Monday, 23 February 2015

February Open Thread: Failure and Challenge

I think the only AIF game I ever failed* at was BBBen's Pervert Action: Crisis. From the beginning I tried to have a diverse set of skills and interact with multiple girls. At the end, I ended up in a relationship with none of them.

A somewhat similar type of failure though happened in Goblinboy's games where I didn't get the ending/relationship I was shooting for. I think in my first play of School Dreams 3 I managed to piss both Becky and Molly off.

Now, for me, those setbacks just made me want to start again from the beginning. What are other people's thoughts on the issue? Does the potential for failure, and challenge** to get better outcomes, make the ultimate result more worthwhile, or just frustrating?

*I'm using a broad definition of fail here, I enjoyed my first play, so on that criteria I didn't fail, but I didn't get what I wanted (which was possible to get)
**I'm speaking generally. I think people generally agree that too much or too little challenge isn't desirable, so I'm asking in general is more challenge good?


  1. Possibly quite a useful topic for me - I'm trying to work out right now what kind of potential failure states I should include in my next game, whether I should have game over states, etc.

  2. Arguably, the possibility of losing (ie failure) is a defining quality of games in general. If you can't lose, you can't really win either. It's important in fiction as well, since a story where it's obvious the protagonist is going to prevail isn't very interesting. That's why there's often a moment in the story where the defeat of the heroes seems imminent (eg. Return of the Jedi, when the Emperor reveals that the Rebels are walking into a trap).

    It's difficult to implement that kind of fictional situation (or anything that relies on the protagonist being fallible) in IF without it seeming contrived or railroaded, but results where the player fails, or at least doesn't succeed 100% in their goals are probably the next best thing.

    The important thing about failure states in games is that it shouldn't feel arbitrary, and the player should realise what they did wrong, ie. it should be a learning experience for them and give them some idea of how to overcome that obstacle, which encourages them to try again. At the same time, you don't want to confront the player with opportunity after opportunity to fail because then it becomes meaningless.

    SD3 is a good example of doing failure states well, in my opinion. It's very difficult to fail unexpectedly during the game (you have to do something that's obviously going to have negative consequences in order to end the game prematurely). The player is more likely to get at least a consolation prize, even if they don't achieve their goals (and those goals are the choice of the player anyway).

    Not sure if that helps at all, or if I'm just rambling. Incidentally, I'd recommend Greg Costikyan's essay "I Have No Words & I Must Design", which among other things talks about the importance of failure (or more specifically, opposition) in games.

    1. The most significant counter point to failure states is just that AIF tends to usually fall in the 'adventure game' genre, and many of the best examples of that genre don't have fail states. This is intended to encourage exploration of possibilities, experimentation with objects, etc. Depending on the style of game and puzzles this may be less necessary.

      I'm personally curious what the zeitgeist is on this topic among current AIF players. Most IF runners nowadays allow an "undo" command after a game over, so that takes away one of the bigger inconveniences of failure that used to be a hassle in AIF games. Although if your mistake was made forty turns ago then it's less useful, and that's likely to be an issue for my own projects.

  3. The "adventure game" genre hadn't been always fail safe, but by now it's very much time that it has become like you have said (frankly I prefer so, I adored "future wars" for example, but how much time lost before finding the gas capsule and that didn't help me to like the game).
    In any case now the majority doesn't want games too much difficult, a game like dark souls would have been of medium difficulty years ago (but to be honest there are even people who adore rogue like games).
    It's difficult to implement a good balance between failure and challenge, and the thing is subjective and changes with the time, years ago I played "Gloria" from the beginning innumerable times for a part of story I wanted to find, but now I'm not sure I could use so much time on a game repeating the same scenes until all choices are right.
    The only thing that I can say is that it's important you put some sexy parts even when you fail (the "a" in aif is important for the community).
    I can replay a game for see more, but I need to have seen something that pushes me to wanting more, a girl has to be very incredible if you continue to date her even if she hasn't not even kissed you once after several appointments.
    SD3 was a good game of doing failure states well, but this was possible also because was a big game with much material whatever roads you choose (and, moreover, for my personal opinion, the failure states has always been one of the strong points in Goblinboy style).


  4. I think the question of whether you, as a player, want to be able to lose and therefore like or dislike this 'option' in an AIF game depends entirely on why you want to play it and what you are looking for.

    For somebody who is looking for a challenge and wants to "earn" his progress, the pretty pictures and sex scenes, failure states may very well be a logical necessity. Without being able to fail, you can't really succeed, as ExLibris pointed out.

    But for somebody like me who's not looking for a challenge but for an interactive story that is well-written and good, interactive (hardcore?) erotic content as part of that story, failure states somewhat detract from the experience. Actually, they make it a lot more likely that I am going to look up walkthroughs or ways to cheat my way around difficulties because I'm just not looking for this kind of gameplay.

    Add to that how clunky AIF can be sometimes when the game is looking for specific keyword or keyword combination but won't react to a synonymous set of keywords which I've been typing in FIVE GODDAMN TIMES ALREADY ("WHAT DO YOU WANT FROM ME, GAME?!") and you see why I like to rely on the services of ExLibris and other kind souls who provide guides for many AIF games.

    It may just be my opinion but I think that the first and foremost requirement for an AIF game to be enjoyable is that it is well-written and sexy. Literally anything else, including gameplay, comes after that.
    A fantastic story with very simplistic gameplay will almost always be more enjoyable than a simplistic story with intricate mechanics.

  5. I would say that, like most "challenging" games, the answer lies in good, fair design.

    I too "failed" my first time through PAF, but I could (correctly) guess that it was due to making the assumption that the game ran on the same system as PAC, which I'd gotten down to a science, so that going after three girls at once was impossible. Armed with more information and better planning, I knew that I could do something different this time, and acquire different results.

    This kind of "challenge" balance is probably most important in games with significant timed or random elements, since those sorts of AIF are more "game-y" than their brethren. When I lose in one of them, it's kind of important to believe that I lost because of my own poor decisions than because the game is being a cheap, punishing son-of-a-bitch.

    Then again, I am a born save scummer, and I openly admit it. The old LucasArts style of "no failure states" gameplay aside, if the game offers me a "difficult choice," my first impulse is to save, then try both to see what happens. That, I think, is one very concrete way that IF has an advantage over other games: saving anywhere at any time, even mid-conversation, is a good way to make failure seem less punishing to the player, since smart preparation can take the teeth out of it.

    ...That's from a gameplay perspective. From a narrative perspective, a "bad end" can still be done in a way that feels emotionally satisfying to the player. Anecdotal evidence is useless, but to make a long story short, I ended a game called Mecha Ace with every other major character besides the protagonist dead, and him living out a lonely life as a small-time noodle shop owner. But, because of the writing and the narrative resolution that preceded it, it felt like a bittersweet conclusion rather than a tragic one.

    1. "When I lose in one of them, it's kind of important to believe that I lost because of my own poor decisions than because the game is being a cheap, punishing son-of-a-bitch."

      I think that's probably the most important thing to keep in mind if you're implementing failure states in a game. If they're triggered they have to be the player's fault. Not the game, and not random chance.

  6. As with most aspects of game design, I think this boils down to the question "does it make the game more interesting, or more frustrating?"

    Instantaneous fail states (ie. you perform the wrong action and it results in a 'game over' message) don't offer much of either in most cases. The availability of 'undo' makes failure no more than a minor inconvenience, but unless the possibility of failure is signalled beforehand (eg. as the consequence for failing to find a bomb or win a combat) it doesn't really add anything either. As mentioned above, the existence of instantaneous fail states can discourage exploration, so in general I'd say that they should probably be avoided unless finding the failure states is part of the fun (as in many MC and TF games).

    I'd make an exception for situations where it's obvious that there will be negative consequences. If a game lets you jump off a cliff rather than telling the player that the PC wouldn't do that, then it contributes to the immersion. A good example would be raping one of the girls in SD3. It's obvious that that's going to end badly, but the fact that the player has the freedom to choose that path makes the game feel deeper.

    Delayed failure states (ie where you do or don't do something that forty turns later ends the game) are a bit more risky. If the player is invested in the game they may be eager to get back on the horse again or simply find that the difficulty makes the game more enjoyable to them. But equally they might be annoyed that the author is forcing them to spend even more time playing this game (especially if they'll have to repeat a lot of the same actions and content) and decide to play something else.

    Failure states where the player doesn't achieve all of their goals, but the narrative still ends in a satisfying (or at least believable way) fall on the interesting side of things for me. They basically amount to plot branching, so the player knows that if they try again and do something different, their playing experience is not going to be entirely the same, which can help sustain their interest.

    1. To clarify, I would personally say that outright delayed failure states, (like not picking up the towel in the Hitchhiker's Guide to the Galaxy game and failing much later because you don't have it) are pretty much always a bad idea. Where it's foggier is in a more roguelike case, or some RPGs, where you might be playing fine for a while but eventually your poor choices (or bad luck) make it impossible to realistically complete the game without restarting.

      If it's luck, based on random equipment drops or something, then I'd say that's bad design for anything but a roguelike. In the more RPG style I personally think it works, and the positive feedback I've got about the Pervert Action games' gameplay tends to suggest that's what works in them.

      All that said, the anonymous poster above reminded me that I did originally get in with the idea of game design that was more about being a type of interactive fiction, rather than a challenging 'game'. So I agree that sometimes good gameplay can actually conflict with telling an erotic story that's designed to deliver an arousing experience rather than a challenge.

  7. Completely off-topic, but I've heard that Google/Blogspot are changing their policy regarding adult content and want to ban blogs containing such soon.

    Since AIF Central has a adult content gate, is this change in policy something that is going to affect the blog (and others like AIF Sans Mystery et cetera) and that everybody involved is ready for? Is it not as much of a blanket ban as it seemed to me?

    1. You can see the details here:

      Briefly, the change is that "sexually explicit or graphic nude images or video" are now banned. That shouldn't affect AIF Central, or most AIF blogs for that matter (although I will be backing up, just in case).

    2. ...and Google have reversed their decision (!category-topic/blogger/jAep2mLabQY)

    3. The reasoning they gave for the reversal is really interesting - basically they said it's okay if people want to post porn on their blog to express their identity.