Tuesday, August 10, 2010

Epic Fail

It's weird to think that there exist solitary entertainment experiences wherein you can experience failure. Not failure through a cipher, but you, yourself, failed to do something correctly. Video games are unique, or at least uncommon, in this respect. Maybe some variants of solitaire have unwinnable deals, but those are wholly random. It's much harder to fail at reading a novel, listening to a song or watching a play.

I think this is actually a very good thing. After simply having agency at all, it's probably the most distinguishing characteristic of games. But it also affects how we perceive games. It may cast games as things to be overcome, rather than understood. Things to be beaten, rather than examined. For a lot of games, this alone can be deep, rich and fascinating. But I can't help but think this might be impeding our ability to communicate through a game's mechanics.

When I talked about why indie darlings are so often 2D platformers, I noted that part of the reason why so many creators opt for 2D platforming as a style is there are probably more players who are literate in that style of game than any other. It's not that the challenge or possibility of failure has been necessarily diminished, but rather that the basic structure is very well understood and internalized. Move from left to right, avoid hazards, gravity and timing behave in certain ways, etc. You place a random person in front of a 2D platformer and ask them how to succeed, and they'd probably have a good guess. Getting the same after putting someone in a random spot in Red Dead Redemption is probably far less likely.

In the comments to that post, Michal Wisniowski remarked that bullet hell shoot 'em ups are also quite prevalent in the indie scene. Which made me wonder what would be required to communicate some kind of tonal/aesthetic message through a shmup's mechanics. A great deal of that style's appeal comes from its unrelenting difficulty and split-second timing. This makes them less accessible than more playable platformers though, even for pretty experienced gamers. I mean, I look at Ikaruga, smell copper and then pass out. I enjoy some games with an intensive manual dexterity component, but shmup's are a bridge too far for me.

Given that there's so much emphasis on overcoming challenges and avoiding failure, I wonder how much cognitive space is really left for a player to pick up on anything more subtle. Of course, there's nothing to say shmup's have to be well suited for these kinds of experiences. But it seems like any type of game that does desire to communicate some kind of message that isn't just through exposition and dialog might do well to consider exactly how much emphasis is being placed on "winning." If too much emphasis is placed on moment-to-moment survival, it could be difficult to encourage thinking about the game as a larger, more cohesive statement.

The other unfortunate consequence here is that it becomes difficult to encourage players to explore outcomes that seem "negative." It's quite rare to find a game that doesn't just jump to something checkpoint-ish as soon as a failure state has been reached. Roguelikes are a rare style of game that, due to a brutal and unforgiving difficulty, failure is almost an information gathering process. Each failure grants further insight into the world and its rules. Most games do this to some extent, but the majority cast failure as an optional thing. It feels avoidable and unintended when it occurs. I'm not sure what it would look like, but I'd be very interested in a game that encourages probing "failure" without the merciless character of a roguelike. I just can't help but think we haven't seen this yet because such a game would seem bizarre and antithetical.

As an aside, Jon Blow gave a good talk at MIGS where he discussed the tension in games between progress (continue the story, allows things to evolve) and challenge (keep things from changing until certain conditions have been met). Listen if you haven't, it's solid.

Games can be thought of as a conversation between creators and players. A great deal of the time, the player is just trying to keep up with the conversation until it's over. We've seen some great experiments that helped players hear and reflect about what was really being said by slowing the conversation down, by using common parlance and familiar words. I think we'll see even more of these experiments in the future and I'm definitely excited. Maybe before long, we'll see games where you "fail" because you want to, not because you didn't do things right. As long as we've still got plenty of good healthy challenge in other games that appear alongside it, I'm totally on board. You?

Labels: , ,


Blogger Unknown said...

Demon's Souls seems, to me, to be exactly the sort of "probing 'failure'" game you ask about. When you first start out, everything you encounter is deadly and you will die. A lot. But after every failure you start to see a little more of how an enemy's attack pattern works, or where traps are waiting to skewer you. It's almost, but not quite, a roguelike and I feel it encourages failure more than almost any other game on the market.
PS. Great post, I thoroughly enjoy your blog.

August 10, 2010 at 2:16 PM  
Blogger Paul Bauman said...

I'm not sure. The concept of a game paradigm without unintentional failure seems to compromise something that makes games what they are (for me).

These aren't just texts that a user is plugging herself into to explore and experiment. There needs to be something in the game that gives the player some sense that they're "doing it wrong." Then again, in textuality there's always the intriguing possibility of the productive (and sometimes intentional) misreading.

I definitely get what you're saying about the fail state needing to be less... terminal somehow. To make it a verb that leads to other states to explore is kind of intriguing.

I agree with what Michael said about Demon's Souls, though I think the gist of your argument takes this a step further. In the case of that game, this concept would lead to something along the lines of the "death" state giving you options you wouldn't otherwise have (much like the black/white world tendency effects).

August 10, 2010 at 7:27 PM  
Blogger Nels Anderson said...

@Michael Yeah, Demon's Souls definitely has a roguelike flavour that fits this pretty well. But even in Demon's Souls is not that the player wants to fail or explore those outcomes intentionally. It's just kind of unavoidable until you start to get your head around teh game's systems.

@Paul But there are many games that have done away with failure to no harm, right? E.g. Flower or The Sims. I'm certainly not saying all games should be this way. I'd like to see that rather than assuming all games must have a punitive failure unless proven otherwise, it's a characteristic like any other that can be added, subtracted, manipulated, etc. It's totally appropriate often (maybe even most of the time), I'd just like to see us question it a little bit more.

August 11, 2010 at 12:06 AM  
Anonymous Anonymous said...

Train is also an interesting – although debatable – take on this subject. If the player understands the meaning of the game and works against its victory conditions, he fails (according to the rules). Yet, he does not terminate nor lose the game.

August 11, 2010 at 12:44 AM  
Blogger Nels Anderson said...

@Hugo There's quite a lot I don't like about Train, but the notion of having objectives the player doesn't actually want to achieve is an interesting way to come at this issue. Will have to think more on that, thanks.

August 11, 2010 at 6:55 AM  
Blogger Fred said...

Maybe before long, we'll see games where you "fail" because you want to, not because you didn't do things right.

This seems to go against the basis of a game. In a game, rules are set, and players are expected to follow them, which leads them to an ultimate conclusion : Winning.
To me, "failure" in a game seem to be a way designers use to "punish" players, who don't follow the rules, aka not doing things right.

I'll confess I've never played an oldschool roguelike game, unless you count Diablo and Demon's Souls,etc.
So, correct me if I'm wrong, but it seems that you still want to fail in such games, to gain more information. The ultimate goal or incentive is still to win, to do things "right".

If players are going to be encouraged to intentionally "fail" in games, there needs to be some incentives involved,no? I mean,as humans we would always want to "win" in games,right?

You can take away the concept of "winning" and "failure", but there must still be an incentive for players to continue with that game? Flower for me was basically the emotional experience of the game. I just felt comfortable and relaxing whenever I play it.

In other words, I think in order to create such a game that encourages intentional failures, there need to be a suitable incentive. I might be making the idea of gameplay seem too "cold" and dependent on incentives.

Sorry for the long comment, this blog post just got me thinking for a bit. I actually dropped by, just to say great job on DeathSpank! A wonderful game, reminding me of my Diablo days, just funnier and cheerier!

August 11, 2010 at 7:02 AM  
Blogger Nels Anderson said...

@Fred I think there's kind of two ways to explore this. One is to remove any kind of punitive failure (e.g. Flower, The Sims. The other is to make failure a more fundamental, meaningful part of gameplay. Braid was very clever in this regard. Not only could any kind of failure be undone with the press of a single button, but that mechanic formed the foundation of the whole game.

Again, not all games need to be this way either. I'd just be interested in seeing the results of more people exploring this nearly ubiquitous feature.

And thanks for the thoughts about DeathSpank. Glad you're enjoying it!

August 11, 2010 at 5:13 PM  
Blogger JPLC said...

A game where failure isn't necessarily a punishment but even desired would be interesting indeed. I think Heavy Rain gets close in this regard. Even if one of your characters dies, the game continues on with the remaining ones, and even if they all die, the story just ends there. It's not a "game over", but more a "that's the way things turned out", and I think that's an important difference.

So I could see a game where you would want to fail being similar. To adjust Heavy Rain for this purpose: maybe if you really thought, as the player, that Ethan was the Origami Killer, you could voluntarily turn him in at any time to take him off the streets and stop the killings. You may lose the ability to play as him, but the choice could appear to the player to be the right or desirable thing to do in context.

August 12, 2010 at 5:40 AM  
Anonymous Anonymous said...

"Games can be thought of as a conversation between creators and players."

I've written almost exactly that a few times recently. It seems pretty clear to me, and it's one of the ways that games differ from most entertainment mediums.

As for punishments, I've written recently about risk and challenge, and how they are different things. Challenge is in the playing, risk is what you have to lose. I've never liked punishing mechanics (like "Do It Again, Stupid" design as Shamus over at Twenty Sided calls it), but I like a good challenge.

In other words, don't waste my time making me do something I've already mastered to qualify for something I'm still learning. Most games are terrible with this.

I've also wondered before about making death a different *mechanic* rather than a failure that needs to be punished. Planescape Torment did interesting things with death, and I'm still a little baffled as to why we don't see that sort of thing more often.

August 12, 2010 at 10:32 AM  
Blogger Nels Anderson said...

@JPLC It seems Heavy Rain is a good step in this direction. Maybe a little too much emphasis on having a "good ending" versus "bad ending." Hopefully it will inspire other people to investigate structures like that further.

@tishtoshtesh Yeah, that's a very important distinction as well. There's still some open questions about how to present that (e.g. I didn't mind the lack of risk in the 2008 Prince of Persia but others did). But that's on thing we can simply improve upon.

And as always, Planescape: Torment is a very good reference. There's a lot from PS: T I'd want to see more of and that's definitely one of those things.

August 13, 2010 at 6:52 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home