Endings Roundup

a_winner_is_you20110724-22047-1nd3wifHere to cel­e­brate the end of the year, mem­bers from The Ontological Geek got togeth­er and rumi­nat­ed for a while on game end­ings, good, bad and ambigu­ous.  As you might expect, there are plen­ty of spoil­ers ahead.  Here are the prompts we were given:

1. What makes a good end­ing for a game?
2. Examine a good end­ing.
3. Examine a bad end­ing.

Here’s what we came up with:

Hannah DuVoix

What makes a good end­ing for a game?

There used to be a day when games didn’t real­ly end. When story was less of a pri­or­i­ty than it is today, the vic­to­ri­ous play­er was given the words GAME OVER, per­haps with a high score table thrown in for good luck. Nowadays, games tell sto­ries. It seems obvi­ous, then, that a good end­ing must wrap every­thing up. This solu­tion, how­ev­er, is tricky. What does it mean to wrap up an expe­ri­ence that can take some as few as twelve hours and oth­ers up to forty and beyond.

When I was a younger gamer, I desired the oppor­tu­ni­ty to play on with my char­ac­ters (par­tic­u­lar­ly in Morrowind) for­ev­er. Now I don’t so much believe that. I feel it is more impor­tant to real­ize that, much as the play­er will not adven­ture in the PC’s com­pa­ny for­ev­er, nei­ther shall the PC be on hir quest for­ev­er. It is only right, then, that a good end­ing would give us a sense of clo­sure and that our duties are done.

Games can change you. They can make you exam­ine the truth about your­self and what you believe. They can speak to you on a fun­da­men­tal level. But, like all good things, games must end. And when they do, it should not be because the devel­op­ers ran out of things for you to do, but because you have reached the end of your PC’s story. Endings, like every other point in your quest, must be inten­tion­al. They must feel as though a stage in your life has ended. Does this mean that they must sat­is­fy the play­er? Put all ques­tions to rest and make the play­er feel as though every loose end is tied up and that they are now and for­ev­er in com­plete con­trol of their des­tiny? No. But they must con­clude the story, and gen­er­al­ly it’s best that they do it with final­i­ty.

Examine a “good end­ing.”

Fallout 3 and Fallout: New Vegas.

One of New Vegas’ strengths is that it comes to an end. The pre­vi­ous game, Fallout 3, had an end­ing as well, but (no doubt because Bethesda is in/famous for end­less games) this was rec­ti­fied with the Broken Steel DLC and the Lone Wanderer was per­mit­ted to wan­der the Capital Wasteland for anoth­er ten lev­els (sur­pris­ing­ly more time than one would think). This deci­sion wasn’t, strict­ly speak­ing, a bad thing. The Lone Wanderer was a young lad or lass, fresh out of Vault life. I could (and prob­a­bly will some­day) go on at length at the injus­tice of the Lone Wanderer’s orig­i­nal fate (which, SPOILERS, is to be either a mar­tyr or an ass­hole), but Bethesda picked up on this and gave hir a reprieve (Spoilers, I guess), a sec­ond chance at life after becom­ing the Brotherhood’s nap­kin. 3s end­ing suc­ceed­ed because it gave the play­er a sense that, even though ze may walk away from the game, hir char­ac­ter will be okay. There is some hope for the Wastes and the Wanderer beyond Game Over.

New Vegas, how­ev­er, was dif­fer­ent. The Courier’s age is never spec­i­fied, but I always got the sense that I was play­ing some­one sig­nif­i­cant­ly older than nine­teen. The Courier has trav­eled many miles before ever meet­ing the Player. Though we are told what hap­pens to the Courier (and the num­ber of end­ing vari­ables, and thus pos­si­ble out­comes to the game, is deli­cious­ly baf­fling), there comes a point, as with all good Western (A word which here means “Cowboy”) sto­ries, where we must let them ride off into the sun­set. And New Vegas does that with a slideshow, allow­ing the play­er to reflect on hir actions and the choic­es ze made and to see the effects those will have on the Mojave. You real­ly get a sense of what you have accom­plished in your play, and you see just how much one per­son can change the world.

So basi­cal­ly, what I’m say­ing is that Fallout: New Vegas is bet­ter than Mass Effect.

Examine a “bad end­ing.”

Hitman: Blood Money

Hitman: Blood Money is one of my favorite games of all time. It is a sand­box of stealth and mur­der where­in you play as Slender Man and take on increas­ing­ly daunt­ing chal­lenges (by which I of course mean oppor­tu­ni­ties to mur­der peo­ple, because we should not for­get what’s actu­al­ly going on here), each with a dizzy­ing range of options for com­ple­tion. The game is fun because it is unapolo­getic; it lets you plan and get cre­ative with how you want to mur­der peo­ple, and gives many sat­is­fy­ing alter­na­tives to sim­ply run­ning and gun­ning (which almost never works). In fact, as in the Metal Gear Solid series, direct com­bat serves more as a pun­ish­ment than a viable option. It can be done, sure, but the game feels designed to make you regret your fool­hardy ways, and will pun­ish you with rat­ings such as “Thug”, “Deranged Slayer”, and (my per­son­al favorite) “Mad Butcher” if you run and gun.

Then Hitman: Blood Money makes a boo­boo. Specifically, it makes you run and it makes you gun.

Throughout the game, we are shown scenes of a funer­al at an impres­sive, seclud­ed man­sion, though there’s no indi­ca­tion who is the guest of honor. Well, it’s for you. Agent 47 is laid out in a dis­play that would please the Band Perry. He is not actu­al­ly dead, though; the play­er can, with a manip­u­lat­ing of the con­trols I con­fess I failed the first two or three times I got to the scene, revive 47 and pro­ceed to kill all the dudes. But remem­ber how I said the game pun­ish­es you for the hubris of run­ning and gun­ning? It gets even worse here. Because half of all the dudes are secu­ri­ty guards who are armed much bet­ter than you are. I think in my gam­ing career only Ninja Gaiden for the XBOX the First made me retry one gor­ram part more than this game.

The end­ing sequence might have been designed to be cathar­tic; after slink­ing and sneak­ing like you’re ashamed to be tar­get­ing and killing unsus­pect­ing peo­ple at their children’s birth­day par­ties or some­thing, per­haps the devel­op­ers thought they were doing us a favor by tak­ing us off the leash and let­ting us run free. The flaw, in addi­tion to the prob­lems with the game itself, is that the deck is stacked against you. Agent 47 is in his idiom in the shad­ows, sneak­ing around and tak­ing the mea­sured, cal­cu­lat­ed approach to force. This sequence has you wak­ing up in broad day­light in front of many peo­ple you’ve pissed off through­out the game with narry a moment to get your bear­ings before the beat­ings begin. Needless to say, it was nei­ther fun, nor cathar­tic, nor sat­is­fy­ing.

Jim Ralph

What makes a “good end­ing” for a game?

Man am I gut­ted I failed to get involved in the extreme­ly suc­cess­ful Boss Roundup young Bill orga­nized. That would’ve been much eas­i­er! This ques­tion has me wrack­ing my poor lit­tle brain and every time I think I’m on to some­thing I think of an exam­ple that goes clear against it. I guess it’s obvi­ous that there’s no clear cut for­mu­la to cre­at­ing a good end­ing for a game, but a decent bet is to main­tain some con­sis­ten­cy with the spir­it of the game prop­er. If you can man­age that, which is hard­er than it sounds, then you’re most of the way there. We all know a great end­ing when we see one, and it fol­lows that usu­al­ly it’s the cul­mi­na­tion of a great game that brings every­thing to a head in the last few moments with­out spoil­ing it. It’s a bit like sex. Or fly­ing. Or, indeed, both. While it’s hap­pen­ing it’s the best thing ever but so help me god if you don’t nail the land­ing (I’ve main­ly gone with the fly­ing sim­i­le) it can ruin the whole damn thing. When it goes right and every­thing becomes still at the end you’re left wide-eyed and breath­less, you spend a moment bathing in what just hap­pened and you feel all at once real­ly quite alive.

What were we talk­ing about again?

Examine a “good end­ing.”

Say, did you ever play Prince of Persia: Sands of Time? Did you like it? ‘Course you bloody did, every­one who played that game did. Partly that’s because it’s a great lit­tle mix of plat­form­ing puz­zlery and excel­lent com­bat mechan­ics, but main­ly it’s because it has a nar­ra­tive struc­ture that’s at once sur­pris­ing, delight­ful and kick ass. The game’s coup de grace moment of weav­ing its own time trav­el­ing theme to turn the whole plot struc­ture on its head is so damned sat­is­fy­ing that you can­not help but smile. There aren’t many games where pulling a stunt like this would work, but the whole inci­dent is inter­nal­ly con­sis­tent and per­fect­ly in fit­ting with the tone of the game as a whole, not to men­tion the Prince him­self. The risk here was that the final moments would come off as over­ly con­ceit­ed and crammed in to man­u­fac­ture a cheesy happy end­ing. Fortunately the game pos­i­tive­ly dances across that tightrope and we’re treat­ed to some­thing smart, charm­ing and actu­al­ly rather funny. Love it.

Examine a “bad end­ing.”

Let me be clear: I love the Metal Gear Solid series. I can’t pick my favourite book, actor or even my favourite yogurt, but ask me my favourite game. Go ahead. It’s Metal Gear Solid. Now, hav­ing said that… Fucking hell, Sons of Liberty. Just… fuck­ing hell. The sec­ond game itself is pret­ty good, though prob­a­bly the weak­est of the series, which means that when that end­ing comes it crash­es just as hard as Arsenal Gear’s meet­ing with the Manhattan shore­line. There are ele­ments I like which, again, are the­mat­i­cal­ly con­sis­tent with the build-up. The idea of the game’s events being a sim­u­la­tion to mimic the Shadow Moses inci­dent works beau­ti­ful­ly into the series’ ten­den­cy towards meta self-reference. And then… I think maybe that turned out to be a lie as well? Or maybe that was right and the other thing was a lie? I just… I…

I mean, look at me, Sons of Liberty. Just look! Do you see how many ques­tion marks and ellipses are in that para­graph? No ratio­nal human being should be brought to such hea­then depths. Whether the end­ing makes any sense is some­thing we can bounce back and forth all day long, but real­ly that miss­es the point. The fact that we could even have that con­ver­sa­tion means it is a nar­ra­tive fail­ure. The real issue here is that when the play­er is bom­bard­ed with a series of plot expo­si­tions, deux ex machi­na, full on char­ac­ter u‑turns and insane AI gob­bledy­gook in the final moments of a game it becomes a night­mare for that play­er to keep up. I’m sure I had it once, I got my head around it all and it seemed clear and bright as a brand new day. Some years have passed and now all I remem­ber is a mess that puts me off ever want­i­ng to start the game again for fear of fin­ish­ing it. I’m all for hid­den mean­ings and ambigu­ous sig­ni­fiers, to which the appli­ca­tion of thought and analy­sis are as morn­ing sun­shine open­ing the petals of a rose. Sadly, Sons of Liberty’s end­ing shuts the rose in box in a safe in a cas­tle guard­ed by bas­tards.

Oscar Strik

What makes a good end­ing?

A good end­ing is some­thing that gives a sense of nar­ra­tive clo­sure to a game, and it should be in pro­por­tion to the inten­si­ty of the nar­ra­tive in the rest of the game. It does­n’t have to be much; if the game is abstract, the end­ing can be too. Super Hexagon is a love­ly exam­ple of this. Not all games need an explic­it end­ing, either. Just pro­gress­ing to a new high score in an end­less game (think arcade clas­sics) can mark enough clo­sure for a play­er.

For games that tell a more elab­o­rate story, a ful­fill­ing end­ing is impor­tant. If we’ve become invest­ed emo­tion­al­ly in game char­ac­ters, we want to know what hap­pens to them at the end. If there is a major antag­o­nist, we want to see it defeat­ed, or alter­na­tive­ly, we want to lose and be defeat­ed our­selves in a mean­ing­ful way. If there is a loved one to res­cue or pro­tect, we want to see them safe at the end of the game, etc.

For games that have a nar­ra­tive, but which rely most­ly on the expo­si­tion of a com­pelling game world—such as many RPGs—the end­ing may again be less impor­tant. It can be very dif­fi­cult to cre­ate an end­ing that lives up to the many adven­tures play­ers have expe­ri­enced dur­ing the game.

Examine a good end­ing

A game that does this very well nonethe­less is Planescape: Torment. While the game has a very com­pelling set­ting, it is a story and character-driven RPG at its heart. Its end­ing is bril­liant in com­bin­ing mul­ti­ple cli­max­es: find­ing and unlock­ing the cru­cial pieces of the per­son­al his­to­ry of the pro­tag­o­nist, a show­down with the main antag­o­nist of the game, and option­al­ly a final bat­tle where all char­ac­ters are at the pin­na­cle of their pow­ers.

Two other end­ings that impressed me great­ly are the last two games by The Chinese Room: Dear Esther and Amnesia: a Machine for Pigs. Without want­i­ng to spoil any­thing, both end­ings com­bine an emo­tion­al cli­max with beau­ti­ful music and a spa­tial or archi­tec­tur­al design that is high­ly sym­bol­ic and per­fect­ly suit­ed to their posi­tion in the game and the story. Although part of me sees through the struc­ture of these end­ings while they are hap­pen­ing, for some rea­son that does­n’t take away much from the beau­ty of them. They just feel right.

Examine a bad end­ing

One of the more dis­ap­point­ing end­ings is that of The Curse of Monkey Island. I love that game, but after hours of solv­ing puz­zles and meet­ing crazy peo­ple, all to res­cue your cursed pirate love Elaine, after the final puz­zle the game ends with the briefest of movies and bang! it’s over. This can’t be any­thing but end-of-budget con­straints, but it’s just so sad.

A mild­ly dis­ap­point­ing end­ing was that of Deus Ex: Human Revolution, which does­n’t real­ly do jus­tice to the con­cept of mul­ti­ple end­ings. You lit­er­al­ly have to select between a few dif­fer­ent but­tons on machines to choose the end­ing you want. Why not just press 1, 2, 3 on your key­board and be done with it? The end­ing is par­tic­u­lar­ly dis­ap­point­ing com­pared to both the orig­i­nal Deus Ex and Deus Ex: Invisible War, both of which are per­fect exam­ples of how to make choic­es between dif­fer­ent end­ings mean­ing­ful by mak­ing the play­er work for them.

Aaron Gotzon

What makes a “good end­ing” for a game?

The ques­tion of a “good end­ing” is a tricky one, now that games’ sto­ries are being judged increas­ing­ly through lit­er­ary lens­es.  These days it’s a bit like ask­ing what makes an end­ing to a film or novel “good.” Obviously it’s wise to remain inter­nal­ly con­sis­tent – it’s not real­ly con­sid­ered Best Practice to give a mod­ern roman­tic com­e­dy flick a Hamlet-esque “kill ‘em all” end­ing, nor do audi­ences expect a sci-fi-inspired thriller like Minority Report or Inception to have a clear-cut happy send-off (e. g., hero gets girl, per­pet­u­ates essen­tial­ist gen­der stereo­types, and exe­unt all while repeat­ing a line from the begin­ning of the movie but mean­ing­ful­ly this time).  We expect our end­ings to fol­low the Code, and to an extent select our lit­er­ary and filmic excur­sions accord­ing­ly.

To an extent, this makes the ques­tion of a “good” end­ing a mat­ter of pref­er­ence.  In my the­atri­cal life, I’ve often heard from both prac­ti­tion­ers and audi­ence that the­atre is meant for enter­tain­ment, for good cheer and laugh­ter, and of course for happy end­ings.  I would argue that the­atre can be, and should be, much more than that by def­i­n­i­tion, but who am I to argue with the endur­ing influ­ence and impor­tance of the satyr play?  My love of Pinter shouldn’t define an ending’s good­ness with any objec­tiv­i­ty.

In videogames, the end­ing is also coded.  We know, gen­er­al­ly, how a Final Fantasy end­ing for­mu­la works (party gains power and kills God, some weird stuff hap­pens, some peo­ple hook up, and the world is saved through some kind of sac­ri­fice), or a BioShock (iden­ti­ty twist, last ves­ti­gial power struc­ture crum­bles, ambigu­ous­ly good poten­tial future for the heroes).  Any vague­ly com­pe­tent nar­ra­tive is of course going to pro­duce the sort of con­so­nance required for the end­ing to match at least the gen­er­al tone of the over­all work, though, so this isn’t the key fac­tor here.

Central to con­ver­sa­tions like these are the things which games do that other media don’t (the ludo-stuff).  Therefore, the inter­sec­tion between mechan­ics (design) and story, how they talk to one anoth­er and coop­er­ate, has to be one of the most impor­tant parts of exam­in­ing why cer­tain end­ings are and aren’t effec­tive.  The ideal sit­u­a­tion is a ludonar­ra­tive syn­chronic­i­ty: both ele­ments mak­ing sense both in and out of the con­text of their inter­re­la­tion­ship.

Examine a “good end­ing.”

Chrono Trigger remains one of the first mod­ern games to employ mul­ti­ple end­ings in a branch­ing sto­ry­line effec­tive­ly.  Its set­ting is car­toon­ish, some­times even down­right silly (your jour­neys to the dis­tant past force you to reck­on with sen­tient dinosaurs and ooga-mooga­ Flintstones-style humans), but its pro­gres­sion is log­i­cal even if its premis­es are too far-fetched for some (“real­is­tic,” mod­ern) gamers.

Not long into the game, you reach your first “hub” and the world becomes increas­ing­ly open and trav­e­lable as the plot thick­ens.  After a cer­tain point, you wind up play­ing in some­thing very sim­i­lar to one of Bethesda’s sand­box­es.  As you employ var­i­ous means of jump­ing back and forth through time to pre­vent a fore­seen apoc­a­lypse from hap­pen­ing, you can see your actions in the past have direct con­se­quences in the future.  Many of these changes are total­ly option­al, and range from high­ly impor­tant and plot-driven to rather incon­se­quen­tial (change the moral fiber of a mod­ern char­ac­ter by influ­enc­ing his or her ancestor’s life in a major way, or, yanno, cause an eter­nal, evil city to hover in the sky so that future gen­er­a­tions treat it as com­mon­place, natch).

Nearly all of these choic­es will have vis­i­ble effects in the end­ing you get, as well.  There’s about fif­teen of them or so, and all of them con­tain vari­ables which you’ll ham­mer into place through your cos­mic med­dling.  You can put the cart way ahead of any beast of bur­den and dash to defeat the main boss of the game at about the halfway mark, if you know where to go and man­age to become pow­er­ful enough by that time.  You can side against the humans in the afore­men­tioned trav­es­ty of a Pleistocene and trig­ger the evo­lu­tion of dinosaurs into the supreme rulers of the plan­et, hav­ing along ago elim­i­nat­ed the mam­malian threat.

The game’s all about the but­ter­fly effect: a small tweak here, a three-sentence con­ver­sa­tion there, and all of a sud­den, his­to­ry is changed for­ev­er.  This match­es the ending(s), all of which are very thor­ough about show­ing you (and hav­ing you play through some of) a Where Are They Now syn­op­sis of the sub­plots and char­ac­ters you encoun­tered, both major and minor.  The binding-together of the story and its themes at the end, the tying togeth­er of its threads (made of, it may be said, cos­mic strings) synchs up nice­ly with the game’s mechan­i­cal con­ceits: lin­ear, back-and-forth time trav­el, and snap deci­sions with long-lasting and some­times eth­i­cal­ly heavy con­se­quences.

Examine a “bad end­ing.”

At the risk of being like, so 2011, let’s talk for a moment about Mass Effect 3.  If you’ve been online in the past two years you doubt­less remem­ber the extrav­a­gan­za of Internet bit­ter­ness caused by the esteemed trilogy’s swan song.  The hos­tile argu­ments brought about by the end­ing cen­ter most­ly on the final fif­teen min­utes or so, which posits an expla­na­tion of the galac­tic threat, the Reapers, which real­ly doesn’t have the slight­est allu­sion to its cred­it in the pre­vi­ous three games.  Okay, fine.  Still fur­ther, at the end our hero Commander Shepard choos­es between three dif­fer­ent out­comes which, though rad­i­cal­ly dif­fer­ent in the­o­ry, are almost iden­ti­cal so far as what the play­er is shown and what game­play steps remain in order to put them into motion.

On the face of things, none of this real­ly seems like a prob­lem.  After all, how many sto­ries (those you find in games includ­ed) have but one end­ing?  Pretty much all of them.  The issue is that Mass Effect, like Chrono Trigger but even more so, makes a huge deal about play­er choice and con­se­quence.  There’s follow-through, log­i­cal pro­gres­sion, a keen atten­tion to detail evi­dent in near­ly every inter­ac­tion Shepard has with the char­ac­ters and places he or she encoun­ters over the course of three games.  With so much lovingly-crafted vari­abil­i­ty appar­ent in the course of over 200 hours of game con­tent, it was under­stand­able that fans who felt promised the same amount of vari­a­tion in the end­ings wound up dis­ap­point­ed and con­fused.

Bioware’s moti­va­tions and process­es dur­ing the cre­ation of Mass Effect’s final hours have been the source of much dis­pute and spec­u­la­tion.  Many think that the writ­ing team felt over­whelmed by the details, didn’t think they could man­age to pull every­thing togeth­er effec­tive­ly, and so opted for a dia­bo­lus ex machine to patch things up for the whole galaxy, Shepard as a go-between.  This would have made the end­ings some­thing like LOST-style reduc­tion­is­tic homages to fatal­ism: what hap­pens is gonna hap­pen, there’s no chang­ing it ulti­mate­ly.  This is a fine theme to include in a story, of course, but it’s clear­ly not Mass Effect’s theme as it express­es itself dur­ing most of its run.

Other peo­ple believe (Indoctrination Theory, and yes I’m sort of in this camp) that the end­ing sequence is actu­al­ly fierce­ly inten­tion­al, meant to illus­trate the cul­mi­na­tion of Shepard’s long and des­per­ate strug­gle to retain con­trol of his or her mind under threat of indoc­tri­na­tion by the Reapers.  Again, all well and good, and there are some visu­al hints and bits of fore­shad­ow­ing dropped along the way which seem to lend cre­dence to this posi­tion, but that’s just it.  The evi­dence is visu­al only.  In the con­ver­sa­tion­al game­play, and all the modes of play save the last few min­utes, Shepard retains as much capa­bil­i­ty as before.  Just as for­mi­da­ble, just as much of a change agent.

We play games to immerse, to feel our way through a story instead of just watch­ing it.  To be asked to put the play­ing on hold, to ignore it, is the essence of what’s been late­ly dubbed “ludonar­ra­tive dis­so­nance.”  Its psy­cho­log­i­cal effect on the play­er is sim­i­lar to the con­cern and sur­prise one might express if one felt a solid-looking wall and found it squishy.

Bringing these two ele­ments into a pro­duc­tive accord will be the next step in the evo­lu­tion of gam­ing as an impor­tant cul­tur­al and artis­tic force.  Perhaps some­times we take steps back, but, as much of the indiesphere (at least) has taught us, we see the poten­tial and we keep mov­ing for­ward.

Bill Coberly

What makes a good end­ing?

Good” in this con­text seems to mean both “sat­is­fy­ing”  and “coher­ent with the rest of the work.”  A game with a defined nar­ra­tive thus has a lot of plates to spin with its end­ing: it has to both tie up the story (char­ac­ter arcs, over­all plots, impor­tant ques­tions of set­ting or Big Questions the design­ers are try­ing to ask) and pro­vide a cli­mac­tic and appro­pri­ate­ly dif­fi­cult game­play sequence (boss fights and Important Choices are pop­u­lar).  One recur­ring prob­lem occurs in games which do not lend them­selves to Big Climactic Moments: stealth games are all about avoid­ing Big Climactic Moments, so how do you end one sat­is­fy­ing­ly with­out just mak­ing it impos­si­bly dif­fi­cult?  This is prob­a­bly why many games seem to sud­den­ly switch gen­res in their end­ings — stealth games become shoot­ers, shoot­ers become rac­ing games, etc.

Examine a “good end­ing.”

The snarky response is “there aren’t any,” because videogames seem to have such a hard time end­ing — bud­get con­straints, time con­straints, and the fact that some obscene per­cent­age of play­ers won’t even reach the end­ing often puts “end­ing well” towards the bot­tom of a game’s pri­or­i­ty list.

But I’ll play: (seri­ous spoil­ers): Gone Home ends per­fect­ly.  The game’s use of hor­ror tropes through­out (dark, spooky house once owned by mys­te­ri­ous and vague­ly sin­is­ter bach­e­lor uncle, flick­er­ing lights, miss­ing fam­i­ly) has been busy set­ting you on edge all game, and the vague por­tents that Something Terrible awaits in the pre­vi­ous­ly locked attic make the final ascent up the creak­ing lad­der a ner­vous one– even though by now you’re pret­ty sure that no ghosts are ser­i­al killers are hid­ing up there, you have good rea­son to think that your lit­tle sis­ter has killed her­self and will be dan­gling from one of the attic’s rafters.

When you get there, no such ter­ri­fy­ing sight awaits — Gone Home is not a hor­ror game, it just wants you to think it is some­times.  Instead, you find your sis­ter’s jour­nal, and real­ize that she’s all right after all, and learn where she’s gone.

Gone Home is about explo­ration, mount­ing unease, read­ing other peo­ple’s mail, and a cer­tain refresh­ing, coura­geous sweet­ness, and the end­ing under­scores all of these themes — you sneak into the attic (your sis­ter’s dark­room, where you are NOT SUPPOSED TO BE), lit all in red, using a key you more or less stole from her, and dis­cov­er that maybe every­thing’s going to be okay after all.

Examine a “bad end­ing.”

Amnesia: The Dark Descent is one of my favorite games I absolute­ly hated play­ing, but its end­ing is dis­ap­point­ing as heck.  After work­ing your way through the bow­els of Brennenburg Castle, chas­ing after a nigh-immortal mon­ster of a man, you have to hide from and avoid all man­ner of ter­ri­ble crea­tures.  Utterly help­less, your best defens­es against Alexander’s min­ions are clos­et doors and run­ning, clum­si­ly, in the oppo­site direc­tion.

Amnesia is about hid­ing, help­less­ness, and insan­i­ty.  It’s a stealth game with no instant-kills, where hid­ing in the dark­ness caus­es your char­ac­ter to slow­ly lose his mind.  So, nat­u­ral­ly, the end-game con­sists in push­ing over pil­lars with no dan­ger but a pre­pos­ter­ous­ly long time limit, in a bright­ly lit room.  It’s incon­gru­ous and bizarre, whol­ly unsat­is­fy­ing and entire­ly divorced from the rest of an otherwise-excellent game.  (While the “bad end­ing” you get from just stand­ing around and look­ing con­fused, allow­ing the vil­lain’s Dark Plan to reach its loath­some fruition, is actu­al­ly kind of sat­is­fy­ing, it’s clear­ly just there to sat­is­fy play­ers who want to “see what hap­pens if”).