Beginning a talk on endings

[Iron_JG]

Following update 57, which offered much appreciated insight into Obsidian's design process, I wanted to ask a story implementation question. Basically, how do you make the ending not suck?

 

My sense is you plan the whole story from beginning to end before doing any major level/encounter design, suggesting you come up with a fully realized ending or endings, at least on paper. I would imagine this is standard for most games, and yet designers still sometimes wuss out on implementing a good ending. I realize 'good' is a vague term, but I'm sure many people here could name endings that felt contrived and very abrupt compared to the depth of the rest of the game. My understanding is that most gamers do not make it all the way to the end of many games they play, meaning it makes a degree of business sense to cut back on the endgame first, since fewer people will see it and, therefore, judge the game on it. But this is a horrible thing to do if you expect a game to have any staying power or replayability with a customer base. It can be hard to start a game if you know it leads somewhere disappointing. Not to scratch at old wounds, but I can't bring myself to play Mass Effect anymore after the original ending to ME3. Bad endings burn out good fans.

 

So, I'd be curious if Obsidian wants to address how it approaches designing the end of the game. Part of this includes:

 

-- How fully written out is the ending before it goes to production? If the production gurus say the writers' preferred ending is too resource intensive to implement, do other parts of the overall narrative (like side quests) get scaled back to preserve a consistent quality throughout the main story arc? Can this occur midway through the production process, shelving non-vital content for follow-up releases to protect the main content?

 

-- What experience do you have with games whose endings have not worked out, or almost didn't work out? Can you give us examples, and what you learned from them?

 

-- Related to those games, how does fan reception of the ending influence decisions to expand the game in direct add-ons to the main story line and characters (your traditional expansion pack, like TOB) or to veer off in an unrelated, spin-off adventure? (BTW, I hate unrelated spin-off adventures.)

 

 

To be perfectly clear, I'm not conflating "good" endings and "happy" endings. My concern is that the ending, whatever its tone and outcome, is well paced, fully realized, and meaningful. A well-executed tragic ending is vastly more satisfying than a half-assed happy ending. It should also leave some questions to mystery, so it's worth reflecting on and (hopefully) provides opening for future content.

[rjshae]

Following update 57, which offered much appreciated insight into Obsidian's design process, I wanted to ask a story implementation question. Basically, how do you make the ending not suck?

 

Hire good writers.

 

Personally I'd like to leave the game with some sense of closure and a feel of having accomplished a meaningful task, but with a few lingering questions and uncertainties to be dealt with in a future sequel.

[Ffordesoon]

It's entirely possible to employ the best game writers in the world and still have a crap ending, for any number of reasons. See also: KOTOR 2, NWN 2 OC.

 

That's not to say that good writers aren't an important part of the process, but they're still just a part.

[Lephys]

Hire good writers.

There is no morality scale in P:E, so they'll have to hire non-aligned writers to write non-aligned endings.

 

[PrimeJunta]

Yeah, I'm slightly nervous about this part. Endgames tend to suck, endings tend to suck worse, and Obsidian as a studio and many of the P:E devs individually have a history of fantastic games with endgames and endings that range from let-downs to disasters.

 

In fact I'm kinda hard-pressed to think of games with endgames/endings that don't follow the "wade through masses of mooks, then have a boss fight where the boss keeps resurrecting N times or until you figure out that you need to smash the statues/close the portals/break the jars/jump through the flaming hoops with a triple Salkow, then have an explanation" template, which I think is frankly bad. Let's see.

 

Fallout. Yeah, Fallout. Now that was a good ending and endgame.

 

Erm.

 

Hm.

 

Let me think.

 

:thinks:

 

BG? No. BG2? Nope. IWD? Nope. NWN or any of the expansions? Please. MotB? Nope. KOTOR 1 or 2? Nuh-uh. VtM:B? Nope. 

 

Okay, Planescape: Torment, with qualifications. The very very very endgame and ending did not follow the template and was in fact brilliant, but there's way too much wading through mooks to get there. I mean come on, that's just lazy and shows a lack of imagination.

 

Yeah, that's about it, really. Why is it so damn hard to make a good ending and endgame?

[WotanAnubis]

When it comes to endings, I think it's important to have a general sense of where the story is going to end up, but it should not be set in stone.

 

Mass Effect clearly had no idea how it was going to resolve everything, so the writers came up with a literal Deus ex Machina. Obviously, that's bad.

 

On the other hand, an inflexibly set ending forces the writer(s) to fit everything else around it and the writing process is usually a bit too messy for such rigidity.

 

So that's how I would approach it: have a certain ending in mind, but leave enough wriggle room for unexpected bursts of inspiration during the writing process of everything else.

 

I don't know how Obsidian does it, but considering their finished games all have pretty good endings, I feel reasonably confident they're not going to screw it up.

[exodiark]

Obsidian usually make unsatisfying endings, but not necessarily a low quality ones.

 

Low quality one would be like ME3 ending since they employ a total asspull DEM. I've never seen Obsidian endings employing a DEM, well maybe once in NWN2 ending, but I'm not sure it's a DEM.

 

Unsatisfying means that there are lots of questions unanswered, usually meant for sequel or dlc baits (kotor2, nwn2 OC) or maybe they didn't want to violate somebody else's lores so they usually leave it hanging on status quo.

 

*spoilers*

If you can help Kreia by doing suicide in KOTOR2, there will be clear conclusion that the force is dead. But doing this will simply angers the star war fans. The same goes to wall of the faithless, they can't do anyhing about it since it's somebody else's IP.

 

But when they are given the approval to do it their way, the ending are usually better. Alpha Protocol's and New Vegas' endings are quite satisfying )

Edited June 29, 2013 by exodiark

[Nonek]

I thought Dungeon Siege 3 had a very good ending, it wrapped up all of the narrative threads and introduced all of the changes that the last legionnaires had ushered in, while setting up further conflicts and hints of something deeper behind the scenes. I think what made the ending for me was not just how personal it was, but the foreshadowing and hints laid throughout the game that suddenly all came together. It made sense and seemed revelatory, something I always enjoy.

 

Other fine denouements in my opinion:

Soul Reaver 2, a whirlwind of revelations and empowerment, self referential and byzantine in its unashamed complexity.

Alpha Protocol, specifically the magnificent bastard ending, where Mr Thornton quietly takes control using the tools other thought to use against him.

New Vega, reactivity up the yin yang, a clear demonstration that games are at their best when interactivity is key.

Ultima 4, 5 & 6 brilliant and moral approaches to conflict resolution, that are underscored by the inevitable return to the mundane.

Torment, obviously.

Fallout, multiple approaches and a believable antagonist, whose pride and hubris in his urge to do good were his undoing.

The Witcher, a brilliant turnaround, a righteous condemnation of a monster that is suddenly revealed to be the child you failed.

The Witcher 2, not in itself the best of endings, but it allows us to do that rarest of things, simply walk away. To me after so many forced nonsensical boss battles, that was gloriously refreshing. Apathy may be death but sometimes it is also freedom.

[PrimeJunta]

@Nonek, the dénouement was better than average, but in at least The Witchers, the endgame gameplay was same ol', same ol'. I.e., mobs of mooks to wade through in sequence followed by a set-piece multi-stage boss battle where you had to push the right buttons in the right sequence to win. PS:T had the same, except that it was possible to talk your way out of the multi-stage boss battle if your stats were high enough. I'm really. tired. of that template. Is fun endgame gameplay really too much to ask? Is a set-piece multi-stage choreographed boss battle the only climactic ending you can give a game? I should hope not!

[C2B]

I thought Dungeon Siege 3 had a very good ending, it wrapped up all of the narrative threads and introduced all of the changes that the last legionnaires had ushered in, while setting up further conflicts and hints of something deeper behind the scenes. I think what made the ending for me was not just how personal it was, but the foreshadowing and hints laid throughout the game that suddenly all came together. It made sense and seemed revelatory, something I always enjoy.

 

Seconding this. Though, I would have been happy with more possible slides.

 

On that note, setting up a choice plot thread for a future entry similiar how Treasures of the Sun set up. Because that set-up and choice was completly awesome and the reason I'm still angry THERE IS NO DUNGEON SIEGE IV DAMNIT.

 

Why did you have to remind me Nonek, WHYYYYYYYYYYYY

[Nonek]

Sorry C2B.

 

In principle I agree PrimeJunta, that's partly the reason why I walked away from the Letho battle, combat is getting as overused and tiresome as superheroes, vampires and zombies.

[rjshae]

It's entirely possible to employ the best game writers in the world and still have a crap ending, for any number of reasons. See also: KOTOR 2, NWN 2 OC.

 

That's not to say that good writers aren't an important part of the process, but they're still just a part.

 

Yes, good writers can sometimes produce poor endings. Then again, poor writers are much less likely to create good endings.

[JFSOCC]

I think a good ending means that the topics discussed in the story must come to some sort of conclusion. That's the purpose of having an Arc.

An arc has a person or setting begin in a certain position/mindset, then challenge that position and mindset, and at the end they will have either changed and/or know more about why they have their position.

 

Knights of the Old republic II, while not properly finished, managed to do this by discussing the nature of the force, its purpose, the way it is affected by the user, and how much the user is affected by it.

At the end, although everyone could have a different answer, most people will have thought about it and likely have evolved and nuanced their view.

 

This is what Obsidian is very good at, in my experience.

 

I think we don't have to worry about seeing proper character arcs and story arcs. And once that is set up properly, I'm sure the ending will write itself.

[Ffordesoon]

A friend of mine once argued that a good ending is one that answers the central question posed by the story. I tend to agree with that, more or less.

 

Crap example: in a romantic comedy about a shoe salesman named Dan and a secretary at a shoe factory named Misha who meet-cute in some cloying/adorable way, the question at the heart of the story is "Can Dan and Misha find love and happiness in each other's arms?" If they can, and the author answered the question honestly and without betraying the audience's trust, that's the ending. If they can't, and the author answered the question honestly and without betraying the audience's trust, then that's the ending.

 

Not that that necessarily works well for games, but I do think it's an interesting point.

[Jarmo]

Don't know if I saw the original or revised ending for ME3 but I liked it a lot anyway.

 

FO:NV didn't have a good ending, huge battle and then choose from a couple of choices none of which appeal much.

Red Dead Redemption had a good ending and endgame.

 

The Witcher had abysmally horrible grind-through-masses lead to the end, which in itself was pretty ok.

PS:T much like the witcher, though the final ending was better and lead to it maybe even worse.

 

Largely, I'd like to see the big-huge-long-end-battle-thing go away already.

Or at least skip the hour long pre-battle leading up to the main event.

[Nirgal]

My ideal ending would be a huge unexplained explosion in which everybody dies. This would be supplemented with a "the end" written in blood with a question mark appearing a few seconds afterwards. 

 

The sequel would involve zombies.

[FlintlockJazz]

Obviously the best ending involves choosing one of three colours while everything else is exactly the same regardless of the choice you make. Obviously.

[Amberion]

I really didn't want to think about the ME3 ending anymore, but I just felt I should chime in here with my 2c. My feeling about ME3's ending is that 'you go through the entire game, work your butt off to gather the people and the materials needed to build this machine nobody knows how it works, you get to the final encounter.... you LOSE. and the enemy decides to give you an ending where you don't lose quite so much.' That was how I felt when I finished the game. Like I hadn't won.