The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
Please vote in the Forum Structure Poll. Polling will close at 2PM EST on January 21, 2025.
[PATV] Wednesday, September 18, 2013 - Extra Credits Season 7, Ep. 2: The Illusion of Choice
SR4 has several meaningless choices made in the second segment, but they're fun to make. They impact nothing in the game, do not come back, ever, but somehow they help flavor the game (for me, at least). I usually make the same choices each playthru, though. End Hunger, Take the Low Road, Party On.
SR4 has a lot of other meaningless choices in it, the first of which is character customization. Nothing in customization means anything in the game. Not the voice, not the build, not the clothing you wear. But it invests you in the character, so it isn't completely meaningless. Early on, I bought the Presidential Suit again and I spent the entire game playing in that. It was kinda fun kicking butt in a nice purple suit. Added the Morpheus glasses later, and I felt like a total badass. It was awesome. Meaningless to gameplay, but awesome to the experience.
I'm shocked Walking Dead wasn't used an example here. That game lived on the illusion of choice and didn't even attempt to hide it. You felt you had agency, but were victim to the inevitability of circumstances. It was brilliantly done and one of my favorite role playing games (I realize it's not an "RPG" by genre definition, but it's most certainly a ROLE PLAYING game).
Unfortunately, players can come to recognize illusory choices with enough experience or awareness, and after that these illusions become glaring. Hopefully we'll get some more clever content designs that incorporate procedural generation and what I suppose might be called procedural dialogue. This may present a localization nightmare, of course.
@Incenjucar: Noticing the illusion won't necessarily stop you from appreciating it, if it's skillfuly implemented. @Drakkon is right, embracing the illusion of choice is fun. Valve games IMO are 100% corridors (thinking of HL2 and Portal 2), but the design and art direction is so great you never feel it on first playthrough, and then when replaying it you get all those little "I see what you did there" moments, appreciating how finely crafted those corridors are.
BTW, @Diamondback11, Extra Credits did a couple of episodes dedicated to the Walking dead a while back.
You should play front mission 3 for one of the weirdest case of real choice, which end up with such large change in the story that it might actually not be a choice at all. At the start of the game you're asked a really simple question that really shouldn't matters, but depending on which answer you pick, the entire game, from start to finish, follow a different narrative, and you're never told of how big the choice you just made was. On one side this kinda waste resource, on the other it was freaking awesome when you'd finish the 40 hour game to find out that there was another 40 hour game on the disk. Pretty big difference when you compare it to today game.
Of course, the moment the player realizes that you are hoodwinking them on the dialog, you stand to lose most of their investment in the story. When I catch a game pulling the stunt where the answers are the same regardless of my choice, I kind of stop reading or deliberating my choices. If they answer is the same either way, all the information is going to be provided by the other person, the hero might as well be mute.
@MasodikTiasma Preventing the player from replaying a conversation and realizing that the choice was not only completely illusionary, but also completely ignored by the game, is hard to imagine.
On the other hand, wasn't it the unraveling of the illusion of choice that lead to the Mass Effect 3 ending debacle? One of - if not the only - major complaints people had with the ending for ME3 was how, in the end, the final decision was between which color the light show at the end was. After so many illusions of agency in which so many people felt their was agency, the sudden destruction of the illusion of choice couldn't be withstood. When, when Extended Cut came out and a different ending cut scene was added depending on the choice, it restored some of the illusion (as well as giving closure to the character arcs), correcting a lot of the anger.
Dreamfall: The Longest Journey did this very well. You feel like some of those decisions are important even though you know they won't effect the outcome much because the decisions provide the context for the story. It lets you build an identity for the character.
Deus Ex is another game that made skillful use of illusionary choice in the narrative. Any choices you make are quickly incorporated back into the main narrative, but there are enough callbacks and reactions to make them feel significant.
Mechanically, on the other hand, it had plenty of real choices, as picking different means of reaching a goal allowed you to completely bypass obstacles on other paths and play to whatever strengths you had chosen through skill assignment and equipment choice.
I fear this episode is made for the sole purpose of putting Bioware under a good light since Dragon Age III is nearing its release date. Did they pay you like creative assemply did to make the episodes about Rome?
The truth is that branching paths which go back to the same storyline after a while simply suck. They are not clever at all. The player notices them, even more if he plays through the game twice or more. I can justify them because of the developing costs but you seem to consider them the coolest thing in the world. They are not.
And please, don't tell me that real branching paths are impossible to make, you said who managed to do almost everything right: The Witcher 2.
The thing is, you probably even know that now you have to choose to do something. But the game doesn't just hit you with the list of options, it makes you to figure it out yourself what to do. This is how it should be done more often. It's not like choosing your meal from menu in Burger King, it's like making our food yourself. This hit me really hard. just... Awesome.
Of course, the whole game is just pure awesomeness.
The 'breadcrumbing' of coins, lit areas, different textures etc to draw people through a space has a couple of key drawbacks - you very quickly lose any sense of where you were going, or what your purpose was. Also, you lose the ability to spatially reason your way through a complex space, because you just get led through it with horse blinders on. Thief was the antithesis of this - you generally had to work out where you were, and where you had to go, purely through environmental information, and the breadcrumbs only really worked if you were looking and had a plan.
Illusion of choice is a useful tool, but man does it suck when it breaks. I noticed this in ME2 in one mission where there's a potential branch point if some nameless character dies. If they don't die, they decide to pretend to be dead. It was just so lazy and obvious and broke me out of the action.
Also, sometimes you really can give the player nearly limitless choice. Just look at Minecraft. You have to be willing to sacrifice narrative and structure to some degree, but procedural content can go a long way to making real choice on a huge scale.
I think this is one of those key areas where you can tell many designers fall in love with their excuses. "Well, we have to do this, so it is totally a good thing!"
Yeah, just keep telling yourself that.
Here's reality: Illusion of choice is bad because it deprives the players of agency. That's reality.
"But TD!" the developers cry, "Illusion of choice is a wonderful thing! Look at Mass Effect!"
Yes, let us look at Mass Effect for a moment. It is a wonderful example of why this is bad.
What happened in Mass Effect? People really ragged on the third game. They ragged on it a lot. But why?
The answer is the illusion of choice. The reason Mass Effect was so engaging was that you had to make big decisions - decisions with potentially unknown moral impacts down the road - and when you made those decisions, the game promised you that they mattered. Did you release the queen, or kill her? Did you blow up the Collector base, or leave it intact for Cerberus? Did you save the Council, or did you let them die? Better to commit genocide on the heretic geth, or reprogram them? Did you destroy the research on the genophage cure, or preserve it?
These were all big, weighty decisions, and you could see that there was potential for disaster either way. And the games promised you that they would remember these things, that these things were big decisions, that they were important. The games got away with these decisions not mattering in the game because they pushed it off to the future.
The problem was, there was no future after Mass Effect 3, so all your choices had to matter then. And they didn't. At all. And that basically told the players, "Your choices don't matter now, and they never did." The ending should have had you fighting alongside allies, getting covering fire from the people you chose to save and convinced to ally themselves with you. Dragon Age: Origins had your allies show up and help you out against the final boss, so those big weighty decisions you made felt like they really mattered, and there were some differences between the way that your allies fought - it wasn't a totally different battle or anything, but it made the game different, and your choice actually did matter, even if the actual cost of said implementation was quite low.
Mass Effect 3 utterly failed in this department. The choices you made didn't actually matter, and that sucked. There was no choice, only the illusion, and that killed the whole thing for people. It wasn't just the ending which had the illusion of choice, it was the entire game. Your decisions were supposed to have weight. Your decisions were supposed to matter. They did not.
It is true that you have limited resources. It is true that the illusion of choice works sometimes. But the problem is that if you give players the illusion of choice about something that actually matters, you're breaking a promise to the player, which is that the game isn't lying to them. WSIWYG is one of those basic, vital promises that you make to your players, that you aren't lying to them, deceiving them about things. Invisible walls are an example of a broken promise to players - if the players hit invisible walls, then you're breaking the promise that they can go over where they are looking. The illusion of choice is the same sort of thing.
I can tell you from experience with roleplaying games - real ones, that is to say, tabletop roleplaying games, not CRPGs - that you should avoid the illusion of choice whenever possible. You basically make a compact with the players - they don't intentionally avoid what you've designed, and you give them reign to make choices. If they feel like their choices don't matter, you've betrayed them. The same is true in any medium.
Incidentally, as far as the whole "highlighting" things goes, I have to say that games have trained me to avoid such things because hey, no one hides secrets there. I go down the obvious path last. I think if you really want to encourage such behavior, you need to be smart about how you do it, and never punish them for following the obvious path. Them missing secrets is punishment.
I think the problem is that game developers have, to various extents, fallen in love with the idea of giving players unlimited choice, and then realized no, we can't do that, so we have to cheat, without going back to question their base assumption in the first place.
I'll be real honest here: Chrono Trigger is one of the best RPGs of all time, and the game gives very little in the way of illusion of choice. You are going from point A to point B all the time, with very little room for wandering around at any given moment, and yet, you don't care. Why? You WANT to keep moving forward, to keep going in the direction the game is telling you to go, and the game doesn't pretend like you have a bunch of choices but you end up in the same place anyway. It isn't about lying to you, it is about being compelling.
A lot of FPS's also have very little in the way of illusion of choice. Many platformers have 1-2 paths through any given area. The list goes on. Many, many games where you have very little in the way of illusion of choice are great games, and people love them. People don't complain that you don't have any choices in Chrono Trigger because the idea that you're supposed to have choices in that game never even occurs to them; the choices you have are how to beat up the monsters that get in your way.
This is true of most SNES RPGs. This is true of many JRPGs. This is true of many games in general. Yes, you can easily make a game terrible in this way, but how does the illusion of choice prevent the game from being awful? FFXIII would still be a bad game even if you had the illusion of choice; it was poor design that doomed it, in the end.
I do understand the trick with dialogue in games like Mass Effect or Dragon Age, and players are okay with most dialogue being mostly meaningless save possibly altering some minor numerical variable one way or the other. But when you make "big decisions", and they don't matter, players get upset. And that is where the illusion of choice breaks down; in moment-to-moment, transient decisions, it upsets players a lot less than in big ones if they don't matter because they don't expect the little decisions to matter that much most of the time.
@Titanium Dragon
The point in Mass Effect that angered almost everyone wasn't aggravating because it only gave the illusion of choice, it was aggravating because the illusion of choice failed and people noticed that everyone got the same ending.
Giving two endings with substantially different content is not that hard. Go try playing NWN2 sometime. There's a branching ending that probably took little to no effort (evidenced by the fact that it didn't get cut when obsidian got pressed for time). The 'evil' ending uses no AI, models, or scripts that hadn't appeared before, but it still felt like the player was making a significant choice. (To betray your party or to do something remotely rational)
I can already see that there's quite a large Mass Effect debate going on below! All I can say is I agree that quite a lot of effort went into making engaging dialogue options and story segments. That, and now I want my own jumbo sized choice monster. That thing is adorable.
I have to disagree with Titanium Dragon's comment about Chrono Trigger and other JRPGs having little illusion of choice.
For one thing, the very first part of the game at the millennial fair and the trial afterwards was a brilliant example of the illusion of choice. Chrono Trigger and other old school JRPGs were good at creating an illusion of choice because they allowed you to independently explore whatever town/dungeon/map you were on.
The choice you made was where you decided to go. The illusion is that nothing really happens until you get to the place the game designers intended you to go. The illusion is further reinforced by providing rewards for exploration, i.e. random treasure chests everywhere. How many times has a game prompted you to go somewhere, yet you purposefully go the opposite direction because you want to make sure you find everything? All of those moments are choices.
FF13 caused so much outrage about "linearity" because it shattered this illusion by forcing you to walk through one-lane corridors and gave you very few choices as far as direction until 3/4ths of the game was done.
@ksmiles:
The problem wasn't just with the ending, it was with the entire game. The first two games, you made choices which seemed to matter, but when the third game rolled around, they did not.
More or less, the problem was that they made promises that they just couldn't keep. It wasn't the ending; if it had just been the ending, yeah, people would have been angry. But the entire game primed you for the ending, because the entire game, your choices did not matter.
Basically, they made the first two games seem a lot grander with promises of future content which not only never materialized, but which they simply were not apparently up to the challenge of producing. The bigger the decision, the bigger the impact, the better it feels to the player... but the bigger the challenge you have, as a designer, of dealing with those choices.
Ultimately the purpose of the illusion of choice is to improve player engagement. The larger the choice you present the player with, the more involved they are - but if you turn around and then make that choice not matter, then the player feels lied to. This is because they were lied to.
The very first "big" decision you make in The Walking Dead is a false choice, and the fact that it was a false choice robbed me of my feeling of agency in the story. I chose "wrongly", the character I tried to save died anyway, and it became obvious to me that the game was trying to cheat me.
If you present a significant choice and don't really mean it, what is the point of presenting the choice in the first place? The cost of people feeling robbed is not worth the benefit.
For minor dialogue options, that's fine because they at least feel like they're saying something, even if it doesn't really matter. But if you have some sort of significant choice, choose opposite the way that the story goes... well, it is just bad. The story very frequently flows in various ways without the player's direct intervention, so why are you involving me only to say "Just kidding! We don't care what you think!" ?
This is the problem. I don't feel that all of these choices actually DO add to games, and I don't feel that players miss a lot of false choices.
If we are going to talk about games like this a good example I think would also be the Elder Scrolls games. Skyrim the most recent one for example. There are a lot of choices that you make in this game, most without any real consequence, even some of the biggest ones dont really have any consequences.
Take for example the fact that you are the Dragonborn. A few characters may acknowledge you but the vast majority will pretty much treat the same way every time even if you do something big for the city or the leader of the city, nothing really changes.
Another good example are the guilds, the Companions, the College, The Thiefs guild, or the Dark Brotherhood. You can pretty much take part in every single one of them, become the leader of every single one all at the same time. No one ever questions you, no one ever calls you out on being the leader of the Companion for example and being a member of the College. There are never any consequences for really any of the things you do, nothing is ever kept with the exception of a few lines of dialogue and maybe an item or two for your decisions.
I disagree with this episode in pretty much its entirety.
With the illusion of choice, I agree with much of what @Titanium Dragon is saying below. If I were to sum it up, I'd say that every choice should have an appropriate payoff based on the scale of the choice. And that appropriate payoff proves that it can't be an illusion. It doesn't matter if the grand story arc reconverges. So long as the story arc does diverge on the choice, then the player is making a choice, a real choice, and so long as events following this divergence fit the scale of the choice, the player is not going to feel cheated.
It's when the game is utterly dismissive of the player's actions by either ignoring them or not giving them appropriate impact that the "illusion" falls through. And that's not the player failing to believe in the decisions being real, but rather the game failing to grant the agency that the player is expecting.
With FPS games, all the mentioned techniques always feel like being led by the nose through the level. They never provide me with an illusion of choice. But that's okay because what's really leading me through the inherently linear gameplay is the narrative, and the urge to see what's next, not some misguided notion that I'm going wherever I like and kicking butt whilst doing it. These techniques only help me align myself with the level and prevent me from becoming lost, and nothing else.
If I'm caring that I'm being directed down a linear corridor at a goal, then the FPS is doing something else fundamentally wrong. Either, the narrative isn't powerful enough to draw me through the gameplay, so I don't want to go down the linear corridor because the other end just doesn't seem interesting enough. Or, they've presented me with a gameplay choice that ignored my input, such as presenting me with enemies laying down razorwire, but they're invulnerable and there's an invisible wall between me and them. I wouldn't have cared if that corridor were not there or empty. But the game hinted at an alternative course of action, but that action turned out to be impossible.
I also think it's entirely possible to give the player all the choice, but not without sacrificing anything resembling a narrative. All you'd need to do is create a Minecraft-like game (infinite procedural world) with characters in it who are themselves procedurely generated with motives and desires, and then the gameplay arises with how the player character interacts with these NPCs and who they try to help and hinder. There would be a specific social order and rules governing how NPCs can react and engage with each other to get what they want, and in such an environment, NPC villians and heroes can arise just through the actions they are willing to carry out (or allowed to carry out) in pursuit of their desires.
The real difficulty here is not the asset cost; you'd have to plan for a lot of conversational paths but you could then reuse the tree for each character with reskins on the conversation to provide some differentiation. Where the agent starts in the tree determines where his choices can move him to, and provide the depth of characterisation, and such a tree I can't imagine being more work than writing the narrative of a traditional storyline, albeit it would be harder to test.
The real difficulty is the level of simulation required to be believable. Getting a city full of agents interacting realistically would require a lot of processing power. We might have beefy enough processors now, but have yet to see someone implement such a game, or we might need more power before such a game is possible.
@dotails
It's not really freedom because the story doesn't progress until you travel to the story missions. And the sidequests do nothing to the story progression.
There needs to be a balance between how much illusion of choice you get and how much actual meaningful choice you get, and that balance needs to be based upon the promises your game makes. Skyrim is a horrible game in that respect, because it promises you a world and a narrative and doesn't deliver. Compare it to a Megaman game and you see nothing even remotely similar is in the second, but it doesn't suck as much, because Megaman promises different things. Megaman for example promises you choice in platforming and shooting, and it's not like failing a jump doesn't kill you or shooting the wrong direction hits anyway. One Prince of Persia tried to do this (removing the game over screen because "it's a waste of time") and that basically sucked all the fun out of the game. Super Meat Boy basically does the exact same thing, but you actually die and leave blood trails around, and that makes all the difference. Is that an illusion? No, it's not. Stuff happens in the fiction, you experience it, it's different: that is enough to qualify it as "real" for the purpose of the game. Is it mathematically equivalent? Perhaps, but that's not enough to justify it.
@TD
What you're complaining about isn't illusionary choice but about choices that don't feel MEANINGFUL. They are completely different things. An illusionary choice should feel just as meaningful as any other choice would, and if it doesn't then it isn't because it's an illusionary choice but because whoever wrote the choice in didn't know how to use illusionary choices. It's a tool - nothing more. If you play video games then you've likely made a million of them and never even noticed because you thought they were real choices.
@discrider I'm going to go out on a limb here, and suggest you know nothing about programming, ai, or game development. You've pretty much said that it's easy to create genuine choice, all you would have to do is create a fully simulated artificial world, and populate it with fully functional artificial personalities, each with procedurally generated goals and needs. Given that the entire computer science community has yet to produce a single ai of this complexity it seems somewhat unlikely that thsi problem is only a little more work than a standard storyline.
I love you guys, Extra Credits, but Mass Effect was perhaps not a good example to use. Titanium Dragon hit most of it already, including how many of us have actually been conditioned to ignore cues. Mass Effect actually grossly penalized players for following cues--if ever you saw a 'Priority Mission', finishing it first would often close out other missions thus depriving you of XP and story details. You truly only stood to gain by waiting on 'Priority Mission: Deliver Medicine to Dying Orphans' until you have completed several 'Deliver Artifacts to Junk Dealer' type missions first. This broke my immersion in the game; the fate of the galaxy was at hand, and the game was only penalizing me for treating it seriously, ignoring sidequests. Had the story turned out better somehow, it wouldn't have been so disruptive, but often the story turned out WORSE by ignoring personal quests that were frankly a luxury to deal with.
Walking Dead is a better example of Illusion of Choice. While I still felt a bit cheated by the I of C - which was quite obvious in that game when characters you specifically saved arbitrarily die - there was also less of a time investment to playing the game... and that arbitrariness fits within the world we were given. The story was still strong when we only had a small amount of control over the world, because in the end, death takes us all. All we can do is choose how we face it, how we live in the meantime.
There is a tendency nowadays to look at how much playtime you get out of a game. I would have actually gotten more playtime from the ME series had ME3 been about 30-60 minutes long, but seriously took my choices into account for the ending. I would have replayed ME1 and 2 over and over again until I found an ending I was satisfied with.
In short, Illusion of Choice has its uses, I'm sure - but not in games where they tout 'Meaningful Choices'. So many choices in ME1 should have led to wildly divergent story paths, and they should have honored that bargain with the players. Even Dragon Age: Origins, one of their earlier works, did it better, and 'Choice' was not a selling point of the game.
@Vinnie555
The Sims where the player is a Sim and where the land is procedurally generated and not just restricted to a lot would be pretty darn close.
You wouldn't need as deep a biological sim, and the random global wants would turn into quest hooks that the player can choose to satisfy and which the sims would work try to satisfy themselves. Jobs could be predefined on creation or on age level up. And the relationship network is there to help determine what actions a sim is likely to take to advance their global needs.
Like I said, the main problem would be getting hundreds of sims automated at the same time. But until we can do that, you could always cheat by making lower classes have only basic pathfinding AI as since they don't affect any larger systems, they don't need as detailed a automation, and then fill the cities with such shells. Potentially if the player befriended/fought with such a citizen then they would receive a full AI treatment as the player has expressed interest in following the character.
Also moving away from a population city would slowly wind back its automation. Citizens would stop being automated at a certain distance because you can't really see what they're doing through walls. Then special character need satisfaction requests would stop outside the city's influence. From then on you may get bulk city automation, that would send notices which indicate that certain supplies or military aid is required. And once the player moves past that influence circle, only bulk region or country automation would be required; wars, weather events, cataclysms, leadership changes, that sort of thing. As each level is passed, the more detailed information informs the direction of the gross area, whilst moving back into the circle of influence reloads all the previous information, informs time passed and tries to propagate prior behaviour forward under the current new environment.
You're not looking to make a fully responsive or human AI. You're just looking to make one that operates believably enough without the direction of the player in their current job and in their current environment, and who can provide incentive for the player to act.
I -really- don't think this is beyond our current capabilities or knowledge, and would make a great game. You just wouldn't be able to tell a story through it, and that's all AAA wants to do.
I have to jump into the discussion although I do it very rarely. I just must remind one important factor in game creation.
You play games for fun, not to make meaningful choices. A world that is purely generated from random mathematical formulas determining everything that happens would take a load of resources to create and has the potential to randomly be very boring to the player. Minecraft has randomly generated worlds, however its selling point is not "explore the unknown" but "You can BUILD STUFF here". We are happy to explore the caves because HOLY SHIT, DIAMONDS! and not because we can find a cave. Finding a cave excites us because those have resources that we can build fun stuff out of once we leave the cave. So basically, Minecraft is Lego blocks which you have to mine out of the ground.
The Sims on the other hand are fun because it works like a doll house.
We also have games like Spelunky which randomly generate levels for us to explore. The random generation there however is NOT used to give you endless meaningful choices. The game just forever stays fresh with that Random Generator since you can never memorize "what will be in this room" so you are always extra careful. Or super reckless - also a choice.
So, since we are on the topic of meaningful choices - you really think a game would be better if the outcome was generated at random? Even if we had the technology, someone would have to be insanely smart to program the AI so that the dialogs never speak gibberish to us. Otherwise the stories would be really simple. You walk somewhere, an NPC approaches and asks "Wanna slay X monsters for me for Y reward?" "Wanna go to X to look for Y item?". I doubt this would be a fun game really. You would actually feel as if whatever you do has no meaning at all since there is no "Greater Goal" you are working towards.
/end of semi-related lecture
Strictly Episode related starts here/:
The "Illusion of Choice" mentioned in the episode should be read by future designers as "softening the blow" on the player. Let's be obvious here - in order to create a good game, you will have to ship it some day. You will NOT create a game that allows the player to do everything. Not in your lifetime. You can however do something so the player keeps having fun despite the fact you had to cut corners.
And that is why designers use "Illusion of Choice". Bad? Good? Meh, debatable. Honestly, as a designer I am ready to do it as long as the game is more fun thanks to it.
Alright, this got very lengthy now. Maybe I will reply later with the rest.
@ThomasWindar
The point of the game would be much like Minecraft, except instead of shaping the terrain, you'd shape the kingdom by aiding/hindering people you like/hate to power.
So the quests wouldn't be "Wanna slay X monsters for me for Y reward?", they would be "Kill off the crown and all heirs so I can seize power" or "Help me rally troops" or "Seek out my adversary" or "Find these resources" with no mention of how you are to acquire them. Open-ended quests which benefit the quest giver and which they'd probably attempt anyway.
I mean, if you really wanted to trim it down, just base it off GTA2 and have the gangs in a constant power struggle against each other outside of scripted quest events. So you always have this tension raging which the player can step into to affect or can completely ignore. The quests there would be less quests and more notifications of imminent powerplays. And we definitely have the AI to do this, as it would only take an RTS AI to control all the independent gang members that make up each faction.
@discrider
Your example in this case is not exactly showing any choice problems.
In the example you show there it looks like this from a gameplay perspective:
- Obtain quest from a randomly generated NPC to find X person/item.
- Interact with Person/item X.
- Return to NPC to finish quest. (optional)
- One of the "Faction Bars" increases.
In every single thing you do.
I'm not trying to criticize you, just pointing out that I don't think randomly generated content is a good idea. There is a limited amount of things you can randomly generate this way that actually make sense and will ultimately lead to the player doing "the same damn thing I always do in this game" just named differently. The only thing changing is the place you have to go and the skin of the NPC/Item. At first it might be fun, but after 3 hours of doing this random searching for stuff you will feel like "I've been doing the same thing for 3 hours now". You would have to have a REALLY GOOD mechanic for it that makes gameplay so awesome that the only thing you care is "when can I do this AGAIN?!" to utilize random generation.
For that example, speaking RTS games - the Meaningful Choices you make are based of the fact that in a "good RTS" different strategies have different effects, changed by what your opponent does. The content in the game is the same, the way it is used however is different because the human behind the other screen is unpredictable. And NOTHING in there is randomly generated. RTS games are carefully crafted.
Ultimately it falls down to the category of "how much you can code into the game before shipping it". And even then, it will always be less than you wanted and you will have to put those "soldiers that block your way" in a place you would otherwise have to put an invisible wall.
If you're talking about my GTA2 example, then progression wouldn't be marked by faction bars, and instead be marked by the faction you're actively assisting gaining new territory or resources up until you've forced the other two factions more or less off the map entirely. It would also be less quests and more seeing two factions clashing in the streets and choosing to help out, with maybe the leaders asking for your services for particular specific forecasted missions, aiming to take control of vital areas.
Such gameplay would be supplemented with the standard GTA2 wreak havoc for no reason gameplay.
If you're talking about Best-RPG-Evar [Working Title -Do not steal], then Faction Bars make even less sense. If you help someone usurp a throne, then the king is dead, long live the king. If you help someone gather an army, then they have an army and march out on the neighbouring kingdom. If you gather materials for someone, then they get more skilled at their job, or can build new items for their town/you, or get wealthier and can hire more apprentices and build a bigger store.
There would be some sort of favourtism shown by those you've helped towards you, so requests to them are more likely to succeed, or they might put up with your antics more, or they might be using you and not care at all and backstab you for no reason. But largely, the things you do to help people would have immediate flow on effects to their business and the environment and not just to some arbitrary construct that determines when the next mission unlocks.
[PATV] Wednesday, September 18, 2013 - Extra Credits Season 7, Ep. 2: The Illusion of Choice
Posts
SR4 has a lot of other meaningless choices in it, the first of which is character customization. Nothing in customization means anything in the game. Not the voice, not the build, not the clothing you wear. But it invests you in the character, so it isn't completely meaningless. Early on, I bought the Presidential Suit again and I spent the entire game playing in that. It was kinda fun kicking butt in a nice purple suit. Added the Morpheus glasses later, and I felt like a total badass. It was awesome. Meaningless to gameplay, but awesome to the experience.
People should learn to embrace the illusions.
har har.
BTW, @Diamondback11, Extra Credits did a couple of episodes dedicated to the Walking dead a while back.
@MasodikTiasma Preventing the player from replaying a conversation and realizing that the choice was not only completely illusionary, but also completely ignored by the game, is hard to imagine.
Just something to think about I guess.
Mechanically, on the other hand, it had plenty of real choices, as picking different means of reaching a goal allowed you to completely bypass obstacles on other paths and play to whatever strengths you had chosen through skill assignment and equipment choice.
The truth is that branching paths which go back to the same storyline after a while simply suck. They are not clever at all. The player notices them, even more if he plays through the game twice or more. I can justify them because of the developing costs but you seem to consider them the coolest thing in the world. They are not.
And please, don't tell me that real branching paths are impossible to make, you said who managed to do almost everything right: The Witcher 2.
http://www.worldofleveldesign.com/categories/wold-members-tutorials/magnar_jenssen/functional-lighting-magnar-jenssen.php
The thing is, you probably even know that now you have to choose to do something. But the game doesn't just hit you with the list of options, it makes you to figure it out yourself what to do. This is how it should be done more often. It's not like choosing your meal from menu in Burger King, it's like making our food yourself. This hit me really hard. just... Awesome.
Of course, the whole game is just pure awesomeness.
Also, sometimes you really can give the player nearly limitless choice. Just look at Minecraft. You have to be willing to sacrifice narrative and structure to some degree, but procedural content can go a long way to making real choice on a huge scale.
Yeah, just keep telling yourself that.
Here's reality: Illusion of choice is bad because it deprives the players of agency. That's reality.
"But TD!" the developers cry, "Illusion of choice is a wonderful thing! Look at Mass Effect!"
Yes, let us look at Mass Effect for a moment. It is a wonderful example of why this is bad.
What happened in Mass Effect? People really ragged on the third game. They ragged on it a lot. But why?
The answer is the illusion of choice. The reason Mass Effect was so engaging was that you had to make big decisions - decisions with potentially unknown moral impacts down the road - and when you made those decisions, the game promised you that they mattered. Did you release the queen, or kill her? Did you blow up the Collector base, or leave it intact for Cerberus? Did you save the Council, or did you let them die? Better to commit genocide on the heretic geth, or reprogram them? Did you destroy the research on the genophage cure, or preserve it?
These were all big, weighty decisions, and you could see that there was potential for disaster either way. And the games promised you that they would remember these things, that these things were big decisions, that they were important. The games got away with these decisions not mattering in the game because they pushed it off to the future.
The problem was, there was no future after Mass Effect 3, so all your choices had to matter then. And they didn't. At all. And that basically told the players, "Your choices don't matter now, and they never did." The ending should have had you fighting alongside allies, getting covering fire from the people you chose to save and convinced to ally themselves with you. Dragon Age: Origins had your allies show up and help you out against the final boss, so those big weighty decisions you made felt like they really mattered, and there were some differences between the way that your allies fought - it wasn't a totally different battle or anything, but it made the game different, and your choice actually did matter, even if the actual cost of said implementation was quite low.
Mass Effect 3 utterly failed in this department. The choices you made didn't actually matter, and that sucked. There was no choice, only the illusion, and that killed the whole thing for people. It wasn't just the ending which had the illusion of choice, it was the entire game. Your decisions were supposed to have weight. Your decisions were supposed to matter. They did not.
It is true that you have limited resources. It is true that the illusion of choice works sometimes. But the problem is that if you give players the illusion of choice about something that actually matters, you're breaking a promise to the player, which is that the game isn't lying to them. WSIWYG is one of those basic, vital promises that you make to your players, that you aren't lying to them, deceiving them about things. Invisible walls are an example of a broken promise to players - if the players hit invisible walls, then you're breaking the promise that they can go over where they are looking. The illusion of choice is the same sort of thing.
I can tell you from experience with roleplaying games - real ones, that is to say, tabletop roleplaying games, not CRPGs - that you should avoid the illusion of choice whenever possible. You basically make a compact with the players - they don't intentionally avoid what you've designed, and you give them reign to make choices. If they feel like their choices don't matter, you've betrayed them. The same is true in any medium.
Incidentally, as far as the whole "highlighting" things goes, I have to say that games have trained me to avoid such things because hey, no one hides secrets there. I go down the obvious path last. I think if you really want to encourage such behavior, you need to be smart about how you do it, and never punish them for following the obvious path. Them missing secrets is punishment.
I'll be real honest here: Chrono Trigger is one of the best RPGs of all time, and the game gives very little in the way of illusion of choice. You are going from point A to point B all the time, with very little room for wandering around at any given moment, and yet, you don't care. Why? You WANT to keep moving forward, to keep going in the direction the game is telling you to go, and the game doesn't pretend like you have a bunch of choices but you end up in the same place anyway. It isn't about lying to you, it is about being compelling.
A lot of FPS's also have very little in the way of illusion of choice. Many platformers have 1-2 paths through any given area. The list goes on. Many, many games where you have very little in the way of illusion of choice are great games, and people love them. People don't complain that you don't have any choices in Chrono Trigger because the idea that you're supposed to have choices in that game never even occurs to them; the choices you have are how to beat up the monsters that get in your way.
This is true of most SNES RPGs. This is true of many JRPGs. This is true of many games in general. Yes, you can easily make a game terrible in this way, but how does the illusion of choice prevent the game from being awful? FFXIII would still be a bad game even if you had the illusion of choice; it was poor design that doomed it, in the end.
I do understand the trick with dialogue in games like Mass Effect or Dragon Age, and players are okay with most dialogue being mostly meaningless save possibly altering some minor numerical variable one way or the other. But when you make "big decisions", and they don't matter, players get upset. And that is where the illusion of choice breaks down; in moment-to-moment, transient decisions, it upsets players a lot less than in big ones if they don't matter because they don't expect the little decisions to matter that much most of the time.
The point in Mass Effect that angered almost everyone wasn't aggravating because it only gave the illusion of choice, it was aggravating because the illusion of choice failed and people noticed that everyone got the same ending.
Giving two endings with substantially different content is not that hard. Go try playing NWN2 sometime. There's a branching ending that probably took little to no effort (evidenced by the fact that it didn't get cut when obsidian got pressed for time). The 'evil' ending uses no AI, models, or scripts that hadn't appeared before, but it still felt like the player was making a significant choice. (To betray your party or to do something remotely rational)
For one thing, the very first part of the game at the millennial fair and the trial afterwards was a brilliant example of the illusion of choice. Chrono Trigger and other old school JRPGs were good at creating an illusion of choice because they allowed you to independently explore whatever town/dungeon/map you were on.
The choice you made was where you decided to go. The illusion is that nothing really happens until you get to the place the game designers intended you to go. The illusion is further reinforced by providing rewards for exploration, i.e. random treasure chests everywhere. How many times has a game prompted you to go somewhere, yet you purposefully go the opposite direction because you want to make sure you find everything? All of those moments are choices.
FF13 caused so much outrage about "linearity" because it shattered this illusion by forcing you to walk through one-lane corridors and gave you very few choices as far as direction until 3/4ths of the game was done.
The problem wasn't just with the ending, it was with the entire game. The first two games, you made choices which seemed to matter, but when the third game rolled around, they did not.
More or less, the problem was that they made promises that they just couldn't keep. It wasn't the ending; if it had just been the ending, yeah, people would have been angry. But the entire game primed you for the ending, because the entire game, your choices did not matter.
Basically, they made the first two games seem a lot grander with promises of future content which not only never materialized, but which they simply were not apparently up to the challenge of producing. The bigger the decision, the bigger the impact, the better it feels to the player... but the bigger the challenge you have, as a designer, of dealing with those choices.
Ultimately the purpose of the illusion of choice is to improve player engagement. The larger the choice you present the player with, the more involved they are - but if you turn around and then make that choice not matter, then the player feels lied to. This is because they were lied to.
The very first "big" decision you make in The Walking Dead is a false choice, and the fact that it was a false choice robbed me of my feeling of agency in the story. I chose "wrongly", the character I tried to save died anyway, and it became obvious to me that the game was trying to cheat me.
If you present a significant choice and don't really mean it, what is the point of presenting the choice in the first place? The cost of people feeling robbed is not worth the benefit.
For minor dialogue options, that's fine because they at least feel like they're saying something, even if it doesn't really matter. But if you have some sort of significant choice, choose opposite the way that the story goes... well, it is just bad. The story very frequently flows in various ways without the player's direct intervention, so why are you involving me only to say "Just kidding! We don't care what you think!" ?
This is the problem. I don't feel that all of these choices actually DO add to games, and I don't feel that players miss a lot of false choices.
Take for example the fact that you are the Dragonborn. A few characters may acknowledge you but the vast majority will pretty much treat the same way every time even if you do something big for the city or the leader of the city, nothing really changes.
Another good example are the guilds, the Companions, the College, The Thiefs guild, or the Dark Brotherhood. You can pretty much take part in every single one of them, become the leader of every single one all at the same time. No one ever questions you, no one ever calls you out on being the leader of the Companion for example and being a member of the College. There are never any consequences for really any of the things you do, nothing is ever kept with the exception of a few lines of dialogue and maybe an item or two for your decisions.
I disagree with this episode in pretty much its entirety.
With the illusion of choice, I agree with much of what @Titanium Dragon is saying below. If I were to sum it up, I'd say that every choice should have an appropriate payoff based on the scale of the choice. And that appropriate payoff proves that it can't be an illusion. It doesn't matter if the grand story arc reconverges. So long as the story arc does diverge on the choice, then the player is making a choice, a real choice, and so long as events following this divergence fit the scale of the choice, the player is not going to feel cheated.
It's when the game is utterly dismissive of the player's actions by either ignoring them or not giving them appropriate impact that the "illusion" falls through. And that's not the player failing to believe in the decisions being real, but rather the game failing to grant the agency that the player is expecting.
With FPS games, all the mentioned techniques always feel like being led by the nose through the level. They never provide me with an illusion of choice. But that's okay because what's really leading me through the inherently linear gameplay is the narrative, and the urge to see what's next, not some misguided notion that I'm going wherever I like and kicking butt whilst doing it. These techniques only help me align myself with the level and prevent me from becoming lost, and nothing else.
If I'm caring that I'm being directed down a linear corridor at a goal, then the FPS is doing something else fundamentally wrong. Either, the narrative isn't powerful enough to draw me through the gameplay, so I don't want to go down the linear corridor because the other end just doesn't seem interesting enough. Or, they've presented me with a gameplay choice that ignored my input, such as presenting me with enemies laying down razorwire, but they're invulnerable and there's an invisible wall between me and them. I wouldn't have cared if that corridor were not there or empty. But the game hinted at an alternative course of action, but that action turned out to be impossible.
I also think it's entirely possible to give the player all the choice, but not without sacrificing anything resembling a narrative. All you'd need to do is create a Minecraft-like game (infinite procedural world) with characters in it who are themselves procedurely generated with motives and desires, and then the gameplay arises with how the player character interacts with these NPCs and who they try to help and hinder. There would be a specific social order and rules governing how NPCs can react and engage with each other to get what they want, and in such an environment, NPC villians and heroes can arise just through the actions they are willing to carry out (or allowed to carry out) in pursuit of their desires.
The real difficulty here is not the asset cost; you'd have to plan for a lot of conversational paths but you could then reuse the tree for each character with reskins on the conversation to provide some differentiation. Where the agent starts in the tree determines where his choices can move him to, and provide the depth of characterisation, and such a tree I can't imagine being more work than writing the narrative of a traditional storyline, albeit it would be harder to test.
The real difficulty is the level of simulation required to be believable. Getting a city full of agents interacting realistically would require a lot of processing power. We might have beefy enough processors now, but have yet to see someone implement such a game, or we might need more power before such a game is possible.
It's not really freedom because the story doesn't progress until you travel to the story missions. And the sidequests do nothing to the story progression.
What you're complaining about isn't illusionary choice but about choices that don't feel MEANINGFUL. They are completely different things. An illusionary choice should feel just as meaningful as any other choice would, and if it doesn't then it isn't because it's an illusionary choice but because whoever wrote the choice in didn't know how to use illusionary choices. It's a tool - nothing more. If you play video games then you've likely made a million of them and never even noticed because you thought they were real choices.
Walking Dead is a better example of Illusion of Choice. While I still felt a bit cheated by the I of C - which was quite obvious in that game when characters you specifically saved arbitrarily die - there was also less of a time investment to playing the game... and that arbitrariness fits within the world we were given. The story was still strong when we only had a small amount of control over the world, because in the end, death takes us all. All we can do is choose how we face it, how we live in the meantime.
There is a tendency nowadays to look at how much playtime you get out of a game. I would have actually gotten more playtime from the ME series had ME3 been about 30-60 minutes long, but seriously took my choices into account for the ending. I would have replayed ME1 and 2 over and over again until I found an ending I was satisfied with.
In short, Illusion of Choice has its uses, I'm sure - but not in games where they tout 'Meaningful Choices'. So many choices in ME1 should have led to wildly divergent story paths, and they should have honored that bargain with the players. Even Dragon Age: Origins, one of their earlier works, did it better, and 'Choice' was not a selling point of the game.
The Sims where the player is a Sim and where the land is procedurally generated and not just restricted to a lot would be pretty darn close.
You wouldn't need as deep a biological sim, and the random global wants would turn into quest hooks that the player can choose to satisfy and which the sims would work try to satisfy themselves. Jobs could be predefined on creation or on age level up. And the relationship network is there to help determine what actions a sim is likely to take to advance their global needs.
Like I said, the main problem would be getting hundreds of sims automated at the same time. But until we can do that, you could always cheat by making lower classes have only basic pathfinding AI as since they don't affect any larger systems, they don't need as detailed a automation, and then fill the cities with such shells. Potentially if the player befriended/fought with such a citizen then they would receive a full AI treatment as the player has expressed interest in following the character.
Also moving away from a population city would slowly wind back its automation. Citizens would stop being automated at a certain distance because you can't really see what they're doing through walls. Then special character need satisfaction requests would stop outside the city's influence. From then on you may get bulk city automation, that would send notices which indicate that certain supplies or military aid is required. And once the player moves past that influence circle, only bulk region or country automation would be required; wars, weather events, cataclysms, leadership changes, that sort of thing. As each level is passed, the more detailed information informs the direction of the gross area, whilst moving back into the circle of influence reloads all the previous information, informs time passed and tries to propagate prior behaviour forward under the current new environment.
You're not looking to make a fully responsive or human AI. You're just looking to make one that operates believably enough without the direction of the player in their current job and in their current environment, and who can provide incentive for the player to act.
I -really- don't think this is beyond our current capabilities or knowledge, and would make a great game. You just wouldn't be able to tell a story through it, and that's all AAA wants to do.
You play games for fun, not to make meaningful choices. A world that is purely generated from random mathematical formulas determining everything that happens would take a load of resources to create and has the potential to randomly be very boring to the player. Minecraft has randomly generated worlds, however its selling point is not "explore the unknown" but "You can BUILD STUFF here". We are happy to explore the caves because HOLY SHIT, DIAMONDS! and not because we can find a cave. Finding a cave excites us because those have resources that we can build fun stuff out of once we leave the cave. So basically, Minecraft is Lego blocks which you have to mine out of the ground.
The Sims on the other hand are fun because it works like a doll house.
We also have games like Spelunky which randomly generate levels for us to explore. The random generation there however is NOT used to give you endless meaningful choices. The game just forever stays fresh with that Random Generator since you can never memorize "what will be in this room" so you are always extra careful. Or super reckless - also a choice.
So, since we are on the topic of meaningful choices - you really think a game would be better if the outcome was generated at random? Even if we had the technology, someone would have to be insanely smart to program the AI so that the dialogs never speak gibberish to us. Otherwise the stories would be really simple. You walk somewhere, an NPC approaches and asks "Wanna slay X monsters for me for Y reward?" "Wanna go to X to look for Y item?". I doubt this would be a fun game really. You would actually feel as if whatever you do has no meaning at all since there is no "Greater Goal" you are working towards.
/end of semi-related lecture
Strictly Episode related starts here/:
The "Illusion of Choice" mentioned in the episode should be read by future designers as "softening the blow" on the player. Let's be obvious here - in order to create a good game, you will have to ship it some day. You will NOT create a game that allows the player to do everything. Not in your lifetime. You can however do something so the player keeps having fun despite the fact you had to cut corners.
And that is why designers use "Illusion of Choice". Bad? Good? Meh, debatable. Honestly, as a designer I am ready to do it as long as the game is more fun thanks to it.
Alright, this got very lengthy now. Maybe I will reply later with the rest.
The point of the game would be much like Minecraft, except instead of shaping the terrain, you'd shape the kingdom by aiding/hindering people you like/hate to power.
So the quests wouldn't be "Wanna slay X monsters for me for Y reward?", they would be "Kill off the crown and all heirs so I can seize power" or "Help me rally troops" or "Seek out my adversary" or "Find these resources" with no mention of how you are to acquire them. Open-ended quests which benefit the quest giver and which they'd probably attempt anyway.
I mean, if you really wanted to trim it down, just base it off GTA2 and have the gangs in a constant power struggle against each other outside of scripted quest events. So you always have this tension raging which the player can step into to affect or can completely ignore. The quests there would be less quests and more notifications of imminent powerplays. And we definitely have the AI to do this, as it would only take an RTS AI to control all the independent gang members that make up each faction.
Your example in this case is not exactly showing any choice problems.
In the example you show there it looks like this from a gameplay perspective:
- Obtain quest from a randomly generated NPC to find X person/item.
- Interact with Person/item X.
- Return to NPC to finish quest. (optional)
- One of the "Faction Bars" increases.
In every single thing you do.
I'm not trying to criticize you, just pointing out that I don't think randomly generated content is a good idea. There is a limited amount of things you can randomly generate this way that actually make sense and will ultimately lead to the player doing "the same damn thing I always do in this game" just named differently. The only thing changing is the place you have to go and the skin of the NPC/Item. At first it might be fun, but after 3 hours of doing this random searching for stuff you will feel like "I've been doing the same thing for 3 hours now". You would have to have a REALLY GOOD mechanic for it that makes gameplay so awesome that the only thing you care is "when can I do this AGAIN?!" to utilize random generation.
For that example, speaking RTS games - the Meaningful Choices you make are based of the fact that in a "good RTS" different strategies have different effects, changed by what your opponent does. The content in the game is the same, the way it is used however is different because the human behind the other screen is unpredictable. And NOTHING in there is randomly generated. RTS games are carefully crafted.
Ultimately it falls down to the category of "how much you can code into the game before shipping it". And even then, it will always be less than you wanted and you will have to put those "soldiers that block your way" in a place you would otherwise have to put an invisible wall.
"Faction Bars"?
If you're talking about my GTA2 example, then progression wouldn't be marked by faction bars, and instead be marked by the faction you're actively assisting gaining new territory or resources up until you've forced the other two factions more or less off the map entirely. It would also be less quests and more seeing two factions clashing in the streets and choosing to help out, with maybe the leaders asking for your services for particular specific forecasted missions, aiming to take control of vital areas.
Such gameplay would be supplemented with the standard GTA2 wreak havoc for no reason gameplay.
If you're talking about Best-RPG-Evar [Working Title -Do not steal], then Faction Bars make even less sense. If you help someone usurp a throne, then the king is dead, long live the king. If you help someone gather an army, then they have an army and march out on the neighbouring kingdom. If you gather materials for someone, then they get more skilled at their job, or can build new items for their town/you, or get wealthier and can hire more apprentices and build a bigger store.
There would be some sort of favourtism shown by those you've helped towards you, so requests to them are more likely to succeed, or they might put up with your antics more, or they might be using you and not care at all and backstab you for no reason. But largely, the things you do to help people would have immediate flow on effects to their business and the environment and not just to some arbitrary construct that determines when the next mission unlocks.