Postmortem: Pac-Man, Iwatani's rhapsody in yellow

Toru Iwatani joined Namco in 1977 and took charge of the game and visual design of 1980's industry-defining title Pac-Man. Since then, he produced more than 50 arcade and video games, such as Point Blank, Alpine Racer and Time Crisis.

He wrote this Pac-Man postmortem for the December 2005 issue of Game Developer Magazine [PDF], and we've taken the liberty of reprinting it below to mark the 35th anniversary of Pac-Man's release in Japan on May 22, 1980.

The original concept for Pac-Man was the result of my desire to create a game that everyone could enjoy.

With a female target audience in mind, I wanted to create a game based on eating (which is why the name comes from the Japanese onomatopoeic word pakupaku, the sound one makes when opening and closing one’s mouth while eating). When I was thinking about this, I was at a restaurant and noticed a pizza with a slice missing. I thought, “This is it!” This was the inspiration, and it became the shape and general concept for Pac-Man.

Around this time, game amusement centers were saturated with games where killing aliens was the main objective. Lots of these games had great concepts that were fun to play, but I felt that none of these were equally accessible to women. Games at the time lacked variety—these types of games had a rather brutal image and a largely male audience.

I wanted to liven up the game amusement centers by bringing female gamers, as well as couples, to the scene. I was inspired by several of Atari’s games in this regard; they had some innovative concepts which taught me a lot about design. I had no doubt that the concept for Pac-Man would appeal to women, even though I didn’t spend a lot of time seeking their opinion on the ideas.

After all, even in the fashion and jewelry design industry, you have male designers creating items for women. I was confident that my creation was something that women would find appealing, so I just used my own intuition. When drafting the original proposal for the game, I kept Pac-Man to myself.

A promotional photo of Iwatani with a Pac-Man pizza, a nod to the game's origin story

When I finally showed the proposal to my boss and colleagues, the response I got wasn’t all that overwhelming. I had already created three other games prior to Pac-Man, so I was already known to be a game designer among my colleagues. As a result, most people were able to look beyond this different kind of concept and let me explore the game’s possibilities. 

What went right

1. Low-pressure environment 

When we were making Pac-Man 25 years ago, we didn’t have the same budget constraints or deadlines that most developers encounter today. Without this kind of external pressure, we were able to create something we were all very satisfied with in an environment that supported creation. Even so, we weren’t able to include absolutely everything we wanted, even in 1980!

2. Small team values

Unlike the large-scale projects most developers work on today, our team consisted of a mere five members, so it was easy to control workflow. Communication problems and team chemistry were not an issue, since we were all so close. In order to effectively manage a large project, everyone on the team needs to think somewhat along the same lines—to work toward a common goal.

If you leave out the communication aspect, your team will fall apart. The process of making games today is much more complicated than it was in the Pac-Man days, so we were able to thrive with a very lean development team.

3. Simple design

We succeeded in making the game simple. When we were trying to bring out the exciting and fun elements in the game, we mostly used a trial-and-error approach. When working on the design though, we had to remind ourselves not to sacrifice the game’s simplicity, since it was the one basic idea we had agreed upon when we began work.

We wanted to make a game that would appeal to all levels of players and all genders, and included settings that would adjust the overall difficulty over time, such as attack waves and run delay zones for the ghosts. We put a great deal of effort into including these settings so it would be challenging for a wide range of players. It took a significant amount of time for us to playtest and to fine tune the various factors to reach the right level of balance.

What went wrong

1. Hard sell

Since so many of the popular games at the time were similar, we had some trouble explaining the basic game concept to both our colleagues at Namco and the general public. It was especially difficult for us to explain the concept of how Pac-Man, who spends most of his time avoiding ghosts, is able to turn the tables and chase the ghosts after eating a power cookie.

Iwatani toys with an original Pac-Man machine in this photograph published by Wired in 2010

Internally, we received a lot of suggestions on how we could improve the game and make it easier for players to understand. For example, the president of the company requested that we change the color of all the ghosts to red, even though they all had independent AI routines and personalities.

Fortunately, the game explained itself when people saw or played it, so our concerns about confusing players were largely unfounded.

2. Artificial intelligence

The one area of the game I would have liked to refine more if given the time is its artificial intelligence. I am pleased with the AI that exists in the game, but I would have liked to implement a system where the difficulty of the AI is automatically controlled.

I anticipated that having a set difficulty curve would not be enough to cover the entire range of people that would play it. We wanted to have a system in which the computer could tell whether a beginner or an expert player is at the controls, based on the time it took the player to make a mistake. A record of these mistakes could be kept on the system, so it could automatically adjust the level of the AI in real time. We just didn’t have enough time in our 15 month production schedule to implement that system.

We also considered a few additional game features, such as gates that would trap the ghosts, but these also had to be sacrificed in order for us to finish on time.

3. Technical difficulties

For the most part, we didn’t run into too many technical problems during development. The developers on our team were all quite good, and as mentioned, we chose to keep the game simple in order to avoid potential technological issues. I think ultimately, we were able to create a simple game solely because there were techniques we didn’t know about, through which we actually could have realized a more complex design. In that sense, it was better that we didn’t know about them! 

The one technical problem we did encounter was with the animation. We ran into a few issues animating the non-game scenes, but managed to still create something compelling with a lot of personality. We had always wanted to create an animation style that people could chuckle at, since we wanted it to appeal to multiple audiences.

We knew that in order for the game to become a successful franchise, we needed a main character that was positive and upbeat, and would make players smile while they played. The same goes for the ghosts—the game needed an enemy, but we wanted something with a cute design. Fortunately, we were able to work around this one technical constraint with some excellent programmers to make Pac-Man come alive. 

Pac to the future

I’m delighted that Pac-Man has gained such popularity worldwide. We knew we had a high quality product after we finished the game when we realized there was nothing that could possibly be added or removed to make it better, aside from the minor issues I mentioned.

Even still, we were uncertain as to how successful the game would be overall. We had no idea that it would become such a big hit around the world. Even after its success in Japan, we didn’t anticipate the overwhelming reaction from overseas. Even young players today know Pac-Man due to its consistent presence on consoles generation after generation, but most of all, I have the simple game design and adorable characteristics of Pac-Man himself to thank for this success.

As the developer of Pac-Man, I was able to get to know a lot of people that I otherwise would not have had the opportunity to meet. These people are all very important to me and I feel fortunate to have worked on such a special game.

All of us who worked on Pac-Man learned some important lessons that can apply to any game developer. First, it is important to make a game that your target player will enjoy, and not just a game that you would like to play. In that respect, it’s important to have a service-oriented mindset. Additionally, developers should believe in themselves and strive for their goals, knowing that they can succeed at anything with bravery, energy, and a sense of mission.

Game Data

Developer: Namco

Publiser: Namco [Japan], Midway [United States]

Platform: Arcade

Arcade hardware: Z80 main CPU, Custom 3 channel 4-bit WSG sound chip

Release date: May 22, 1980

Number of developers: 5

Development time: 15 months