Curmudgeon Gamer
Curmudgeoning all games equally.
01 December 2005
Story vs. Choice in Konami Games (Part 2)
This is part 2 of the piece I started yesterday. The idea is to examine how games balance storytelling with player choice. Today's games are Castlevania: Symphony of the Night and, the most interesting example, Silent Hill.

Castlevania: Old School

Castlevania: Symphony of the Night has a straightforward structure: there are essentially two paths, each of which splits exactly once to provide four endings. See the structure in this diagram. In keeping with the terminology used for Silent Hill's endings, I've called them Good+, Good, Bad+, and Bad.

The endings all have clearly delineated criteria.
  • If you miss an important item, the Holy Glasses, and therefore kill the apparent final enemy, Richter Belmont, then the game ends there with what is considered the worst ending, Bad.
  • If you find the glasses but don't realize (or chooses to ignore) their importance and still kill Richter, then you get a slightly better ending, Bad+.
  • If you find the glasses and use them to see that Richter is being controlled by the evil priest Shaft, then that battle actually marks the midpoint of the game and there is a second castle, as large as the first, to explore after that point. If you subsequently get to the final battle with Dracula with less than 196% of the entire game explored (each castle counts as 100%) then you get a Good ending.
  • If you do reach Dracula with 196% or more then you get the best ending, Good+.
Compared to Metal Gear Solid and Silent Hill (discussed below), this is a rather daring design decision. A player could miss half of the full game and not realize it, ending the story at a point distinctly earlier than two of the other endings. In fact, I do not believe that the Bad or Bad+ game conclusions make explicit that there are other alternatives.

That is, the player determines the outcome of the game, but not necessarily by explicit choices. One of the endings may be reached out of ignorance in Castlevania, whereas Metal Gear Solid informed the player that a choice would be made. Whether the player chooses to believe that Ocelot will kill Meryl is an angle that I've chosen to neglect.

Castlevania also betrays its old school platform game roots by relying on map exploration percentage as a determining factor in the game's ending. A player who completes the game with more than 100% but less than 196% need not know that there is better ending; likewise a player who never finishes with less than 196% need not know that a worse ending exists. This criterion is wholly artificial, with no importance within the game's narrative.

Within the gameplay, the player has a great deal of freedom. Once parts of the castle are accessed, they may be revisited at any subsequent time. The player may dress and arm the character, Alucard, as desired and even purchase items from a store. With the exception of obtaining and wearing the Holy Glasses, however, this gameplay has no effect on the final outcome of the game.

Silent Hill: Real Choices

Silent Hill is by far the most interesting of these Konami games. It has more branches in its choice diagram, seen here, than any of the other games, although it only has four meaningful endings just as Castlevania does. (There is a hidden fifth ending, the UFO ending, which does not fit into this discussion. Silent Hill purists might disagree.) The endings are typically called Good+, Good, Bad+, and Bad.
  • If you obtain a liquid from the motorcycle tank during the game, then you will obtain one of the Good endings.
  • If you do not obtain the liquid from the motorcycle tank, then you will get one of the Bad endings.
  • If you obtain a different liquid from the hospital and you use it on Cybil in your battle with her, then you get a Good+ or Bad+ ending (depending on the motorcycle liquid choice).
  • If you either fail to find the liquid in the hospital or you find it and don't use it on Cybil, then you get the Good or Bad ending (depending on the motorcycle liquid choice).
One important point: the liquid in the hospital is found before the liquid in the motorcycle, so the choice diagram (seen here) is does not reflect the order of events in the game itself. However, this is the simplest structure for the diagram, making obvious the two Good/Bad paths through the game, with the Good+ and Bad+ as refinements of that key decision.

There are several interesting points about this structure. First, two endings (Good and Bad) can be reached by more than one path. Also, the ending scripts don't necessarily indicate that there are other endings possible. Furthermore, the game itself doesn't require the player to find the items required for the Good, Good+ or Bad+ endings, nor does it necessarily make explicit how to use them. If this were Resident Evil, the game would require you to find and use every important item.

This last point is important, I think: What appear to be inconsequential choices (or omissions) throughout the game gravely affect the game's ending.

If Castlevania was daring for letting the player miss half of the game, I think Silent Hill's design is even more unusual. Despite the somewhat contrived device of finding these magic liquids, the designers did create four logical endings of the story. To get an appreciation for the full scope of the game, it must be completed more than once; this is very likely to happen since the unaided player may well miss the details required to finish it with the Good+ ending the first time through.

In one way this design is rewarding: it shows that player choice is important. It is also frustrating, in the way real life is frustrating, as we often don't know how important our small choices will be until a great time has passed.

While experiencing Silent Hill, the gameplay and environment are actually rather limited. Artificial walls and gateways restrict the player's movement and the number and types of environment interactions are very limited. This simplicity belies the game's sophistication.

It is interesting to point out that Silent Hill's sequel, Silent Hill 2, continued to develop the idea of player choice. The endings are not necessarily considered as Good or Bad, but are simply alternatives to each other. The player is left to assign value to the endings. For example, in one ending, simply called "Water", describes the protagonist committing suicide, ostensibly to end the unhappiness that brought him to the town of Silent Hill. Moreover, how an individual game's ending is determined appears to depend in some part on cumulative action and fuzzy criteria. If you look at a knife in your inventory and allow yourself to be injured often, then the "Water" ending is likely, but not guaranteed. This heuristic evaluation of a player's behavior is an interesting twist.

More Thoughts

There are many games that I've not discussed. For example, there are surely role-playing games which have a richer array of choices. One example that's been mentioned before to me is Knights of the Old Republic, but I have not played this game and I have no means to do so at this time. Online games like Worlds of Warcraft don't have overarching narratives, but if we view levels of player stature as endings, then clearly the player has nearly unlimited freedom in achieving those ends.
--Matt Matthews at 11:05
Comment [ 9 ]

Comments on this post:

I'm missing the thesis here. But my observations:

- multiple endings are almost always superficial dressing on a game
- konami game designers need to be kicked in the nads


By Blogger Michael, at 01 December, 2005 16:54  

There isn't any grand thesis. I found it entertaining to see the tension between telling a fixed story and providing multiple endings.

One thing is that games which tout lots of player freedom really mean the freedom to choose how to do inconsequential things. In MGS, how to get through a room of guards. In Castlevania, what armor or weapons to use against which enemies. That kind of thing.

What the game designers reserve for their own use is the power to steer the narrative.

I would agree that multiple endings can be superficial (see my previous rant about Ace Combat 2), but I think the endings in these Konami games are actually fairly essential, especially the endings of Silent Hill.

So there. ;^)

By Blogger jvm, at 01 December, 2005 17:58  

I guess if by consequence you're talking about material that usually has nothing to with the game per se, sure. But I'm eight billion times more interested in the choices that affect mechanics and gameplay rather than pick which poorly voiced, robotic cut-scene I have to watch.

Or even worse in the case of these bonehead designers, whether I play HALF THE FUCKING GAME.


By Blogger Michael, at 01 December, 2005 18:03  

Rather than trying different behaviours to get different endings I so much prefer real decisions that affect the story. The best example so far I found is the game you mentioned: Knights of the Old Republic. You make choices all the time and they affect you in lots of ways, both immediately and long-lastingly. The decisions in fact have IMPACT - in a sense that I really ask myself "How would I react in real life?" instead of "I do this to get good/bad ending".

What I see in the examples you mentioned is a quasi non-linear story. In the older games, if you don't follow the story, you get Game Over screen. Now, if you don't follow the story as author planned, you get "bad" ending. Not really an evolution, just an enhancement. In some cases quite fascinating, though. And not necessarily bad.

By Anonymous Anonymous, at 02 December, 2005 05:29  

Since Michael had a snit on IRC when I didn't reply to his comment...

Yes, the Konami Castlevania people are kind of nuts. The design of Harmony of Dissonance is so bad that I might never finish it, especially since the map is basically useless. And did I tell you that I recently found out that I did not get the good ending to Aria of Sorrow? Turns out that there's a better ending that no one told me about...


By Blogger jvm, at 02 December, 2005 15:02  

This comment has been removed by a blog administrator.

By Blogger rufbo, at 02 December, 2005 20:59  

"Online games like Worlds of Warcraft don't have overarching narratives..."

Not true. What's interesting about WoW (vs., say, UO), is the games' narrative. I make a distinction based on narration to classify RPGs like these -- UO is a true simulation, eg, (especially in its first incarnation; you should read the way the economy was supposed to work) and so is Second Life in some senses. WoW and friends are games with pretty clear ties to traditional fantasy storytelling.

In brief, WoW is built on quests that, though not required, are essentially intergral parts of the game. I talk about it a little here:

The quests serve not only to introduce, textually, no less, an overarching narrative but also serve the practical purpose of leading characters from one part of the game world to the next. For a very long time, there was nothing of the sort in UO -- and for a long time after that, there were only NPCs that asked you to lead them to another town for a small price.

In WoW, you certainly have a great variety of branches, so to speak, for how you choose to experience the world story, but for the most part similar characters will have experienced similar quests. Every so often I stand near a particular NPC in the game and try to create a lame parody when players come up and finish that NPC's quest. The NPC in question has some long lost love come back, but only for a second, and they word bubble some typical star-crossed lover crud about missing each other, etc. I like to jump up on the railing and yell stuff like, "They said the same thing to me in March! Don't believe it! They're using you! It never changes! It's a big practical joke! You're being exploited! YOU'RE BEING OPPRESSED!"

Like I said, lame. But you get the point.

Anyhow, a rambling post to hopefully help correct that oversite.

By Blogger rufbo, at 02 December, 2005 21:20  

For what it's worth, the situation you describe sounds similar to how Everquest Online Adventures, for PS2, was designed, at least when I was playing. Right at your base, where your player spawned when he was created, there were several NPCs who were highlighted from the beginning as people you could get tiny quests from. Once a quest was completed they'd either offer another one or send you to a new NPC somewhere.

As you ventured out, you'd meet real players and they'd tell you about other stuff they'd seen around, so you'd find other quests that way. Or you could just go wilding through the countryside and kill stuff.

From what you've described, WoW is more sophisticated, since the quests I remember from EQOA were like "I need a beetle leg. Find one, kill it, and bring me a leg." That's not much more sophisticated than an auto-generated adventure from the Adventure Construction Set. Perhaps this was a design decision in EQOA to drive you to actually talk to other humans or face death by boredom.

Since I'm not a WoW player or observer, are there larger story arcs than just a loop? That is, going out from an NPC and returning to the NPC when you were done?

The reason I ask is that I was attempting to think of these narratives like directed acyclic graphs. By that, I mean that there is a flow from node to node, with parent nodes and children nodes. The acyclic part means you can't have loops where you come back to a node more than once.

The situation described of finding an NPC, going on a quest, then returning to the same NPC to receive a reward is similar to a loop like this. I will grant the analogy isn't perfect: if you get rewarded, level up, or what have you, then the gamespace may have changed because new quests may be available to you that weren't before. In that sense, it won't be the same node you're returning to, will it?

Of course, it's a half-broken analogy, since if new things are available to you when you complete the quest, then you've not actually returned to the same node in the graph, right?

If I had to draw a distinction between WoW and these Konami games, I'd probably say that the "directed" (from directed acyclic graph) part is what's different. In the majority of games, there is a starting point and a small set of ending points, and you cannot go back the way you came along the way from the starting point to an ending point. Moreover, the larger narrative (like "Stop Metal Gear!") is cover for smaller ones (like "Change the color of the PAL key!").

There are, presumably, many starting points for WoW, such as starting as a different race/class. I was under the impression that, other than reaching a maximum level, there isn't really an ending point for WoW. Or, if you wish, the ending point is that point at which you stop playing for good (if, indeed, you can).

That is, each of these Konami games has a designer sitting back saying "here's a tiny set of stories I want the player to be able to see." In WoW, as you say, you're coauthoring with Blizzard for your own character's story.

If you want to try to put those two approaches under one theory, that's fine by me. But I think they're different enough that I'd be tempted at this point to draw a line between them.

By Blogger jvm, at 02 December, 2005 22:17  

You've confused me. WoW does have an overarching storyline to which many of the quests contribute. With multi-players unfortunately (for now) comes multi-plicitous experiences of the same timeline, which intertwines the storylines in paradoxical ways like the bit I described in the last comment.

I'm not sure what happens at the end, but from what I've heard you're in part right. There are not a predetermined set of aboslute endings. How that distinguishes the WoW narrative outside of (watch out) meaning that we could insert the adjective "deleuzo-guattarian" (let's oversimp and say that means "always in process") I'm not sure.

Is there more to WoW than quests? Sure, you could wander around aimlessly smacking down Muckrakers for years. But this is more Privateer than Elite (minus those very few when you hunt down someone to gain a rank) in the sense that a subset of the quests are directed and do discover a plot/narrative. Blizzard just hasn't written the ending yet.

By Blogger rufbo, at 03 December, 2005 13:15  

Contact Us

Subscribe to
Posts [Atom]



Warm bile sold separately:

Browse Curmudgeon Gamer Memorial Library


Internet game search:

Classic: 02/2002 to 10/2005

This page is powered by Blogger. Isn't yours?