Building Stories, Telling Games #42:

Don't Look Too Closely, Part 3

by Travis S. Casey
September 20, 2002

As promised last time, I'm wrapping up the "Don't Look Too Closely" series this time around.

But It's Magic!

In fantasy games, working around "realism" problems can be made much easier by the fact that you're dealing with magic. Every fantasy world pretty much defines its own rules for magic, so you can set up the rules in such a way as to make abuse difficult. This does require thinking about things beforehand, though — if using a spell in a certain way worked yesterday, but doesn't today, players aren't likely to happily accept a "that's just the way magic works" explanation.

Unfortunately, none of us are perfect. You can try to think of every way in which players can abuse an item, but you're often going to find that players will think of something you didn't... no matter how hard you try to limit them. So how can you work around it?

Well, the ideal answer is playtesting. Unfortunately, that's not always a practical answer — once a game is up and running, if you want to add a new magic spell or item, you may have no other real choice than to put it straight into the game. (Being able to run two games — the "real" one and a "playtesting" one would be ideal... but, of course, that significantly increases the resources one needs.)

One thing that can be done is to first put out a new magic effect in a limited form. For example, many fantasy games have "charged" magic items that only work a certain number of times, then stop working. Giving out a new magic item in a charged form before giving it out in a permanent form can give you an opportunity to observe it in action before it becomes a permanent part of your game. The same can be done with spells — simply give out a charged item that does the equivalent of the spell to start with.

Once you've had a chance to see it in action for a while, then you can go and create the "real" spell or the "real" form of the item — and not have to worry about any problems your first try had haunting your game forever. Since it is magic, you can easily rationalize that the spell form is a bit different from the item form, or that the permanent item is a bit different from the charged version. Why? It's magic. If we understood it completely, it wouldn't be magic any more.

Getting There Isn't Half the Fun

One area where people are very willing to suspend disbelief, and where it's generally not a good idea to enforce a lot of realism, is in travel. In a game, spending even five minutes going through areas one already knows in order to get to somewhere interesting just isn't much fun. Further, having even semi-realistic travel times gets boring very fast — in a real medieval world, it's likely to be a half-day's walk to get to the next village over.

Few people want to spend even two minutes looking at artificially-generated scenery going by, or reading descriptions of the outdoors. (They might be willing to spend longer if they're very, very well done — but few, if any, games can boast having either one that well done.)

A second problem is remembering your way around. This can especially be a problem in text games — our brains seem to be somewhat "hard-wired" to handle landmarks and directions, but we don't have that advantage when dealing with things like remembering that you have to go "n n n e n e" to get from "Dave's Bar" to "The Stone Fountain". Because of that, quite a few text-based online games have "speedwalkers" of one sort or another — items or commands that let a player associate a particular command with a memorized sequence of "moves" to get from one place to another.

Some game designers have proposed allowing players to use landmarks in text-based RPGs — this is sometimes called "landmark movement" (and the "n e n n" type movement called "direction movement"). However, many people advocating landmark movement have gotten caught up in keeping it realistic — e.g., having movement take place at a particular speed, so that you type "go to Dave's Bar" and then wait to get there. Also, if you want to keep it realistic, there's the problem of path-finding — how do you get to Dave's Bar from here? Path-finding is still an active area of Artificial Intelligence research — it's not something that can be solved 100% easily.

So what to do? Well, text-based computer games aren't the first games to have this problem — those of you who are old enough may remember that in the late '70s and early '80s, adventure gamebooks were popular. Those books had numbered sections of text, and in places the player would either make a decision or roll dice to resolve something, and based on the decision or result would be told to go to another numbered section. Moving around in most of these gamebooks was handled by having the player pick a direction to move in, and giving them a paragraph number to go to... the equivalent of direction movement.

Some of the gamebooks, however, used landmark movement in at least some places. For example, you might be told that at this point you could choose to go to "Dave's Bar", "The Palace", "The Temple of Ra", "The Temple of Set", "The Market Square", or to "Leave Town by the North Gate" or "Leave Town by the South Gate". You'd pick one of these to go to, and go to the paragraph that put you at the starting point for that place. To make things more visually interesting, the places you could go might be shown on a map of the town with numbers printed on or beside them, instead of simply listed.

In going from one to another, you simply "jumped" there as if you'd teleported. Your character was assumed to have walked, ridden, or whatever to get there, of course, but the journey was assumed instead of shown — much like a cut from one location to another in a book or movie. If nothing interesting is going to happen along the way, there's no need to bother with the mechanics of how someone got there. Once at a particular location, movement within it would generally be handled on a direction-based system. (Although elements of landmark-based movement would often be mixed in... e.g., you might be offered the choice to "go north through the door with the lion head carving" or "go east through the door with the unicorn carving". In addition to adding a bit of color to the setting, that gave players another memory cue to help them remember how to get somewhere.)

Doing this in a text-based computer game is simple and obvious — when the character exits a building, simply offer the player the list of choices for where to go next. They choose one and bam, there they are. This does eliminate any sort of time factor... but since travel times are already severely distorted on most online multiplayer games, does it really matter that much?

Some may object that this makes "random encounters" along the way from one place to another impossible... but the old gamebooks had a way of handling that as well. If you chose to go to "The Palace", for example, instead of being directed straight to the opening text for the palace, you might be directed to a piece of text like this: "Roll a die. If it comes up 1-4, you make it to palace without incident; go to X. If it comes up 5, go to Y. If it comes up 6, go to Z." The same sort of random check, and stopping somewhere "along the way" if necessary, could be done in an online game.

This technique also allowed for something else — places that weren't in the list, but that you could learn about. For example, you might be able to find out from a beggar in the market square where the Thieves' Guild was. The game would handle this by telling you, "whenever you're offered the choice to go somewhere else in the city, you can go to the Thieves' Guild by turning to paragraph X instead. You may wish to write this down on your character sheet so you do not forget it." In an online game doing this sort of system, one a different menu of places might be presented to the player depending on events that have happened to the character.

So... what was my point again before I got sidetracked? Oh, yes... a movement system like the one I've just described certainly isn't realistic — but it's convenient for the player, and it avoids "boring stuff". Because of those two factors, it's unlikely that many players would complain about it.

Recent Discussions on Building Stories, Telling Games: