Multiple Endings

[+ EraYachi 180]

So, let's put a fresh idea on the table. Or at least, an old idea, but fresh for those who are new/just wanna discuss stuff.

 

I've got a game with multiple endings. Best/Good/Normal/Bad/Worst. It might happen I knock out the 'Worst' one, but really, it's just a variation of the Bad ending but has a slightly different outcome.

 

What I wanna discuss is the lengths we can, and will go, to integrate 'player choice' into our games. How many of the player's choices actually affect the ending? And not just choices, but actions and decisions play a part in altering the final outcome. A system like this doesn't even need a script, though there are ones that exist. All you need are a few variables and the willingness to create extra content that covers the different possibilities that occur after each decision. 

 

Use my game for example (without direct story spoilers, and changing some facts/details of the ending, of course):

 

• Best
• - Three main characters all survive, in one form or another. Protagonist escapes. Character C, D, and E, who were involved in subplots, all live and the 'great evil' is defeated.

• Good
• - Protagonist survives, but one main character perishes forever. Protagonist escapes. One or more of Characters C, D, and E die/turn traitor, and the 'great evil' lives to plague others.

• Bad
• - Protagonist is trapped forever. One main character dies. C, D, and E are all trapped forever (were never helped, aren't there to help in the end), and the 'great evil' lives on.

• Worst
• - Literally everything dies, protagonist goes insane. Trapped forever. Everyone is forgotten. Misery everywhere.

 

This is actually a pretty basic form of the multiple ending scenario. You have 1 or 2 variables that increase or decrease whenever an event occurs, and you have a conditional branch deciding on the outcome of future events based on how high/low those variable are.

But it can get so very complicated. Some games have up to 12 or more endings, and then there are separate 'endings' for different subquests and plots that aren't related to the main storyline. We're talking Mass Effect level decision-making and story impact (but don't get me started on the ending of ME3...). This could be the very backbone of games that are basically visual novels, and despite it being story-driven, it is in fact a form of gameplay. It's a mechanic.

Also, whether you have one or multiple endings depends on your type of game. Is it important to have choices that affect the game's outcome? It adds an element of replayability, that's absolute. Or is it important to have just one ending, like in the Uncharted series? Both examples are great forms of storytelling.

[Cait 385]

I like multiple endings for horror or survival horror games, because it does give you the ability to save or stop person x's death, plus, adds replay ability, I think. Especially, if you might face something different like in Castlevania game where you had to have a special object to see the true enemy controlling someone you didn't really want to kill.  I've been trying to figure out a way of not only adding multiple endings, but a way of recording which ending you discover.  Sweet Home had a multiple endings and it was based on who survived, which is an interesting way of dealing with that.  I would like more action based reasons, too.

[Gui 61]

Let's say you make a game based upon Romeo & Juliet and you give to the players the possibility to choose that, at the end of the game, the lovers do not commit suicide. What will you get? The basic, boring soap opera love story and bye-bye to the eternal classic...

 

Worse: what if you give to players the choice of making the youth fall in love or not right at the start? There will even not be a story to begin with...

 

There can be only one story, with just one ending. If players want to tell a tale, they write their own.

 

I almost never replay games anyway.

[Cait 385]

Yeah, there are those types of people that don't replay games, but there are those that do.  I think another really good example of multiple endings had to be Chrono Trigger, and that it was when you found the last boss and what you had done before hand the caused the ending to change.  I do like multiple endings that isn't tracked on or forced on a player.

[+ CVincent 234]

 

I do like multiple endings that isn't tracked on or forced on a player.

 

I feel like this is one of the best ways to handle the idea, especially for horror games as you also mentioned. I'm experimenting a bit in my game with these ideas..

 

If you're going to have multiple endings then it's important to make the player feel like it will be worth playing through the game to see. That said, I think it's not always a good idea to make the player want to see all of the endings. A part of the experience of the game is playing through it and getting the result you end up with, the player should feel like they deserved that ending.

 

One example of a game that I think had done this fairly well was Witch's House. The endings were all pretty solid, and I'd be happy getting any one of them but I probably wouldn't do a second playthrough for a while.

 

 

 

This is actually a pretty basic form of the multiple ending scenario. You have 1 or 2 variables that increase or decrease whenever an event occurs, and you have a conditional branch deciding on the outcome of future events based on how high/low those variable are.

 

I'll have to try applying the use of this method to see how it'd fit in my game, It sounds like this would be the optimal choice considering as how you could have any number of events affect those variables. This gave me an idea that it's possible apply this method to any part of the game and not just the end alone. I'll probably end up doing this for character sidequests which will leave things fairly open ended.

[magic2345 252]

Depending on the game, I might play it over to see the other endings. In the games I've played, there are two ways to do multiple endings that I know of.

 

1. Choice-based

You absolutely cannot change something once it happened. Once the choice is made, you are stuck with it. This forces the player to go through everything again in order to see all the endings.

 

2. Action-based

There are some objectives that the player needs to accomplish to see an ending. You are able to see most, if not all, of the endings, depending on the amount of work you do.

 

If you are going to make different endings, better make the story different somehow, so that it will be like a fresh new game for the player. Don't just kill off a few characters, but then make the story the same.

[Enneagon 20]

It depends. Multiple endings can add replay value, but not necessarily the good kind of replay value. The worst thing you can do in this instance is force the player to go through the game again not because they want to, but because they feel obligated to in order to see all the content, which turns the majority of the game from a fun experience to a chore. This is why I prefer endings that branch horizontally instead of vertically: different outcomes with no clear "good" or "bad" ending. Tiered endings can work if they're based on gameplay, as in "I only got 1700 points so I got the Okay ending, but now that I'm better I'm going to get all 3500 points and get the Ultimate Golden ending!" In that case, the player has some agency in what happens at the end. If what determines the ending is a text box reading "Will you save the villagers or let them starve?" you might as well stick a flag in the game saying "You will have to make this choice a second time." The player isn't making that choice based on what it makes sense for their character to do, they're arbitrarily choosing one knowing that they'll have to make the other choice anyway. There are more comfortable middle grounds, but you get the idea. More neutral choices, such as the protagonist ending up with Boy A instead of Girl B, are better for helping the player feel like they're crafting the experience.

[Yuugami 252]

My two cents on something like this is that, like what some people said above, it depends on the storytelling method and the overall level of immersion the developer wants to build up. Three forms of multiple endings come to mind.

 

1. Alignment (Law/Neutral/Chaos): You will this one pop up in western RPGs, because often times the gameplay is derived off a tabletop ruleset; Knights of the Old Republic was based off the Star Wars tabletop rules, Dragon Age: Origins is more or less a spin-off of Bioware's earlier D&D titles like Neverwinter Nights, which in turn adapted the 3rd edition ruleset, etc. A jRPG example, and an overall simplified version of alignment, is also found in Shin Megami Tensei titles. The storytelling element is fairly obvious: the character really only drive the story, but as a mechanic for gameplay, it's definitely the most mathematical. Shin Megami Tensei 4 could actually be calculated, with specific choices being given a certain value, especially since you can't go back to these choices and "farm up" alignment. You'd have to be in a certain range (like -22 to 22 is the whole range) when you reach the point of the game where your alignment gets locked in. Being aware of this, if you were aware of which choices (required or optional) would give you the absolute max form of points, you can just adapt your choices and figure out how many more "desired alignment" choices you need for your ending. Often times, because of the math behind the alignment mechanic, these games can get super tedious to play through multiple times, since you'd basically have to start from the beginning.

 

2. Optional Ending b/c Sidequest: I will use Dark Souls and Dark Souls 2 for the example. Basically, in both games, you'd have the option to achieve another choice on how to end the game if you exhaust all the dialogue from a certain NPC. In the former, this would be Darkstalker Kaathe and in the latter, this would be Aldia. If you do not choose either one, you more or less end up taking the good ending (becoming the next flame or monarch or w/e), but you can choose for a "dark" ending, where it ties in more with the lore because of the Chosen Undead's relations to the Furtive Pygmy. These endings are much easier to achieve, replayability-wise, because as long as you covered the necessary dialogue, you only need to keep a separate save (note that this isn't actually possible in DS and DS2, you have to clear the game in NG+ for that opportunity) and battle the final boss again.

 

3. Route-based Endings: Most common in visual novels, because VNs are text and picture-based games, the replay value would have to be more than "I liked this novel, I'll read it again", since VNs are marketed as video games and treated as such. Routes are like branches in a tree - there's a common trunk, but eventually it'll split off into different arms (the routes), and these arms will have smaller branches (route-specific endings) as a result. VNs often have a "skip" function that makes replaying them much faster, but this function tends to only apply over text/scenes you already went over in a previous play through - if you got a bad ending in one route, and assuming the bad ending takes place over the bigger half of the ending (like the Bad Ending in Heaven's Feel route of the Fate/Stay Night VN), then you wouldn't be able to skip the text for another ending in that same route during your next play through. It's a very fair replay mechanic though.

 

I can't think of any more methods off the top of my head.

Edited March 23, 2015 by Yuugami

[+ Chaosian 617]

I think the absolute worst way to determine an ending for a game is through binary decision that has no telegraphed outcomes - if that becomes the case it's very easy to have your endings become a text version of I Wanna Be the Guy. The Crooked Man was made absolutely brutal because of these. Choices for endings can come up a lot better than simply using the "Choice" menu system.

 

One infrequently explored avenue of choice I really do enjoy in monitoring the player's choices during the gameplay. I think this was best done in Metro 2033, where the actions the player did would sometimes be deemed "moral", or when they explored and learned about the game world - the screen would very briefly flash. Pretty much half of players that play it still have no idea what this means when it happens. Thing is, spoilers, that the default bad ending of game has the player character committing genocide on an innocent species. If the player has enough Moral Points however, something like half of them in the game, they are given control over their character before they do the action, and can stop themselves - getting the good ending.

the white chamber does something similar and Lone Survivor takes this one step further by making essentially EVERY action you do in the game effect the ending. Eating crackers vs eating cheese and crackers, effects your ending.

 

There's advantages to this, as well as disadvantages. If players do not know they are being 'judged' it's very easy for them to feel cheated. However, it gives a far more honest assessment of the player's performance of the game. A great method if players need to earn a good ending, and not pick one. Because as we know, more players choose to be heroes over villains.

Edited March 24, 2015 by Chaosian

[+ Lyla 41]

I'm not sure if this type of post is considered on topic with this or not ... but I did want to chime in with the fact that I'm using multiple endings in my upcoming game as well. While I'm not 100% set on what choices affect the character, when they do make these choices, it effects a in game gauge reading DP (Destiny Points). By the end of Chapter I, depending on the range of their gauge, it'll split into three different chapters. From there on, depending on the gauge, it can split into 3 more potential endings - giving you a total of 12 endings to end up with ... of course, if I wanted to get more ambitious, I could have a third chapter.

 

I suppose the advantage in this type of system is that you have to earn your ending and it feels like an accomplishment when you get a good/great/perfect ending. However the disadvantage is more backtracking in order to experience all endings ... but also different parts of the same story. At least that's the goal. ^.^

[Gui 61]

Yeah, there are those types of people that don't replay games, but there are those that do.  [...]

To be more precise, I replay a game when its gameplay makes me want to replay it. Or its atmosphere. Or whatever other factor, except the story.

 

If it is just to get another version of the same base setting/universe/characters, I do not see the point, really. Because stories are a sequence of narative elements such as revelation, betrayal, turnaround, coup de théâtre, etc. In theory, you can mix them in every order you want, you'll still get a story; but this is like saying that a story is some sort of techno music track that you can mix and remix and remix again ad nauseam, and also to the point that you don't even recognize the base components any more – hell, I'm sure you could even script some algorithms which could do such things... My point is that some mixes are more personal than some others, this is why some books/movies/comics are worth to be remembered and some others are not.

 

Besides, creating is deciding: what becomes of the writer if he lets the player decides for him? What becomes of his role, his worth? Is he still a writer?

 

And isn't writing a core element of developping an RPG?

 

- - - - - - -

 

What if, instead of having multiple endings, you have multiple paths (but which all lead to the same conclusion)? After all, it has been said many times about stories that the journey matters more than the destination; also, it is a common ordeal for a writer to have to choose through which character a particular scene must be depicted – ie: which point of view...

 

I'll take another example to illustrate this point. Let's say you make a game based on the Lord of the Rings – yes, another one... You begin as Frodon, meet the rest of the party and go onto your journey to the Mordor. While passing through the Moria, you lose Gandalf: that's when the game gives you the choice of changing your point of view; you can now decide to play as the rest of the party or to play as Gandalf fighting the balrog and regrouping with the party later. If you decide to stay with Frodon and Co., you'll get another possibility to choose a different branch later, when he and Sam decide to leave the human, the elf and the dwarf to follow their own way. And so on all along the story...

 

This could be a way to get replayability while preserving the uniqueness of the story and also, last but not least, while preserving the fun factor – because I believe it is the main point of a game. If only the end differ from a series of choice and, as was mentioned by some people here, you don't know until the end of the game if you'll get or not a different ending than the previous game, you may think that you wasted your time, especially if the gameplay is not that great to begin with. But if the experience differs right from the start, you'll know you're going for a different game although it is still the same story – it simply unfolds from a different point of view.

[amerk 1,122]

I only replay a game for multiple endings if content has been added. Replaying the same game a second time, especially right after I just finished, only to traverse the same dungeons, visit the same towns, tackle the same quests, and essentially viewing the same story (outside of trying a varying degree of choices) would bore me too much. And if I wait too long to revisit the game (once I've hyped myself up to give it a second go), chances are I'll wind up playing through the same choices again.

 

Now if new dungeons have opened up, new story segments and quests and things to do have been added, I'll try the game again (mayhap even right away), and even consider altering my choices for a different outcome.

 

I've really only played one RPG Maker game that did this right (to some extent). The game is called "After", a commercial VX title. It was pretty amateurish, but I was lucky to have gotten a free copy at the time for doing some beta testing. The game features 8 possible endings. But really, only 4 of them mattered the most, since the other 4 were almost the same outside of one particular playable character that slightly altered quests or story for the other 4 endings. In the 4 that mattered the most, you decide which side to join about halfway through the game. This really opened up the game since each side had their own viewpoint of the story and completely different objectives and dungeons to visit, making replay worthwhile (in spite of the amateur design).

 

Added content is what drives me to replay a game. Multiple story viewpoints and endings is icing on the cake. But multiple story / ends by itself is usually not enough.

[Cait 385]

I like to think that a really good example of multiple endings is to do to Chrono Trigger style that depends on what you do or do not do, when you fight the boss and then, you have games that each ending you get is recorded to view later, and then, you have visual novel approach that your goal is to get all of the endings to get to the true ending.  But, I think that there is no right way to deal with it as you won't please everyone with all your choices.  I happen to like multiple endings and heck, it doesn't have to be all that complicated either. 

 

Sweet Home's ending depends on who survives, rather simple.

Castlevania - sotn (I believe) you fought Ritcher Belmont, but if you didn't fight the right enemy, you got one ending, but if you fought the right enemy, you continued on. 

Chrono Trigger - had lots of endings, based on when you fought the boss and what you did. 

Corpse Party - it seems that you gather endings to view later.

 

But I do think one thing, that no matter what ending you get, it shouldn't be all that hard to figure out.  The various endings should be something you chose, or fail to do, even failing to save someone's life.  I do not think that every game needs it, but occasionally would be nice.

[amerk 1,122]

I would agree with CT, but only because very few games can match it, at least in terms of time travel. It's really a rare exception of time travel done right, and it makes sense that things you do in the past will drastically change the future. It's also a case where new content does not necessarily need to be added, since it was there in the first place, but do to choices and actions some of that content may or may not be available in the playthrough.

 

All of that said, very few developers I think could aim for such quality. For a stand alone developer looking at the prospect of multiple endings, I'd say either make a New Game + matter, or use a minimal amount of multiple endings, making it possible for the player to go back to a previous save file without having to redo everything again.

[Toykiller 2]

I'm not sure I am approaching this in the right mindset, as I'm a brand spanking new game designer with no previous experience. I am, however, an artist and habitual storyteller.

 

From that, I would say never take multiple endings outs of the equation. I would probably go as far as to say you should make the conscious choice to take multiple endings out, rather then put them in. Either way, it should be decided before you start laying out tiles or laying out event scripting.

 

Composition will make or break a piece of art, pacing can kill a great story, but video games are unique in the fact that they take the audiences input as a core component to their entertainment value.

 

This means choice. They are interacting with your piece all on their own. This is nothing new in entertainment. When Bram Stroker's Dracula was released, it was a cheap knock off of a more popular novel. Once the first Red Scare happened, it became an alagory for communist power. The audience gave it new meaning and it became, retroactively, a classic. One that we will sometimes violently defend when it's classic devices are altered. <inserttwilightjoke_1-1009>

 

Video games however are even more sensitive to this then most kinds of media. We, as developers, have to personally take into account what they will do. We can't have a glowing sword do less damage then a non-glowing axe. The players will have numeric proof of its nonsensical nature to common story tropes.

 

The choices a player makes is part of our composition into their enjoyment. Unlike a movie or a graphic novel, we can alter their experience to their desires.

 

Imagine having a Private Detective game. Early in the story, the player has an option to shoot a bad guy or investigate undetected. If that triggers a switch, we can have more action style 'Fugitive' based game or a deeper slower more tense style 'The Departed' intrigue and mystery. And they can flow back and forth. Maybe they land the same place, maybe they don't. Maybe in one you get the notorious mobster put behind bars, and the other you take over the mob empire. Don't limit yourself to moral spectrum endings. Limit yourself to player desired outcomes and suit the game play to get there based on each. That way, when Timmy plays the game as a kid, he can shoot out and blaze through the criminal underworld and smile at becoming the best, but as an adult, he can examine and learn and uncover the secrets hidden behind the unassuming laundry mat on Brayden St.

 

But, maybe what I'm saying is unrealistic.

[Gui 61]

I believe it is, yes.

 

You know, what you long for already exists since around 40 years now: it is nothing less than tabletop role-playing game, plain and simple...

 

For having played such games during a while, on both sides of the DM's screen, I came to the conclusion it is impossible to tell a tale while having players being able to do everything pleases them – sooner or later, they'll ruin the story in doing everything you want except what you expect from them.

 

Another element worth remembering is that no actual A.I. can compete with a human being, so it is not possible to get a video game which can provide as much improvisation and depth as a person, and it will probably not be possible before a long time. That's why I think what you said is unrealistic, unless you want to dedicate your entire life making a single (almost) perfect game.

 

Concerning the impact of the player on the game, that's a tricky one. Because, yes, it's the main interest of video games compared to other medias; but this doesn't mean that any player can choose interesting paths, at least story-wise (I send you back to the Romeo & Juliet example above), and especially when he doesn't know where these paths will lead him, contrarily to the author: it all depends on the options you give to him, how much you control him basically – in other words how you restrain him, ie: how you diminish his impact.

 

You made an interesting point about the fact that plenty of classic tales were, at the beginning, popular stories transmitted orally; but what made of them classics are their definitive, written form: actually, noone remembers their initial shape, except some experts in old, primitive litterature or historians. And were they that interesting to begin with? For having read some of them, I think it is discussable.

 

Keep in mind that some sort of darwinism may very well be at work here. I mean by this that the less interesting versions of these old tales have probably disappeared because they simply weren't good. The written ones we now know have been selected, for lack of a better word, by well established authors of their time, people who knew what they were doing.

 

Is this the case of any player?

[arekpowalan 2]

A recurring idea of multiple endings in my mind is the one with a paired couple. One of my current projects has multiple love interest whichs, after building up certain relation point with a certain woman, will "fix" the ending, leading the player to different story paths that deal with different people doing different things in a similar fashion how visual novel games play. For example, when the protagonist start dating with Woman A, he will continue working with her in the same guild, but if he dates Woman B, he will quit his guild and becomes a royal guard to protect her, etc,

 

This, of course, would change how at least half of the story is played out, and the player will end up in a different ending depending on the soul mate. Nevetherless, I absolutely hate pointless and bad endings, so there won't be such things as the hero going crazy and ending the world. At the most, it will be different final dungeons and bosses. At the least, different women will help the hero fight the final boss.

Edited March 26, 2015 by arekpowalan