NOTE: This is The Column, a regular feature on Apolyton where anyone can write about anything to do with Civilization or the gaming industry as a whole. If you feel like writing, please visit the article submission page.

PREVIOUS ARTICLES

#220 WHAT MAKES A GOOD CIV GAME (PART 2) In the second of four installments, Spiffor l'Omnivore breaks down the second two of eight points needed towards building a good Civ-style game. #220 WHAT MAKES A GOOD CIV GAME (PART 1) In the first of four installments, Spiffor l'Omnivore breaks down the first two of eight points needed towards building a good Civ-style game. #219 A VERSATILE GAME Rasbelin recalls the comforting familiarity of Civilization II. #218 I HAD A WHALE OF A TIME Rasbelin reminisces about the days of MicroPROSE now all in the past. #217 "CLASH OF CIVILIZATIONS": A PRESENTATION OF THE GAME Mark Everson motions for your attention to talk about a gaming project that he is currently leading. #216 THE FINAL FRONTIER Rasbelin runs through the Stella Polaris project on which he serves as Project Manager. COLUMN ARCHIVE

5. Challenge
Civilization games' very nature make them more inclined to [single-player (SP)] than [multiplayer (MP)]. They were first developed for longlasting games, and it is difficult for a non-fan to find other people dedicated to play for several days. At their core, Civ games are very different towards MP than Real-Time Strategy [(RTS)] games are. In CivIII, the modes allowing fast action are recent and are a twist of the existing system. In RTS games, fast action is the very principle upon which the game is built. So, despite the growing emphasis on multiplaying in the gaming industry nowadays, Civ games as we know them will remain deeply tied to single playing and to the [artificial intelligence (AI)].

A good AI is needed to make a good Civ game. While it is fun to crush the AI with the same old technique from time to time, a Civ game gets quickly boring if it is not up to the challenge. Soren has made a fantastic effort in CivIII's AI, and we can hope his skills and the general progresses on AI behaviour will make even more challenging AIs in the future. However, a good AI can come in the way of the game's complexity. Firaxians told several time many [Alpha Centauri] features weren't present in CivIII because the AI wouldn't know how to use them, which created an unfair advantage to human players. They have also put a great deal of effort in hunting down every exploit of a feature they could find. There are two ways to fight exploits: to fight the exploit itself, or to fight the feature that's being exploited. CivIII offers examples of these both ways:

- IFE (Infinite Forest Exploitation) was an exploit from the first version of CivIII, as it entered the stores. Everytime a forest was cleaned, the nearest city got 10 shields. With many workers, it was possible to clean a forest and replant it on the same square several times, giving tens of shields to the city nearby each turn. The AI didn't know this trick, so it was a pure advantage to human players.

In response, Firaxis forbid any cleared forest to give shields more than once in the game. It effectively killed IFE without killing forest-cleaning. It effectively killed the exploit without killing the feature.

- Unit trading. There were known abuses of unit trading in Alpha Centauri and CivII, and Firaxians feared those abuses would be repeated in CivIII. The abuse was to give obsolete units to other civilizations, whose upkeep costs would prune them dearly. Maybe there could have been answers to these abuses, like the impossibility to give/trade units that are obsolete for the receiver, or to give the AI the ability to disband a unit. We'll never know: Firaxis has chosen to scrap the feature altogether. The quest for a better challenge (i.e. a less weak AI) has cost CivIII a cool feature.

I personally don't think there should be a fanatical war against abuses, except the very easy and very tempting ones. If the player wants to play using tricks against the AI, so be it. If the players in MP want to play using abuses, so be it. Abuses aren't a problem if they aren't too easy to figure out. The fanatical war against abuses [in my humble opinion], has cost CivIII unit-trading and tile-sharing with allied civs (units from different civs on the same tile). I don't think it helped making CivIII better.

6. Various levels of micromanagement
There are different kinds of players of Civ. Some are control freaks who want to manage everything in their empire; some like playing fast games without performing "dumb" tasks; some are interested by one aspect of the game and don't want to be bothered with more "trivial" aspects; some are newbies who must not be confused with tons of actions to take. There are also players who change style as time passes by. A good Civ game must not alienate any of these players, for all of them can find a good entertainment in Civ. However, there seems to be an opposition between control freaks -who want to have as much stuff as possible to manage- and newbies who must not have a too steep learning curve. There is one solution to please all kinds of audiences, it is to allow various levels of micromanagement.

In other words, it is automation. There have been tremendous progresses in this domain since CivI, which match the progresses made by the AI. In CivIII, it is now possible to automate workers, cities, exploration, deal renegotiation... It seems the AI takes over these tasks and does them as well as an AI-controlled Civ would perform. Actually, the automation seems to use the same AI. But the concept can be taken even further, where basically every aspect of the game can be automated (i.e taken over by the AI). Tired of dispatching your many defensive troops across your cities? Automate military moves. Tired of negotiating deals to have your coal? Automate trade. Tired of selecting the target tech every 4 turns? Automate research.

A good automation, besides having a solid AI, must be easily and quickly revocable. It also needs an input from the player in form of priorities. CivIII and Alpha Centauri both allow the player to give "priorities" to their city governors, and to give precise limitations to automated workers. In short, a good automation reconciles hardcore micromanagers and more casual players, because it allows the latter to care only about what interests them, and because it allows the formers to micromanage at their heart's content. A good automation solves the tension between adding stuff and removing stuff.

---------
About the author: Civilization enthusiast.

Comment on this article!
This article was originally posted as material in this thread between April 11-12, 2003, but this version has been modified to correct previously published errors.

The opinions expressed on this page do not necessarily reflect those of Apolyton CS. They are just the personal opinions of the writer.