Opinion – Resolving the procedurally generated story problem

It’s the little things that count.

A few articles ago I talked about the infinity game and the difficulty in generating compelling story content using an algorithm.  Stories require elements that are difficult to render down into discrete blocks and therefore require much more care and planning to combine than procedurally generated staples like loot or maps.  In this article, I’d like to discuss an existing stepping stone that can help take the load off of story writers seeking procedurally generated content.  Specifically, I’m talking about story infused elements.

Stories develop organically in games outside of the narrative written by the developers.  Even a story barren game like chess inspires amazing tales of clever strategies and narrow defeats.  Terrible games can similarly create developer separated stories, though usually not for any reason developers want them to.  The point is that games create stories outside of the strict confines of the narrative established by the developers.  They do this by providing game elements that players may use to craft tales of interest from.  Players imbue these elements with meaning which they often share with their peers.  This is the entry point through which developers may turn their procedurally generated elements into procedurally generated story elements.  All they need to do is make it easy.

I played a round of Crusader Kings 2 as a cantankerous, militant duke who succeeded in uniting England after a series of bloody battles and rebellions.  At the end of my character’s long reign, I looked to the next generation only to discover that my next in line was a blood thirsty psychopath with zero talent and a number of failed murder attempts on her record.  Under her, the kingdom would surely fragment.  The next in line after the demon child was a brilliant, charming, and incredibly capable woman who was beloved by all.  Should I have my king murder his eldest daughter to let her sister inherit and thereby preserve the kingdom?  Should I step back from killing a child and let her develop unhindered but with the understanding that England would probably fall apart once more?  Such are the stories of Shakespeare and it was mostly generated procedurally.

One of the great things about Crusader Kings 2 is how it imbues gameplay elements with a real sense of narrative and meaning.  Much of what I described (my character’s martial ability, his daughter’s psychopathic nature, her sister’s saintly disposition, etc.) are all numerical elements of the game combined via an algorithm to produce a variety of scenarios.  Crusader Kings 2’s genius is describing these elements in such a way that they may combine to form an intricate story without the developer having to write one.  CK2 describes its procedural generation mechanics in such a way as to create a structure which the player can fill out with their own narrative.  CK2 never told me that the king in my game was contemplating murder, but it gave me all the elements upon which I could hang that tale.

The ultimate goal of procedurally generated stories is to make it possible for games to invent complex narratives without the player’s inputs.  Understanding that developers aren’t there yet, the infusion of gameplay elements with meaning brings in the player and helps reduce the load on the procedural content in crafting interesting tales.

Mod Dev Diary 1 – Things I learned

This is a teachable moment.

I’ve never made a mod before and I never had any plans to.  I see my role in the gaming ecosystem as a consumer, rather than a creator, yet the reality is that I have embarked on making a mod for the greatest game of all time, Crusader Kings 2.  My goal is simple: make a new religion with gods that have a real impact on the game world.  Whereas the existing game focuses on the political and personal realities of gods, I want to make gods that destroy kingdoms, cast spells, and generally wreck the world.  I’ve had a lot of fun so far, but there are definitely a few things I’ve learned about game development along the way.  Here are some of the biggest.

Planning is key

To start my modding career, I poked around the game files and tried to create smaller mods.  As I gained knowledge, I worked on elements that I figured would be in my religion mod.  It didn’t take long to realize how vital planning was to the whole endeavor.  A mod of any complexity will have interlocking parts that will effect each other and the broader game state.  It’s easy enough to deconflict any couple of elements, but creating a set of interlocking pieces is extremely difficult without a plan.  This became very apparent to me as I was trying to create the levels of status within the religion.  I created a mockup of each status and even wrote the events and leveling conditions that worked the player through each level.  As I considered the leveling scheme, I immediately thought of what it would take to go through as an acolyte of one of the gods.  When I started on the next god, my leveling scheme conflicted with the first.  Inevitably, the leveling scheme of the third god conflicted with the first time.  Planning is definitely needed.

You can’t power through

My work habits largely derive from my academic career.  Like many, I became a master of plowing through complex lines of thought in large chunks of time, usually at the last possible moment.  That doesn’t work anymore.  It’s not that I’ve changed, my brain is still the delightful caffeine addled machine it’s always been, but the type of thinking required benefits from space.  Coding the mod requires intricate work where even a single incorrect line will derail the whole operation.  Detailed thinking, not broader lines of thought, are what is needed.  I’ve literally spent hours staring at the screen looking for an errant line of code and been totaling unable to identify it.  Alternatively, if I walk away for a day and come back, the answer is obvious.  This takes time.

Fantasy names are hard

Like, super hard.  It’s frighteningly easy to come up with extremely stupid names and incredibly difficult to come up with smart ones.  I’ve elected to work off of Latin terminology, but, even then, I’ve discarded a ton of names that make me embarrassed of my own creation.  I can see the temptation to create nonsense works or just use the English word.  I ultimately had to change my goal from creating fun and memorable god names to not terrible god names.  The goal is to avoid breaking the player’s immersion rather than to elevate my game in any meaningful way.  That being said, Furya and the Elemental races is still dumb.  Sorry, Vin Diesel.

Opinion – The State in Games

You are what you program.

Whether we like it or not, every video game reflects an opinion of the world.  Every aspect of a game is a hint about how the developer views a particular topic.  To be sure, sometimes elements of realism are sacrificed on the altar of gameplay, but even that is a reflection of the developers mind.  Realism is sacrificed, but elements the developer deems core to know that a thing is a thing are preserved so that the audience can identify whatever item or concept the developer is trying to convey.  Case and point, guns.  Guns in an FPS never jam, don’t have weight, and have magnetic ammunition that leaps into your pocket when you walk over it.  These unrealistic aspect of guns are included because they are deemed inessential to the idea of a gun.  Today I’ll be looking at the idea of a state.  Many strategy games model the state, but they rarely reflect a state as it actually is.  Many have a different idea of how states function as compared to their real counterparts.  Here are a few:

Civilization V – Centralized dictatorship

Putting aside the immortal sovereign, the Civilization games function as a dehumanizing centralized dictatorship focused on control of major cities.  Decision making is done entirely by the leader without much input from independent interests or other government agencies.  The only part of an empire that actually has a voice is the population based on generalized notions of happiness.  As a result, very little gets done without a direct command.  Everything from basic public buildings to countryside infrastructure is dependent on the player making the order and releasing the resources.  Even supposed changes in government rarely restrict the leader from doing what they want to do (I do recall that democracies could prevent the player from declaring war in Civ2.  Ahh, the glory days.).  Of course, most real governments have interests other than a generalized notion of “the people”.  These interests play an important role in prioritizing resources and are largely ignored by the Civilization series.  Especially in dictatorships, the ruler must appease these interests or face serious problems down the line.

Rome Total War 2 – Bureaucratized Oligarchy

The Total War series attempts to model various interests in the empire.  The Roman Senate plays a role in determining the support a given leader has and special characters, such as generals, each have their own traits and ambitions.  The opinions of the people matters, but, like the Civilization series, only insofar as they cause trouble.   If the people are happy, then the player must respond to a small group of specialized interests.  To do so, the play has access to an efficient and effective bureaucracy that is able to maintain control over a massive empire.  Orders from the center are reflected quickly in the priorities of the provinces with only a corruption increase to show that the bureaucracy has interests of its own.  Modern bureaucracies are only half as effective as the Rome Total War 2 ones.  The idea that a pre-industrial society could have detailed control over such a massive amount of territory is ludicrous and likely reflects the developer’s bias towards modern governance.  Like the Civilization series, Rome Total War 2 also doesn’t model the institutions that would inevitably develop to handle such a difficult task.

Crusader Kings 2 – Legalistic Feudalism

The last game on our list is the most realistic attempt to model an actual government.  Rather than rely on a faceless, obedient, and super capable bureaucracy, Crusader Kings 2 creates a government and society built on individuals with their own interests.  Individuals have titles which reflect their area of control and the individuals they rule.  Those individuals will negotiate with their superiors, fight wars for them, and rebel.  Without direct input, NPCs will develop their own territories and conquer new lands.  It’s a fascinating system, though not without its flaws.  The most notable is the high degree of legalism inherent in the system.  While medieval lords were a far more legalistic bunch than they traditionally get credit for, they didn’t stop crushing victories against their infidel foes because they promised they’d only crusade in a certain area.  Furthermore, borders of provinces shifted all the time whereas Crusader Kings 2 acts as if they were set in stone by Charlemagne.

There are plenty of other things to nitpick.  All three games (and all strategy games that I am aware of) assume an instantaneous response to all orders regardless of distance.  Every leader has a crystal clear view of all resources they possess and the capability to move them wherever they need to be.  I could go on.  Still, these feel like conceits to gameplay and less like a world view.  The governments outlined here seem reflective of a viewpoint, or a blind spot, of developers who didn’t quite understanding what they were saying when they made their game.

Failing where no man has failed before.

I suck at Minecraft.  This is theoretically impossible as Minecraft has few limits and almost no goals beyond not dying yet I suck at it all the same.  I can survive so that rules out the last bit.  The reason I suck at Minecraft speaks to the core of the motivation behind playing Minecraft.  Whereas most games have a distinct goal (kill this dude, conquer this thing, etc), Minecraft is notable for being without one.  As a player, I’m motivated by the goals provided for me and have trouble making them for myself.  I suck at Minecraft because it demands the player decide their own direction.

Well, it’s not that dramatic.

Most games I play have a direction.  This helps guide my play and give me ideas on how to get the most from a game.  I know what the game wants from me, and I just have to figure out how to get there.  The directed experience is nice, because it gives the player a clear goal and allows for greater expansion of story elements that may require the player engage in a specific activity.  Elements I like in games, such as narrative and character building, are much easier under the direct control of a developer.  Of course, directed play has its downsides.  Players who wish to engage in other activities don’t have an option.  This can be particularly frustrating when there are parts of the game they like and parts they hate.  Then the player must go through the annoying bits to get to the gameplay they actually enjoy.  Another issue is the lack of emergent gameplay.  Some of the best story and gameplay comes from the interaction of game systems.  As those systems are restricted, so too is the chance that they will combine into something unique and interesting.

Minecraft takes the opposite approach.  It dumps a ton of tools in the player’s lap and says “You figure it out.”  Player set goals are the only goals which can become exhausting at times.  The true sandbox games wants the player to effectively create their own fun while providing very little if the player loses motivation.  The game effectively quits when the player no longer wants to invest the time and energy into creating things from its tools.  Of course, that also means a motivated player has considerable freedom to do as they wish.  If the player has a goal, or is good at creating them, then the open sandbox is an ideal way to create their own fun.  With tons of tools and plenty of potential projects, a solid sandbox game can provide near limitless play.  As long as the motivation is there, so is the fun.

The best games are the ones that include both sandbox and directed elements.  This gives the player things to work on while also leaving them the option of branching out and accomplishing something strange and interesting.  This gives the player a task to accomplish when they want something more directed and the opportunity to do something different when that doesn’t appeal.  In my personal favorite game, Crusader Kings 2, the player has the opportunity to do just that.  The game never explicitly sets a goal, but rather provides tons of clear goals for the player to choose from.  The player can do something more directed, such as conquering a kingdom, or freeform, such as converting the Holy Roman Empire to Hinduism.  It’s up to the player to decide what experience they want, but both options are there.

Opinion – Wrapping up 2014

Arbitrary rankings for goodness!

2014 has been a bad year for video games.  From the controversy surrounding the Gamergate issue to the rash of overhyped games that failed to deliver, it’s hard to see this year as anything but a failure.  Still, there have been some successes.  In this article, I’ll go through my most disappointing game, but I’ll also highlight three games that show what the year could have been.  Let’s hope 2015 is better.

Most Disappointing Game – Civilization: Beyond Earth

This was a crowded field.  From the disappointing games I played like Borderlands: The Pre-Sequel and Farcry 4 to the others I mostly heard about like Destiny or Assassin’s Creed: Unity, 2014 was chock full of over promises and underdelivering.  Even then, one game stood below the rest.  Civilization: Beyond Earth.  I initially met the announcement of a Firaxis developed space Civilization game with a sense of joy and excitement.  We hadn’t seen a proper successor to Alpha Centauri ever and the premier turn based strategy development house was taking it on!  Unfortunately, the final product proved to be an extension of Civilization’s flaws rather than a proper development of the concept.  Firaxis failed to infuse the game with the story or the strategic depth of its spiritual predecessor.  Even without the comparison, Civilization: Beyond Earth was just a soulless game without much to recommend it.  There were certainly worse games this year, but few failed to live up to their potential like this game.

Best Surprise – Hearthstone

Blizzard Entertainment is one of the most public and successful developers, so it’s hard to say that anything they do is a real surprise.  That being said, a deep, complex collectable card game is incredibly difficult to make and Blizzard had no background in it.  Even without the experience, Blizzard created a fantastic experience in Hearthstone that only the beta players really saw coming.  From the variety of strategies to the slick interface, Hearthstone is the digital successor to Magic: The Gathering.  To be sure, the game has a ways to go.  Blizzard has a great deal of design space to explore, I’m not convinced it has a grasp on how to build cards for arena, and it doesn’t seem to have a clear strategy for rolling out new cards.  That being said, the base game is fantastic and accessible.  We now know that Blizzard has the chops to put together a compelling card game.  The question is whether it has the ability to maintain it.

Best Updates – Crusader Kings 2: Rajas of India, Charlemagne, and Way of Life

Any who have played Crusader Kings 2, know of the game’s ridiculous depth.  Even the vanilla version allows a player to control thousands of Christian leaders across almost 400 years of history.  Had developer Paradox walked away from their 2012 hit, it would have remained a great game.  What makes it truly one of the best is Paradox’s continued commitment to updating and improving CK2.  This year, we saw releases that expanded the world to India, introduced a story line around the Carolingian kings, and created an RPG-lite system for character development.  Paradox made an already deep game even deeper this year and shows no signs of slowing down.  If you haven’t had a chance to play this game, overcome its (admittedly, vicious) learning curve and dive into one of the best strategy titles available.

Game of the Year – Dragon Age: Inquisition

It’s telling that my best game of the year is also a heavily flawed one.  DA:I suffers from a number of bugs, odd pacing issues, and a generally uninteresting cast.  Even with its problems, it still provide the most compelling experience of the year.  DA:I is effectively two games.  It combines open world exploration with a dedicated scripted story line with the challenging, complex choices that we’ve come to expect from the Dragon Age series.  More than any game of the year and most games made, DA:I provides a clear sense of the world the player is inhabiting and the people who live in it.  The history, culture, and society are all center stage and intricately woven into the gameplay.  It’s telling that, as much as the first Dragon Age frustrates me, DA:I has tempted me to go back, endure that frustration, and truly make the story my own.