I’m trying out different things right now; the rapid prototyping of the Source Engine pleased me greatly, but I realized that I probably didn’t want to make a first person shooter (or extensively modify the engine to support another genre of game). So, I’m trying to return to Wiki World.
The big problem I had with it was trying to figure out what kind of combat system to put in it. I’d planned Wiki World to be this really open, player-controlled game where players created everything, from the environments to the people; however, combat really does not mesh well with that formula. Combat often boils down to counting numbers: health, movement speed, damage, damage over time, etc.
So, perhaps the answer is to do away with the combat completely. The problem, then, is that essentially what I’ve created is a roleplaying engine. Now, one might argue that all games are roleplaying engines; in Starcraft, you take the seat of a commander in charge of his armies. In an FPS, you take the role of whoever you control. And so on. But no, I mean a different kind of roleplaying engine - that kind where you sit around in a chatroom and pretend you’re elves, or dragons, and the like.
Of course, it’s a nice concept, although almost certainly something someone’s done before. Either way, that’s not what I set out to make; I’d really like some core mechanic rather than simply roleplaying. So, combat it is.
The problem with allowing player creations in combat is that it is all too easy to leave some imbalances in the system. Players, after playing for long enough, will almost certainly find such imbalances and exploit them, and soon, no one can get by without knowing such imbalances. This not only makes the game hostile towards new players, who have no idea what the optimal builds are, but also limits the creations to more experienced players so that everyone is using the same two or three builds with no room exploration elsewhere. For an example of this, look at Diablo 2.
Of course, the other direction is to simply be extremely conservative in the choices available, either limiting or dampening the controls given to the player. This definitely resolves the problem of newbie-friendliness. However, the two-or-three builds problem is still there; the only difference is that instead of players finding them and possibly feeling rewarded, the builds are forced upon the players by the developers, leaving the players feeling like there’s no depth to the game. Generally, even those builds will not be completely equal. In a better world, each build would be suited to a different playstyle and the builds would complement each other. That is rarely true. For an example of this, see Diablo 3.
My original plan for Wiki World was, like I said, to have the players dictate everything about the units they control. I wasn’t clear on what genre of action the combat would be, but I was thinking some kind of turn-based-strategy, isometric-tactics hybrid. The problem, as stated above, is that both letting the players have all the controlsand still keeping balance in check is nigh impossible.
For example, say I had a generic “spell” template. This could be a fire spell, an poison spell, a slow time spell, whatever. It could even be a spell to transmute someone into a granola bar. There’s two problems here: (1) How does one create such a generic spell template? One can generalize spells into certain categories, like damage or stun, and then add some time or distance components to the spell, but you’ll never really be able to get the full spectrum of spells into just one template. (2) How does one compute the costs to these spells? Oblivion, in particular, had this problem; they had very predefined categories of spells like sleep or camouflage, but I cannot remember an instance that the cost to cast the spell was better or more reasonable than a stock, store-bought spell.
I realize I’ve just been iterating over the problems, and no potential solution. The problem I have is this: no potential solution is in sight. I believe I’m going to have to resort to predefined spells, stats and such, like in Final Fantasy Tactics. Yes, this seems like an easy way out, but really, this is a one man operation and I’d like to at least get something done. They say you can’t aim too high on your first few projects, and they’re right.
Let that be the moral of this story; don’t bite off more than you can chew.