I create a lot of prototypes for games, and there's a few things I've learned. One thing is about the focus of the rules and the way the world develops. Please keep in mind I'm not talking about just video games, but also tabletop RPGs, board games, card games, anything.
As I create prototypes, I find myself focusing more and more on a single core play element. For example, if I create a prototype for an RPG, I might create just the combat system, flying free from leveling and equipment and so on. This is not really a bad way to do it, but it does come with some baggage as to how the rest of the game develops. I thought I'd touch on some fundamentals this process seems to reveal.
Most games focus on several simple systems that interlock. For example, in an RPG you have a simple tight-combat system (rolling to hit, damage, armor, etc) combined with a simple tactical system (movement, buffing spells, area effects, etc). Outside of that, you have an equipment system, a leveling/character power system, a skill-check system, and often several other systems, depending on the focus of the game.
This is true of most other games. A first-person shooter might combine a simple tight-combat system (aiming, shooting, damage) with ammo management, weapon management, level layout, enemy AI response, and whatever else the game chooses.
(For this post, we'll put aside whether or not non-play elements such as graphical design, story, interface, etc matter in this way.)
By creating many simple layers that interlock, these games create an end result which is reasonably deep and interesting. The player plays by running up and down the different layers, spinning each in turn. It creates a good pacing dynamic: the player will never be caught on a single play type for very long.
The other end of the spectrum are games where there are very few play mechanics, but they are much deeper. The easiest example is a fighting game. Aside from choosing a character, nearly all of the rest of the play is on the field of battle. Sure, some newer fighting games have weird world-exploration modes or equipment-equipping modes or whatever, but let's ignore that for now. The game is about the fighting.
The fighting is fundamentally a simple play system: move in small increments, and attack at the right time. This is made complex because there are many kinds of attacks - not just throws vs punches vs fireballs, but low attacks vs high attacks, fast attacks vs slow attacks, attacks you can cancel out of. For purposes of this analysis, defending will be considered automatic as long as you aren't moving forward or attacking.
This can be thought of as a simple framework with a lot of simple components. Each individual attack is a module which pops into the framework and casts a shadow on the playing field. These simple components create a pattern that you play within, jockying for the right position and the perfect moment to launch one of your techniques. Taking the enemy's own capabilities into account as well, launching your attacks into gaps in their patterns, avoiding the points in space where they are strongest...
The end result is that a relatively simple play mechanic with many modules develops into a very complex space, as opposed to the more common method, which is that several relatively simple play mechanics are layered to create complex space. Think about it in terms of a projector shining down from "above" the game. The layered approach is a bunch of stacked transparencies that cast a shadow beneath them. The modular approach is a bunch of small bubbles and lenses put down on a single surface, distorting the light into halos and shadows and blotches as it passes through. Both are complex, but they approach it from very different angles.
This isn't boolean: there are plenty of games which are vaguely in the middle somewhere. However, while building prototypes, it began to seem like these are the two fundamental approaches: building a layer of modules, or building several flat layers that stack.
This also has a lot to say about world-building and the fiction your game can support. A game with many simple layers can draw fiction into any of those layers and it will be valid. For example, a treasure-hunter's guild in an RPG relies heavily on the exploring-for-treasure mechanic and all but ignores things like leveling and combat and equipping and whatever.
On the other hand, a game with focused "module-style" gameplay doesn't have disparate layers to draw from, so fiction either has to be crammed in as color or directly relating to the fundamental play. This is why games like StarCraft or Street Fighter have such tightly focused narratives: all the fiction and color revolves around the core gameplay, which is pretty limited to combat. You can have a treasure hunter in a fighting game, but it's pure color.
On the gripping hand, you can go a lot deeper with your fiction in a "module-style" system. If you introduce a Taekwando master in an RPG, he's not going to be very distinct from any other "monk-class" character. But in a fighting game, the fine distinctions are very clear, to the point where a fighting game's characterizations often rest almost entirely on the fighting styles of the characters.
I guess that's all I have to say about that. You?