Breathing life into Seaman

Yoot Saito's Dreamcast game Seaman is the kind of title that makes people ask, "How did this get made?" While most of the industry was still coming to grips with 3D action games, Saito made a virtual life simulator where players monitor an aquarium, maintaining an ideal environment for a man-faced abomination to evolve from pollywog to fish to frog and beyond. As if that wasn't odd enough, the game included a microphone peripheral to facilitate conversations between the player and their virtual pet, who had a distinctly unfriendly personality. To shed a little light on exactly how the game got made, Saito delivered a post-mortem at the Game Developers Conference Wednesday.

After a bit of biographical background, Saito began by asking the audience what their preference would be if a gaming executive offered them the choice of three projects to work on: an improved Dungeons & Dragons game, an adventure game using licensed Hollywood characters, or a "rare beast of a game" like nothing that came before it. Saito said his choice was the third one, deciding to take the one that fit him the best rather than the one that seemed "most correct."

"I'm a game creator who's good at making new titles because I'm not very good at making sequels," Saito said, "even with my old titles that became hits... When you try to create something completely new, you're always going to find yourself lost at some point in the process. I want to talk today about how you're going to find yourself out of that situation and find a way to accomplish your work."

In 1997, Saito said his company had been working on two games: SimTower and a high-res tropical fish simulator on the Macintosh. The company's staff had put a huge amount of research into the fish simulator, observing and digitizing fish in aquariums within the office. As the manager of the company, Saito complained about it during a company lunch break, saying they weren't putting that much effort and that many resources into a simple aquarium simulator. He suggested that it would be more entertaining and dramatic if the creatures evolved and grew legs and arms and eventually went up on land.

The employees kept picking his brain about the idea, and Saito obliged. He decided that the fish should have the attitude of a famous Japanese comedian who ended his TV show each week by looking at the camera and snidely asking what the audience was looking at. He thought it should be like a Sea Monkey, but with the face of a man, hence Seaman.

Even after the lunch, the idea stuck with Saito. He told his wife about it and showed some sketches of the character, which she was disgusted by. A few days later, she encouraged him to pursue the game in earnest, which surprised him. He pointed out that she hated the sketch, and she admitted she did, but that doesn't mean she wasn't interested in it.

"The opposite of hate is not like, but indifference," Saito said. "That concept was a very big change in my life."

Saito decided that the character of Seaman would have to be hated by most people, rather than liked. However, he thought that it would be easy for that hatred to be switched to fondness, and the more people hated Seaman, the more preoccupied with the character they would be.

Saito settled on pursuing three "opposites" for the game, ways it would be different from the usual hit games. First of all, it would have non-cute characters. Second, he wanted the game to be a confusing mix of fantasy and reality. Finally, he wanted it to work differently than other titles from the user's perspective, essentially breaking the fourth wall. Instead of having users project themselves onto the game's characters, Saito wanted to have them feel like they were being observed from the characters in the TV, something the game's aquarium metaphor made easy to understand.

Still, Saito felt he needed guiding theories to lead the project through development, or he would have yielded to other people's suggestions along the way. Creating something that's never existed before is a battle against indecisions, Saito said, and he needed his theories to provide a focal point in the distance to strive toward.

Saito theorized that to make the game a hit, there had to be a strong theme running through the conversations with Seaman that could capture people's interest. Ultimately, he decided that the players' daily lives would provide the thematic center for the discussions with Seaman, because the one thing everyone should be interested in is themselves. He started designing conversations with an idea that players would be raising a pet, only to eventually discover the pet was raising them all along.

Saito also theorized that a game could be successful by appealing to people beyond existing Sega fans, that players would respond to the idea of "raising" an AI pet, and that using his own voice for Seaman would provide a welcome sense of mystery to the game (not to mention making recording re-takes easy and cheap).

In the end, his theories proved true. Seaman was a hit in Japan, with a player base consisting primarily of women, many of whom purchased the game and the console together.

"If you asked me what the most important thing is, it's not building theories," Saito said. "It's that you execute it not only to explore your theories in your mind, but test your hypotheses in reality and prove it by completing your titles. There are probably several hundred people who come up with an idea as unique as yours, but if there's a way to show off your inner strength, it's by bringing that to reality. Please do not be scared to be different from the existing conventional games. I encourage you to swim against the stream. Do not try to predict the future, but just make it."