Alfe Clemencio of Sakura River Interactive is the author of highly branching visual novels in Ren’Py. His previous work Fading Hearts features a wide range of possible player paths and outcomes; now he is working on an ambitious RPG project called Don’t Save the World. His Indiegogo page describes Don’t Save the World thus:
Don’t save the World: An RPG is a game where the effect of player’s choices are so strong they can change the genre of story and game. Live a life of adventure (RPG gamplay) or a normal life of running a shop (management-sim). Say “No” to saving the world!
…Near the end of the game gamers might be given the chance to slow down or stop the hero from defeating the dark lord.
I will guarantee that some players trying to be “good” will try to stop or slow down the hero.
In this scenario you are not the hero and won’t be defeating the dark lord. If the dark lord isn’t stopped then all the lands will be flooded with monsters that will bring the cities and towns to ruin. The hero is definitely a good person and is trying to do good.
It’s because moral choices like this that morality meters won’t work for this kind of game. Can you figure out why gamers trying to do good would do something like stop the hero?
Here he talks about that project, about the challenges of managing highly-branching narrative, and about the moral elements he is hoping to explore in his new work.
ES: It appears that Don’t Save The World is mixing together several genres of game mechanic. Are there also several genres of narrative? Do you get a different kind of story if you choose to play the management sim vs the RPG?
AC: The are several genres of narrative in the game. However just because you play more of the RPG doesn’t necessarily mean you’ll get an adventure story. This also goes for the management sim game elements. A player could just use the RPG elements to gather resources or making traveling easier for business. Highly advanced RPG players could figure out how to go through the more adventurous and dangerous plotlines while using mostly management-sim game elements and extremely low-to-zero grinding. The latter’s example was inspired by players who do low-level and/or “healer-only” runs in traditional RPGs.
ES: How do the different game genres embedded in Don’t Save The World benefit from being presented together in one story? Why does the management sim (e.g.) benefit from appearing alongside an RPG?
AC: There are four reasons that come to mind for this:
– Variety in gameplay
– Cost savings on art/content assets and marketing
– targeting a wider audience
– to have a certain theme in the plot that’s harder to express otherwise
Including more game genres increases the variety in gameplay at a fundamental level. Players will have simply more to do. Two games worth of gameplay mechanics might actually be a little bit intimidating to the player if not presented in smaller chunks.
Producing one game with two game genres inside it can be cheaper than producing two separate games of separate genres. Art assets, marketing, and other overhead costs of producing and selling a game are shared. However it’s still possible to have costs be more than more than two separate games.
You can target a wider audience with multiple game genres in one game. However you need to be really careful with including multiple gameplay genres. Do it wrong and only the players who like both game genres will enjoy it. It’s the difference between targeting players who like genre A “AND” B and targeting players who like A “OR” B. This is where player’s choice is important as they get to choose how much of A and B they would like.
As for the themes that can be expressed by having two separate game genres, one example would be how a business person compares to a traditional RPG hero. Comparison about what a player can potentially do in either role is almost guaranteed to happen. Then that’s when conversations may get interesting.
ES: Are there things about the story that the player can only understand by replaying the game in several different styles?
AC: If the player were to never have a conversation with another person about the game then yes. Without knowing about the different points of view, the player may be missing information that is required to understand certain things about the story. However what usually happens is that they go online somewhere to compare experiences or get information.
It’s not as big of a spoiler if an ending is not guaranteed to happen like it is with static non-interactive media. A player can still feel tension in the interactive story even if they know all the endings. Why? Because there is always a chance that they missed something and what they expect might not happen.
ES: This is an experience I had somewhat with Fading Hearts — finding that as a player I sometimes felt like I had ended the game missing key information. How much do you think it’s fair to expect players to construct the meaning of your story from pieces outside the game? How does that affect your design decisions?
AC: A general goal when designing is to make any and all possibilities/choices/challenges/mysteries available and solvable to the player when they first encounter them for the general case. For Fading Hearts about 90% of those challenges met that goal so there is still room for improvement. One out of ten challenges isn’t good enough and several people usually get hung up on the same ones.
However some of mysteries are actually designed to be extremely hard and subtle so that some players will never realize there was clue to a big secret. To solve these mysteries requires a decent amount of in-depth story analysis. These aren’t required to enjoy the game and don’t impact gameplay but to those that do figure it out, it’s a great feeling.
“Trial and Error” strategies are generally discouraged and several challenges are designed specifically against them. Getting players to formulate theories about how the forces of the plot work is what we try to do. One good example was the elusive worst-yet-best ending in Fading Hearts named “Sole Survivor”. If players tried to map out and try all the possibilities in a branching story diagram, they still wouldn’t be able to consistently reproduce that ending.
However the writer for a walkthrough of Fading Hearts nailed the exact causes for that worst ending. She formulated a theory about how that event happened after thinking about it and tested it out. Then discovered that it worked. As a game designer it made me very happy that someone thought about plotline and discovered its secrets themselves.
ES: One of the big challenges for games that offer multiple play types (open world RPGs, sandbox games) is giving the player significant freedom while still signaling important plot elements so that the player can follow the story arc if he wants to. Is that something that has been tricky for this project, and, if so, how are you resolving it?
AC: It’s not quite as tricky once two different techniques were figured out of placing plot hooks. Broadcasting plot hooks and person/place/thing plot hooks.
Broadcasting a plot hook involves presenting to the player the public knowledge about a recent major event in a story arc. The player will likely not miss a broadcasting plot hook. The insider details of the story are usually left out and the information might be inaccurate. Good examples of these would be things like an in-game newspaper, town bulletin board, or random overheard gossip. These usually give the player an indication of where the interesting things are happening as there is typically a location mentioned.
For plot hooks that are persons, places or things they usually give the player more of the in-depth information that isn’t public and a chance to get directly involved in that plot arc. These plot hooks are more likely to be missed since they require the player to be in a specific area at least. What can be done to help mitigate this and make it a bit more interesting is to have more than one same or different plot hooks to the same story arc.
One simple example is to have every item of a fetch quest be an entry point of that fetch quest. A player will more likely be able to stumble upon at least one of the items naturally and the quest can start from there.
ES: You talk about a choice at the end of the game where some good-aligned players will be motivated to try to stop the hero. This reminded me of some of Victor Gijsbers’ work in interactive fiction as well as tabletop storygames that focus on exploring knotty moral problems by asking the player to choose the least evil of several undesirable outcomes (rather than having a clearly demarcated Good and Evil). I know you don’t want to give too much away, but how does your approach compare with this?
AC: I may actually try to tempt players into the mindset of “lesser-evil” when there may be another way that requires a very high degree of skill in either gameplay genre. Also the player may be on the receiving end of an NPC who made a choice of the “lesser-evil”. Won’t always happen but the chance is there if the conditions are right.
But back to the question.
There are two more factors that are added to the moral choice equation that brings it at little closer to how moral choices are made in real life. They are “possibility of incomplete information” and the “doubt of ability to execute a choice”. I’ll explain these in relation to the “stop the hero” moral choice.
Usually in video games and non-interactive media, all the required/relation information that the player will ever get or need is guaranteed to be already given for a player to make a moral choice on the first playthrough. That may not be the case for this game. They may be making a choice under “incomplete information”.
ES: A common complaint about games that present moral choices with incomplete information is that the player can feel that he’s been duped by the author or set up for a fall — maybe the game didn’t give him a chance to find out that additional information! If it’s a moral failing in real life not to investigate a situation fully before acting, then a game that doesn’t clearly present the opportunity of investigation has rigged the situation against the player.
What kinds of design choices are you making around this? Are you most concerned about being fair to the player, or about setting up an instructive failure for him, or something else?
AC: The key issue what is the definition of a “fair chance”. For non-interactive media like in golden age detective novels I believe, the reader must be presented with all required clues to solve to be “fair.” However this changes with interactive media. A “fair chance” now can just constitute highlighting where to find the required information and giving the player ample opportunity access it. Something like always checking an updated town bulletin board. However the players must also agree that it was a “fair chance” even after they find the answer.
As for what is going to be actually used in our games, the player should always have a chance to beat a challenge or had ample notice that some sort of challenge was heading their way. Even on their first playthrough. Player understands the dark lord is a threat to the kingdom that he is in. Having in-game research on the dark lord and the other hero and the suggestion of doing research should be considered fair enough. The player doesn’t need to but understands that they risk missing something they need by skipping the research. There might be even more reasons appearing later to do research to add even more of a fair chance.
A popular and interesting case study of “fair chance” of having info required in making a choice is in Deus Ex: Human Revolution. There is one particular choice in that game which has a lot of players falling for the trap and a lot of players being amazed at how many fall for the trap. Reading on the forums there appears to a very staggering list of clues and reasons subtle and/or definite spread in emails, dialogue, and even visual clues if you pay attention to your surroundings when certain things happen.
The conspiracy theorists caught on in the initial prompt to make the choice. Many others did not despite the large amount of clues spread everywhere. This shows the wide different range of amount of clues needed for every player needs to be taken into account.
ES: So you were saying, about how a player might respond to incomplete information…
AC: For any player that doesn’t understand why good-aligned people may try to attack the other hero, they can perceive most of their good endings as a happy ending. When they do understand why, a lot of their good endings may suddenly feel like they have a tint of sadness and regret that nags at them and won’t go away. Especially since they know there is something that can be done. It’s gets worse when they hear of other players gushing so brightly about their “good” endings.
If players just try to attack the other hero blindly just because they heard it’s somehow good, it won’t turn out well. They will likely get one of the worst endings possible. Other things need to be done as well.
So then the player decides to stop the hero somehow. How does one do that? There are multiple ways actually. But does the player have what is needed for any of those methods? Do they have the combat skills or influence? Can they discover another way that doesn’t require any of that? This is where “doubt of ability to execute a choice” comes into play.
For most games with choices, you really don’t question if you can do a perform a given choice or not. Don’t Save the World is a bit different as players may have to figure out how to follow through with their choices. It could be just some random rumor because it seems kinda hard and outrageous. Now you’re doubting the possibility of it so you don’t try in the first place.
ES: It’s clear you’re imagining a lot of different ways (both mechanically and narratively) that a player could experience the later parts of the game. How are you approaching your plot structure, and keeping content generation from getting way out of control?
AC: One of the more interesting parts about the plot structure is that the player can cause delays in the triggers of plot events. Stop or delay the hero, for example. Inter-subplot dependencies with dynamic timing of event triggers can become a nightmare if you don’t plan properly however. An extreme example is if two characters exists in many subplots and both can potentially die separately in the story before or after many events. The combinatorial of potential states is quite scary if you do the math. However, since you are the creator of the plot rules you can just choose to do certain things to make it easier. Fading Hearts’ word count is less than a novelette of 75K words despite the high degree of player freedom and choice.
Math proofs and number theory are used very heavily to help understand what sequence of events is possible and not possible. Using a technique like this helps reduce the the “possibilities” to a more manageable level yet be nearly invisible to the player.
Here is an example. So let’s say person A can come to visit every 9 days and person B can come to visit every 10 days should the player invite them. This is starting from day 0. They are both moral enemies and should they ever visit on the same day they will both kill each other. Nothing else can kill them in the plot. Many plotlines are actually dependent on them. There won’t be any weird plot trickery where they come back as a ghost instead doing the exact same thing as if they were alive. From what day do you start to need thinking about the possibility of them being dead?
The answer is day 90 as it’s the least common multiple of 9 and 10. So for all the events that only can happen before day 90 you can safely assume that they are alive. No need to write a separate set of dialog to handle the “Key character is dead” case till after day 90. That’s a ridiculous amount of time and work saved. All because you choose 9 and 10 as the delays.
ES: Is there a particular moral theme or problem that you’re interested in exploring through this game?
AC: A moral theme I want to explore is how morality can be very complicated. The story of how multiple people trying to do good end up fighting each other is one example. Causing the player to be one of the actors in that story while having so much freedom and free will is one of the goals of this game.
I also want to explore themes of moral business choices, the extremely different perceptions of money that exist, gender roles, and how perceptions can change greatly with additional information.
ES: Does the player have different types of moral choice depending on whether she’s chosen to play the game as a management sim or as an RPG?
AC: There are a decent amount of moral choices that are available to both play styles. However the difficultly and methods in reaching them are different for the different game genres. There are a few that will only happen in one of the gameplay genres. Nearly every day in the game you can chose which style to play. You can choose both but every action takes time. Time advances all storylines a little. So you better be good at making your actions count.
Emily Short's Interactive Storytelling 
I kind of felt like the interviewer didn’t understand RPG games very well. All I kept thinking about when reading the interview was the Fable series. All of those have a management sim aspect, though usually only as a way of making money to supplement the main RPG quest line. And now that I’m thinking about it, all complex RPG’s have a management sim aspect. That’s why people go “farming”…Usually it’s limited to crafting and there’s no story tied to it, is the only difference I see.
Well — it was more that I was trying to get at what Alfe thought the difference was, given the stress he puts on this. (I did ask him if he’d played the Fable series, since I thought that would be relevant to both the moral angle and the gameplay variety, but since the answer was just “no”, I dropped that from the final article.)
I think maybe you’re the one that doesn’t understand… The subject was “the interactive narrative of Don’t Save the World,” not the entirety of Don’t Save the World ^_^ So of course there weren’t going to be any questions about the battle system and whatnot. This particular style of interactive story is really the heart and soul of the game, and what will set it apart from other RPGs. Emily did a great job of asking informed questions about that, and I enjoyed reading the interview.
Interesting. A two-channel mechanical approach with conventional RPG play sitting alongside a sim-shopkeeper game isn’t exactly novel – it’s basically what the “Atelier” series of console games does, or the recent-ish PC title “Recettear” – but I can’t think of one that’s really gone in for the branching-narrative approach.
I pretty much love the “Atelier” series of Gust’s games. So you’ll probably see a lot of influence from those games. Although the Atelier series focuses on much more of the “item-crafting” in which it Gust was the pioneer and still #1 in IMO.
However the “item-crafting” in Don’t Save the World will be critical in triggering a plotline that will never occur otherwise. I can make the story reference actions that the player perform in-game in some cases. Why? Because players usually behave in certain somewhat predictable ways that is more consistent than you think.
Players will always gravitate towards getting better weapons and gear right? Why not make a story about that and have it triggered if the player makes certain really powerful weapons?
But the underlying reason for better weapons and gear is better damage-per-second and survive-ability right? What if some combination of skills and mostly average weapons actually does more damage than the most powerful weapon? Then players will figure this out and start using exclusively in fighting.
Figure out what those skills are and listen in for players doing that. Have certain plotlines trigger based on that. Then you have a strong connection between gameplay and story.
Of course this requires the writer to understand the mechanics of gameplay as well. So if you get writers that “don’t really play games” then you need to do a little more explaining to them of those expected strategies. Ideally the writers should spend time playing with the game to discover strategies so that those strategies can be referenced in their writing.
But if you bring some award-winning novelists to write and they refuse to play the game? Well… you’re kinda stuck.
That sounds really interesting…but doesn’t it make the most powerful weapon useless unless it sells for a fortune or triggers some sort of plotline? And if that’s all it does, you wouldn’t need to make it “the most powerful weapon” because that would be essentially untrue–it might as well be some ceremonial weapon. Players are going to feel cheated if the “most powerful sword” doesn’t live up to its name (and presumed difficulty in finding or making it), not to mention that it’s going to look like poor balance on your part. I think it might be better to say an inventive combination of X, Y, Z can be serviceable, but if you really want to be the best, you’re going to have to go to the trouble of acquiring the best. There are players who won’t be satisfied until they’re the most powerful they can be, so that in and of itself is reward enough for them ^_^
I suspect you’re looking at it the wrong way ’round. Grinding for the biggest numbers and employing system mastery to find the mechanically optimal combinations are two distinct modes of play that appeal to different subsets of players – and you generally don’t *want* achieving the latter to require engaging in the former, since the system-mastery types often get into it precisely to circumvent the grind.
@David: Well, Alfe said this: “What if some combination of skills and mostly average weapons actually does more damage than the most powerful weapon?” Not that the combination would be serviceable enough to beat the game with for people who like doing low-level playthroughs, but that it would actually be BETTER than the so-called “best” weapon. I understand he wants to reward people for understanding and mastering the finer details of the system, and I think that’s a great idea. But the way it was outlined means he’ll be doing at the expense of another subset of players–let’s call them the Disgaea class–and it’ll also relegate the “best” weapon to little more than a gold bar or plot item. Maybe I didn’t make it clear enough, but I made my suggestion with the hope that it could make both camps happy.
I suspect that, in practice, the approach you’ve outlined would simply force the folks who go in for system mastery to do a bunch of grinding first. That’s really not ideal – you’re not making both groups happy, you’re making *one* of them happy at the expense of punishing the other.
@David: I’m NOT saying you’re going to have to grind if you want to be competitive. Grinding is bad. I’m just saying that you can’t call it the “best” sword if it can be surpassed by combining X, Y, and Z. And if X, Y, and Z are easier to obtain, not to mention they’ll eventually become common knowledge among the fanbase, the “best” weapon becomes almost pointless to even include. Skill should always trump raw stats, and skill should always be able to compensate for low levels. I agree with all that. But I also think effort should be rewarded. If the same skilled player acquires the “best” sword, it should be better than the creative combination of X, Y, and Z even if it’s only by a little. Otherwise, it’s just a glorified plot trinket.
I don’t agree. There’s no particular reason that grinding must yield larger rewards than system mastery. It’s become conventional to implement reward structures that incentivise grinding over system mastery in multiplayer online games because folks who like to grind are more valuable in terms of long-term subscription revenue (i.e., they’ll spend longer and pay more playing the same content), but a single-player game need not operate under the same constraints. Indeed, single-player games emulating the conventions of MMOs without considering *why* those conventions are set up the way they are is one of the major pitfalls of modern game design.
Because system mastery is easier than grinding thanks to the internet. As soon as one person discovers the uber X-Y-Z combination and posts it, many if not most people will use it without even understanding why it works because it’s the path of the least resistance. Even if you personally won’t look up the information, you can’t deny that a lot of people will. At that point, you have to ask yourself what the purpose of the “best” weapon is now. Is it just meant to activate a quest line? Is it just meant to be sold for a lot of money? If those things are true, is it really the “best” anymore? Unless it’s a weapon with maybe a one-time function to expend all its true power for something, it has little to no use as a weapon. No one who acquires it would ever use it as a weapon, so it might as well be a gem-encrusted gold sword that can’t even be equipped.
I thought I mentioned somewhere that there’s a way to get the “ultimate weapon” extremely early in the game. Probably before you could get the things for the highest DPS you can get the ultimate weapon.
Ah… I guess I didn’t mention that the examples relate to the game.
Also since the storyline and gameplay are going to heavily intertwined, explaining certain strategies may accidentally spoil some of the best reveals in the story. People are usually self-enforcing when it comes to story spoilers.
And within the conversation between Alfilaria and David someone mentioned something that is a trigger for one of the most epic plots that is solely made possible by the in-game actions of the player. This is what I said before about player actions being more predictable than game designers realize.
@Alfilaria: The notion that the only “legitimate” challenges in gaming are those that challenge the player’s patience is a disease upon modern gaming. Sure, if you design your games under the assumption that every single person who plays it will be following an Internet walkthrough by rote, the grind is all that’s left – but why would you inflict that upon yourself and your players?
Well, having the best weapon available off the bat should be interesting XD I guess we’ll just have to wait and see how Alfe makes it work! I’m just a bit utilitarian sometimes–so many developers spend time programming classes, skills, and items that serve little to no purpose and wind up ignored by the playerbase, which I think is sad. So I think a game revolving around true choices should strive to make as many items, equipment, and skills as competitive with each other as possible. If X, Y, and Z are easier to get and do more damage, and we all concede that everyone will eventually start using that combination, the “best” weapon will need to have something going for it gameplay-wise, especially if it’s difficult to get. If it’s not valued for its gameplay usage, it’s just a glorified plot trinket, and it’s pretty annoying when the supposed “best” sword isn’t actually the best. Plus, it violates suspension of disbelief when the Legendary Sword of Plot Awesomeness is worse than the sword you won playing poker at a casino.
@David: You keep putting words in my mouth… I don’t know if you’re doing it on purpose, I’m that bad at explaining, or you really just misunderstand me that badly. Either way, I respect your position even if you probably don’t understand mine, much less give it any credit. So I hope we can just civilly agree to disagree.
http://www.sirlin.net/blog/2009/11/23/migs-brenda-brathwaite.html
This article is interesting as the board game also has some parallels with Fading Hearts and DSTW in how tragedy happens. With player choices that they make completely of their own will being the absolutely key in making the tragedy? It’s going to be so much more powerful as there was a way to stop it.
And I know players will try to break the game balance. There will be stories waiting for them.