Roller Coaster Tycoon: EGIS

Until now, I discussed topics like game fantasies, game context, themes. Generally I focused on things which could be considered narratologist in nature. I consider the debate Narratologoy vs. Ludology pointless and refuse to picks sides. Hence, is time to start some “ludology” for a change.
I thought about the following concept for some time and I’m glad to finally get it of my chest. It is about how games die…

It all beigns simple and fun…

Let us take look at Roller Coaster Tycoon. In case you do not know, Roller Coaster Tycoon is a great game by one of my favorite designers Chis Sawyer where you build your own theme park. There are millions of different details you can adjust and customize. The joy comes from tinkering with your roller coasters and finally watch tiny people hop on board, scream, enjoy the ride and vomit afterwards. It is quite like building one of those miniature train model dioramas but with more vomiting.

Every time I play the game, it is great at the beginning but as the game progresses, I seem to loose more and more that initial joy. The game turns gradually from being fun to being a chore. The gameplay seems to fall apart, it looses stability and the game dies. It is quite disappointing since the game provides the feature of an “endless mode”. The name implies a game mode where the fun goes on and never stops. However even if it is technically possible to continue playing the game forever, in reality you might want to kill your theme park very soon and start a new one from scratch.

I have noticed similar problems with other games. For example Civilization, X-Com, Final Fantasy Tactics Advance and of course, Sawyer’s other titles – Transport Tycoon and Locomotion. In all those games you tend to arrive at a point where starting a new game will be more fun then continuing playing the old session.

Here is a model that explains this phenomenon. A game can be seen as a collection of systems which interact with the user and process data. For example, the graphic engine is a system that displays data on the screen for the user to read. Those systems are designed to operate at a certain, optimal range of speed and data complexity. However, we know from flow theory, that in order to create a flow state, a game has to increase the challenge for the player as the player learns how master the game. This means that in every game, the speed at which the systems operate and/or the data complexity the systems process will gradually increase. Ultimatively one after another, the systems will be pushed outside of the parameters they were designed for. They will gradually stop working or work differently. I call this phenomenon “End Game Instability” – EGIS.

In an updated flow chart, flow is finite.

Roller Coaster Tycoon is a perfect case study. Take the graphics engine. While the isometric view reveals a great deal of charming details at the beginning of each session, very soon it becomes impossible for the player to see anything because of all that details. It dies a painful EGIS death. It is not a pretty sight since it happens gradually and momentum might carry the player deep into the EGIS area before he realizes that he stopped having fun a long time ago.

See the full screenshot in its EGIS glory.

The symptoms of EGIS are intriguing. Players often develop strategies to lessen the impact on the delicate systems in order make the game work even at high parameters. For example, Transport Tycoon offers the possibility to build road vehicles and trains to deliver goods. Every vehicle needs to be replaced by the player manually after 50 game years. If the player has trains, the process is simple since there is mostly just one train on each route. However, road vehicles require a lot more care because you use at least 4 trucks on each route and each needs to be replaced seperatly. Late in the game, all you do is replace vehicles. Experienced players will choose to use trains only and refrain from using road vehicles completely.

Game designers also try to help players deal with systems vulnerable to EGIS by bypassing them with special functions. Mostly they involve some sort of automatization. If a game has such function it is a clear indication that the game suffers from EGIS. A good example is the inclusion of the “Automate Worker” function in Civilization III as the management of worker units becomes a chore late in the game. Roller Coaster Tycoon allows players to quickly turn off certain details of the graphic display like trees or supports. The problem with these solutions is that they render a part of the game obsolete. Why building a controllable worker unit in the first place when the first thing the players does is to give up that control? Why including all those graphic details in Roller Coaster Tycoon if players will turn them off anyway?

In any case, systems vulnerable to EGIS are wasted time for game designers and players. Game designers will spend time building workarounds and players will spend time using those workarounds. If everything works perfect, the vulnerable system will remain unused anyway so it might just as well be excluded from the game. Here are some ideas how to battle EGIS:

  • Build more robust systems. Test them to make sure they work even late in the game. If a system cannot be made stable enough for end games, drop it and replace it with more stable features instead of adding workarounds.
  • Especially pay attention to EGIS when using procedural content. Consider switching to handcrafted content if EGIS strikes. Handcrafted content is immune against EGIS since it is static.
  • Quit the game before EGIS occurs. Test the game to find the limits and design winning conditions accordingly.

Oh Will Wright, you smile now but EGIS is SO out there to get you and your players…

EGIS reveals a special characteristic in games. They are suicidal in nature. They encourage the player to push them to a point where they fall apart. Game designers often do not realize that. Instead they envision games as hedonistic perpetuum mobiles capable of delivering entertainment for a unlimited period of time. Hence, they will implemet silly features as “endless modes”. Wake up! All Games die! Do your game and your players a favor and prepare – let the game die in dignity.

Roller Coaster Tycoon

Krystian Majewski

Krystian Majewski was born in Warsaw and studied design at Köln International School of Design. Before, he was working on a mid-size console project for NEON Studios in Frankfurt. He helped establish a Master course in Game Design and Research at the Cologne Game Lab. Today he teaches Game Design at various institutions and develops independent games.

2 responses to “Roller Coaster Tycoon: EGIS”

  1. Yu-Chung Chen

    Does EGIS occur (you sure love to come up with nice terms :) when the “flow indicator is too high”, or does the experience line drift off the flow channel?

    Let me rephrase what I gathered from your article. I think there is a finite amount of skills to be learned and refined in each given game. After its mastery, any attempt to rise the challenge without a corresponding skill development becomes a mechanical task, which easily leads to fatigue.

    Therefore, the horizontal axis of the diagram, indicating abilities, should end somewhere. Maybe the EGIS is then on the right side of the diagram: when the challenge continues to rise while skills cannot be further developed (i.e. going more to the right).

  2. Krystian Majewski

    Yes, I like to come up with new terms. I think linguistically, we lack the proper tools to talk about game design. We need to invent new terms.

    EGIS occurs when the challenge level of a game is higher then the game was designed for. The flow channel fades away in the EGIS zone as the rules of flow no longer apply.

    You are right. I too thought about adding a second EGIS area at the right edge of the flow chart. While the upper EGIS deals with the limits of the game systems, the right EGIS would be about limits of the humand body and mind – cognetics and ergenomics. The crazy japanese Tetris players come to my mind ;-)

    By the way, I wanted to add a nice example of how a game deals with EGIS – in Warcraft II, III and in Starcraft there is a limit on how many units you can build. With the editor you can remove the limit so techically, the game is capable to support a higher number of units. It seems like the game designers decided that higher numbers of units have bad influence on the gameplay and thus, introuduced a limit.


Game Design Reviews is a Blog used by a group of game designers from Germany to publish and discuss their thoughts on various games. The blog consists entirely of reviews of games. Each review focuses on the important game design ideas and concepts of that particular game. We also run a second, more informal Blog called Game Design Scrapbook.


follow Krystian on Twitter
follow Yu-Chung on Twitter
follow Daniel on Twitter