, as far as they know, are being kept secret, but they are actually being revealed to their opponent without the owner of the secrets knowing that they were being revealed. That was kind of the core mechanic I was really interested in exploring. The diamond trade seemed like the perfect setting for a game where you couldn't trust the people that you’re sending out in the field. They’re coming back with a package of five million dollars in rough diamonds, how do you know that they are bringing them all back? So it seemed like the perfect setting. I wasn't trying to make some kind of statement. I didn’t have some bone to pick about the blood diamond trade. I was just like, 'Wow, this is a great setting for these kinds of mechanics that I wanted to use.'"
Clearly conceived as a two-player / two-DS experience, sometimes you might not have someone to play with, or perhaps just want to sharpen your game. To solve this Rohrer included a single-player mode as well. However, as you might imagine, recreating a realistic experience of playing against a human opponent using artificial intelligence for a game driven by uncertainty is a tall order. Rohrer recognized this, and did his best to account for this while crafting the game's AI.
"AI for game design is just really hard. In this case I sort of went in kind of cocky, because I’ve studied AI, and I’ve made AI game players for chess in the past, and I know a lot about AI. I’ve done AI research. But, trying to make an AI that works in a game with hidden information, and this uncertainty kind of stuff was really much harder than I thought it was going to be. There are some papers that people have published about this kind of thing, and people have come up with ways to deal with uncertainty, and I thought I was just going to be able to apply one of these things and it was just going to work. The DS is also really limited in terms of CPU speed, so the number of simulation steps, and the number of exploration steps in terms of the search space and so on that you can do while the player is waiting for the AI make a move is not very many. I had to scale back from these research paper ideas that were out there, and implement a bunch of hand tuned heuristics to help the AI out so that it would be making more reasonable moves in a shorter amount of time.
Doing this creates a multiple possible reality simulation where it doesn't know how much money you have because that’s a hidden variable. So it runs a simulation where you have one dollar. It runs a simulation where you have two dollars. It runs a simulation where you have three dollars, and so on, through all the possible range of what amount of money that you could possibly have, given the moves that have taken place so far. It averages out all those and tries to pick the best possible move, on average, based on how much money you might have, as example of what it’s trying to do in an ideal scenario. All that simulation in multiple possible worlds takes a lot of CPU resources, and takes a long time. The AI is really just a practice opponent. It’s not like Chessmaster where it’s going to be able to beat you for the rest of your life."
In the end, Rohrer's goal with Diamond Trust was on par with all of his other games – get it into the hands of as many people as possible so that they can experience his creation. He's confident that even though his game is being released on the DS that there people will still get to enjoy it.
"At this point the DS is at the end of its life, but this is three years later. When I started out on this project it wasn't as crazy to release a DS game. But still, people have access to one somewhere. It’s probably easier for many people to get a hold of this, than maybe even an iPhone. Finally, at this point, the iPhone is neck and neck with the DS for number of units that are out there. But three years ago, far more people had access to a DS than an iPhone. In terms of consoles in general, this one is pretty accessible one for most people even though it’s so old. Because it's so old is what makes it accessible. It’s also very inexpensive to get a used one, if you go to Gamestop and buy a used DS for like $50 or something. That was part of the calculus I was doing when I was deciding to make this game in the first place. It has two small screens - 256x192. That’s not a lot of art assets for one person to create. It’s got this automatic wireless kind of thing that makes it really easy to make multiplayer games that just work without having to monkey around with servers that’s hard for one person to maintain over the long term financially. It’s hugely popular and widely available. All those things together just made a lot sense for me. Whereas trying to make a game for Xbox Live Arcade wouldn’t really be quite as sensible for one person to try to do on their own."
While the Kickstarter rewards are officially on their way to backers of the game, you can still purchase a copy of the game directly from Rohrer for $30 . If you are interested in grabbing a copy, act fast, as of the time of writing this, there were only 4510 copies left, which, when compared to a run of a game released by a major publisher, is basically sold out.