To be completely honest, I haven't given game design much thought the past few days. I've been working my day job as a mild-mannered waiter, and fetching someone another cosmopolitan doesn't allow for much deep thinking on game design (though it does encourage one to really, really, really want to get out of waiting tables); on this subject, I recommend you read http://waiterrant.net/ for the low-down on how horrible this profession really is. Thus, I've decided to start doing some work on game design. Besides, I promised my friend Sara I'd start writing again.

All this is preamble to today's subject, random number generating systems. This is the heart of any game system -- the way in which we generate numbers to determine the success or failure of an action, randomly. Which means dice. Oftentimes, I'll hear gamers expound on their favorite system, saying this-or-that one is more "realistic" or "better." That may or may not be true. After all, all we're doing is randomly generating a number. A game in which players randomly draw a number out of a hat is really functionally the same as rolling dice, after all.

In the end, when we talk about "systems" (by which I mean game systems), what we're talking about is the core mechanic of determining success or failure. Sure, there are many other sub-systems that comprise a game's mechanics; how you generate basic attributes, saving throws, experience points... All those elements comprise a game's rules, but I think it's the core mechanic that's the most important. As much as I like

*All Flesh Must Be Eaten*, I absolutely cannot stand the Unisystem's central mechanic, for example. (What makes that game great is the zombie design rules.)

Not a lot has gone on in this area over the last thirty years. But there's only so much you can do. The basic equation is: If [die roll] is equal to or greater than [target number], then success. That's all it is. (You can express it as the converse (equal to or

*less*than), but you're really just saying the same thing.)

Originally, in D&D, this target number was defined by character class and level. If you were a first level fighter, you had to roll over a number defined on a "to hit" table. A first level thief had a different target number. Moreover, that's pretty much all we were concerned with -- whacking the monster with your sword. Modelling any other action was out of the question because there was no skill system.

That changed pretty rapidly, however. Soon, we got skill systems that modeled swimming, or climbing walls, or sneaking around, or hacking computers. But the central die mechanic remained the same. Roll a die and compare to a value. How we determined this value could only be defined in so many ways. With some systems, the target number was defined somewhat arbitrarily, that is to say the designers created a table upon which you cross indexed values. Or the target number was defined by the character's innate skills; roll over or under a number on the character sheet. Some systems only accounted for the character's skills. Others included base attributes and skills. Eventually, this target number was stripped out and placed firmly in the hands of the referee; he or she defines how difficult a task is by selecting the target number. This is where modern game design stands now:

If [die roll] + attribute + skill ³ [target number] = success

I see I've left out the question of the random number generator. Some games use six-siders, some 1d20, some use percentile dice. Big deal. There is functionally no difference between using 1d20 and using percentile dice. Both produce a straight-line progression, which is a fancy way of saying the statistical chances of rolling a one and rolling a 20 (or 100) are equal. The only time this changes is when you add extra dice to the mix, because now you're creating a bell curve. Rolling 2d6 produces different statistical likelihoods. The average roll on 1d6 is 3.5 (or rather, either a 3 or a 4); so the mean on 2d6 is 7, with a cluster of results around 6,7, and 8. There are fewer results of 2 or 12 (the ends of the bell curve). So the majority of your rolls are going to be around 6-8, with a few at the ends of the curve (the 2s and 12s). But you know all this. And I've likely lost the math-challenged among you.

(This is why I both like and dislike the Storyteller system; you're rolling a number of dice equal to your attribute and skill, and looking for the number of successes, but I don't possess the math skills to calculate the statistical chance of success rolling 10d10.)

All of the above is a long, roundabout way of saying: I have no idea what to do

*vis a vis*game design. On the one hand, there's no point in reinventing the wheel; it's all just sublime recapitulation (I love that phrase, it's from

*The Name of the Rose*). Rolling over or under. Including attributes and skills, or not. It really doesn't matter. On the other hand, I really hate the idea of just using an existing system. It offends my design sensibilities. Moreover, I don't want someone saying "oh, it's just the Savage Worlds engine." A game's mechanics define the game. Playing

*Call of Cthulhu*should feel different from playing D&D. (This is, by the way, the reason I was so discouraged by the Open Gaming License; every game felt like I was playing D&D. Imagine playing chess, Monopoly, and Chutes & Ladders using the rules for Battleship....) And yet that core game mechanic remains the same.

Thus, I conclude that it's not the central random number generating mechanic that's really important. It's all the other attendant sub-systems that define a game.

## No comments:

## Post a Comment