I am currently struggling a bit with the concept of failure. In a normal game failure is part of a gameplay mechanic, it is part of the loop that the game wants you to go through. For instance if a fall down a hole in Super Mario I have to restart further back and retry. The game is designed around this and it expects the player to fail sooner or later. When playing a Super Mario, I do not really mind failure as such, it is part of the experience to retry challenges and to become better at them. It is classical "theory of fun" stuff.
When we play a game that is not about mastery, as Michael alluded in the first post, we want to make sure that the player has a smooth experience as possible. Once we start using mastery loops all sort of problems pop up, that I am sure all of you are familiar with. The most important one is that it takes away focus from the narrative experience and the game's fiction gets viewed as systems.
However, there are instances when not having failure creates the opposite effect. In our new game, we will have dangerous elements that the player is supposed to fear and see as threats. That the player view them as threats is a core part of the experience and we build a lot from this*. So it is crucial that the player has these feelings. Now, it is pretty easy to build up a sense of threat without having a failure. Knowing that the threat is out there is enough and so forth. But it is not possible to sustain this state indefinitely, there comes a point when you need to show your cards. When this happens, you either have something or the player will call your bluff. For instance, in The Path, there is not really a fail state, BUT eventually the wolf will get you and your journey will be over. This is not really a normal fail state, but it is a sort of bottleneck or possibility collapse. I think The Path solves this very nicely, failure = end of this part of the game, and it works very well from the set up. (Sidenote: There is a lot of brilliant stuff in The Path, that I wished had gotten more attention, perhaps time for blog post on it
![Wink](https://notgames.org/forum/Smileys/default/wink.gif)
)
This sort of collapse does not just happen when the player is out to "troll the designer", but it can happen even when the player is playing her role perfectly. Of course, we can do things to delay it (player is merely hurt by the threat, etc), but these are just stalling tactics. Sooner or later we need to own up to our claims.
How to solve this? Here is a just a quick list of thoughts:
*Just do it the gamey way and restart from checkpoint! Focus on making it unlikely this state occurs. But with enough threats the likelihood is pretty high of it happening.
*Do plot branching and make the death matter. This makes things seamless, but it is extremly hard especially if we have more dynamic situations where we are unsure of the possible states that can occur.
*End this scene, and continue to next. Only works in some cases (like The Path) and might destroy the narrative.
*Have some story related thing happen, eg inflicting pain on an NPC. This can be really hard to sustain though, and really just postpones the owning up part.
*Make failure compulsory. Sort of like how it works in the path, but feels like cop out. Problem sometimes is also that we want to avoid the failstate as much as possible.
*Threats are often best when the player has most of it in their minds.
*Remove the threat. This just is not possible all the time, sometimes the entire experience hangs on the player feeling threatened (as in our case)
How to solve? I dunno. I do not think there is one good solution to all this, but I at least want to get some better way of thinking about it.
*This is actually a new route we have taken very recently, but have to get in on that some other time.