One of the more interesting concepts I’m trying to implement in Sampi is one I’m calling ‘useful complexity’.
There are two basic schools of thought when it comes to item distribution in CRPG games. The first is the ‘puzzle’ mentality. If you find an item, it’s meant to be useful. You usually need to find and collect most of the items in an area before advancing, and certain distinct items are needed to continue in the game. Adventure games, most interactive fiction, and some other games (Resident Evil comes to mind) adhere to this quite strictly. It simplifies things, as there’s no way for the player to become distracted from the goal.
The other mentality is the ‘flavor’ pattern: Add as many items as reasonably possible to create a sense of immersion. The ‘Elder Scrolls’ series of games is a good example along with most roguelike games, as well as some games with procedurally generated content, such as ‘Elite’ (The old space-trading game). These games have many, many items, more than the player could ever collect or even see. But most of these are just for flavor. They serve no function, enough so as to be entirely useless. There might be 20 swords at a shop in Morrowind, but you’ll only want the best one. The rest are dead bits. There might be 30 commodities at a space station in Elite, but you’ll only care about the one with the best price difference to the next station. And Every time you come across a new piece of armor in a roguelike, you check to see if it’s better than the one you have. You’ll only trade up, so by the end of the game there are hundreds of discarded armor pieces left lying around!
There’s complexity, to be sure, but it’s all fluff. The ’solution’ is still a puzzle. You still have to find the ‘best’ item. And now you have to sort through a hundred lesser ones to find it. While these extra items can add interest to a game, make the virtual world a bit less artificial and sterile, I can’t help but think that it might be possible to do more with all of these items. Instead of having plates sitting on the shelf, make the characters use them, and have them become annoyed if plates aren’t available. Rather than having several hundred weapons, out of which only a dozen are useful, give the characters preferences for them, make them useful in different situations, make them break often enough to cause real problems. The complexity needs to be made useful.
This concept extends far beyond items. Let’s say you need to get through a door to continue in the game. In most games, this means finding the one item (usually a key, but also perhaps a battering ram or lockpick) that will get you through. Perhaps there might be a way around that you need to find. Sometimes there might be two ways to get through, but it’s still a *puzzle* regardless. With useful complexity, there would be a dozen ways to get in a door. Find the key, hide and wait for someone to come out, drill the lock, smash it down burrow underneath, etc. And most would be valid for any door, so that you don’t have to hunt and guess. More importantly, with useful complexity, you shouldn’t feel the need to get through the door at all, if you don’t want to. You could go make your own door. Or stage a protest outside the door.
The idea is to reward creativity, rather than luck and perseverance. There is enough value placed on those two in real life as it is. I don’t see why we would want it in our entertainment.
The whole problem with this approach, of course, is that nobody wants to micromanage that kind of detail. If there are a dozen way to get through a door, you’d have to remember all of them, and then choose the one you actually have available at the moment. And nobody has that kind of time or concentration.
My solution is to have the game do the memorizing for you. You’ll walk up to the door. It’s locked. The game will remind you of the possible options, suggests several salient ones, and then give advice on how to best accomplish the one you choose. Thus, rather than trying to ‘beat’ a puzzle, the player is only left to decide which option is the most interesting!
In other words, the player will try to pick the option that makes the best story. Which brings me to the next concept: I want Sampi to be an interactive story more than I want it to be a ‘game’. More on this later.
