Flash of Steel header image 2

The joys of ceding power

February 10th, 2006 by Troy Goodfellow · 5 Comments · Uncategorized

Some of my favorite games are those that don’t let me control everything.

This may sound a little odd coming from as devoted a devotee of the god-game as I am. The whole god game idea suggests that some benign (or malicious) intervention from the player can set things aright. The player sets the agenda, builds the empire, raises the cash and kills the foe. Most games, in fact, leave all important decisions to the player. Leveling up a character, choosing a weapon, opening a door, picking a starting pitcher…not a lot of stuff is put out of your reach.

There is a risk in not giving the player complete control. What if those factors out of his/her reach seem arbitrary or unfair? What if even truly random events make the player feel he/she is being punished for trying something creative?

Soren Johnson’s Civ IV afterword mentions that the early religion mechanic was designed to have the various faiths travel along the trade routes from city to city – much as they did historically. Christianity travelled the Mediterranean and Silk Roads, Muslim traders exported their beliefs across the Indian Ocean, Buddhism and Hinduism led to a number of syncretic faiths wherever they met. The problem with this design decision was that unless you let players control the trade routes – a micromanaging nuisance in Civ II – they couldn’t control the spread of their faiths. So, missionary units were introduced. Player control won out.

Even when games go through the trouble of developing complicated AI preferences, they often default to player control. Supreme Ruler 2010 gave the player a cabinet to help make decisions. They would even override some of your decisions if they didn’t match the AI’s ideological profile. If you raised taxes but had a conservative finance minister, you might find that he has fiddled with the sliders. Great idea. The problem, of course, is that there were so many menus that you might not even notice. Add that domestic policy is a pretty important component of your larger strategy and you can see why most players clicked a box that said “do not override my desires, people”. The mere inclusion of such a choice made the painstaking work in programming this code superfluous.

Now take a look at Take Command : Bull Run 1861. This is a civil war battle game that sort of looks like Manassas: Total War. You control the troops and tell them what to do. But the beauty of the design is that the game is much more fun when you don’t control the entire army, but only a brigade or a division. It is much more satisfying to be a smaller fish, command your men and then see the entire battle develop around you. Once again, the option to control everything is there, but the design pushes you down to ground level.

Probably the best example of the tension between tight and loose reins is The Sims. Sure, the entire game is based on your bossing around a cartoon game who speaks gibberish, but the game limits you to controlling one Sim at a time. So while you tell Mama Bear to get the porridge ready, Papa Bear is out in the yard seducing Goldilocks Goth. If you turn “free will” off altogether, the game loses much of what makes it a magical title. Juggling a family, keeping them mindful of their priorities, raising the kids…all are too easy if they just do what they are told all the time. The Sims is all about recognizing the limits of player power.

Recognizing these limits is something that gamers used to accept more readily. The business cycles in the original SimCity were very basic and mostly immune to player influence – especially at the early stages of the game. SimEarth, SimAnt and a host of other Maxis games were more about making an environment for things to work in than giving those things orders to fill. Similarly, Populous was about building a setting for your Holy War – but you never controlled the soldiers.

In his epic alternate analysis of what went wrong with Master of Orion III, Bruce Geryk suggests a government management system where the virtual viceroys that run your planets would have good and compelling reasons to ignore your orders. A planet that makes a lot of money on interstellar tourism might balk at being told to shift into war economy mode. I think Geryk envisions some sort of pop-up that would tell you that Zarquon 5 has refused to fulfill your orders because of some reason and then you would have to decide whether it was worth your trouble to fire Zarquon’s governor.

I think this would have been an interesting, if ultimately disastrous, design decision; not because it is a bad idea, but because I don’t think most gamers would have appreciated Chancellor Spiff carving out his own agenda.

Computers today allow the player to control more. Interfaces have steadily gotten better and more intuitive (AoE 3’s monster UI excepted), processing power has increased exponentially, and better graphics resolution means you can fit more words, numbers and flashing icons on the screen. Because the player is able to control more, most players and designers think that they should. So, soldiers in Rome: Total War will instantly fall back in line if you think they are going after the wrong target. Superpower 2 lets you set every social policy for your government no matter how negligible the effect is. There are never communications failures in wargames. Troops that show initiative are inevitably told to stand down. The player is the game.

Note that I am not adovcating for a game that plays itself. That would be pointless. Like MOO3. I do think that strategy games work well when the player is constrained in his/her possibilities. You can do this in a boardgame way by simply allowing a certain number of “moves” or “actions”, but this can only work in a turn-based environment. But what good is telling me that my Gauls are “impetuous warriors!” if I can restrain that bloodlust with a mouse click? Maybe “impetuous” is now a fancy word for “takes orders well.”

In many ways, the game philosophy of limited power is an analog to the fact that governments and generals don’t control everything. Even the term god game is a little silly, since I have doubts that God himself is controlling everything down here – and if he is, I may have some complaints to register. As great as games are at making you feel like a deity, they can also remind us of the everyday constraints that even the greatest minds and powers in the world have had to confront. Interesting decisions aren’t just about outcomes; interesting decisions are often about how you deal with stuff beyond your immediate control.

Please visit the Round Table’s <a href=”http://not-a-real-namespace/http://blog.pjsattic.com/corvus/round-table/”>Main Hall</a> for links to all entries


5 Comments so far ↓

  • Patrick Dugan

    I agree and I think the “Gamer” market has such exacting expectations because they’ve been cultured by decades of conquest-oriented gameplay to demand precision in their influence. Most people (casual and non-gamers) don’t want that sort of control because they don’t want to go through the trouble of becoming such a master, they’re more interested in the experience of play for its own sake.

  • Duncan

    I think that the problem arises in what the user is initially allowed control over. If (using your last example) you are given the ability to command your impetuous Gaul warriors, and the interface is set so that you click and they move then certain actions are expected. You quickly establish that the Gauls are under your direct control. Which means that you can stop bad or unwanted actions with a click as easily as you can force them to do something.

    However, if the interface is designed so that you order and they respond in their own way, then that establishes something different. You expect that the units will follow, so long as you are a good leader. Or perhaps you can discipline willful units into submission. The more you bend them to your will, or the more they trust you, the more likely they are to follow your orders (and perhaps the better or worse they will perform).

    Further down the scale would be something like Majesty (http://www.majestyquest.com/) where you have influential control, managerial control, but no direct control of any units. You can set things to do, goals and incentives, but the units drive themselves. You can’t stop them from being stupid and getting themselves killed, but you can control what types of units you recruit and sometimes provide magical support.

    The interface often determines how much control the user expects. If you give them a reason to believe something will work, they will expect it to work that way all the time. If you take the control away later, then the user will be confused about why. If you want certain agents in your game to be able to act on their own, you have to establish their independence early on and then hold them to a consistency. If you want the user to reclaim control, then it should be done in a way that makes sense in the context of the game (fire your advisors for insubordination, then the next ones will be more wary – but, perhaps, less helpful). If you break context, you takes something away from the game.

  • Troy Goodfellow

    Great comment, Duncan. Player expectations are everything, and these need to be established in the context of the game design. However, players also come in with expectations drawn from years of other control schemes in other games.

    Majesty is a great example of a hands-off game – one that put you in a world and then gave you very little direct control over the things going on it. It is the ultimate detached-ruler RTS.

    In many ways, it is more similar to a city builder like Caesar or Zeus. It really is a kingdom simulator.

    As fondly remembered as Majesty is by many gamers, though, the fact is no one has picked up the cudgel for RTS that let you act as a theater commander more than a shepherd. You can see the successors to CnC, StarCraft, AoE, Myth, etc. but where are the Majesty followers?

  • Chris

    This is a topic close to my heart…

    The whole concept of Ghost Master was to allow the player the experience of *leading a team*. The haunters make their own tactical decisions, you make the strategic decisions. The low level work is all done for you (although you can set orders if you want more influence).

    I guess if I had the perspective I had now I’d have realised this was a dangerous move for a commercial game. But, I do believe we succeeded in most of the design goals for the game, and it’s certainly not like other games that are out there.

    It’s tough to go against player expectations, though. The hardcore gamer community, as Patrick alludes, want the precision and detailed control in sufficient numbers to make it a problem if you choose not to deliver it.

    All the more reason to find ways to explode the market. If there’s one thing we really need, it’s ways to get games into the hands of players who enjoy them without having to go through the conventional hardcore gamers first. It’s a tough goal to achieve without a big marketing spend.

    I also want to fervently agree that ‘God game’ is a bit of a misnomer – almost everything we’ve seen so far gives the player the capabilities of a powerful wizard. Nothing has really quite managed the experience of being a God of even Olympian influence, let alone a god of Brahmanic or Yahwehian proportions. If only Spore were top down instead of bottom up. :)

    Take care!

  • EU: Rome Updates

    […] leaving it to the people on the ground to work things out. Though I’ve written before about the pleasures of ceding power, I wonder how well this will work. Even in CK you had quite a bit of control over who would succeed […]