Maybe Mundane

Communication is a big part of Ruins. It is the core mechanic behind the game’s initial concept and is integral to the gameplay and experience. Because of this, much of my design work has gone into planning the dialog system and devising a way for the player to interact with Susan that feels natural, rather than game-y. One way I have decided to do this is by abandoning the traditional dialog tree for a social artificial intelligence.

I first decided on the use of social AI while reading about Comme il Faut, a system that simulates social games based on the psychological needs and roles of the game’s actors. Because the only dialog that will occur in Ruins is between the player and an NPC, I didn’t think such a complex system was necessary. But it made me realize that if the player could have multiple responses to a single line of dialog based on their current state, why shouldn’t an NPC?

So I started working on a system that allows the AI to choose different responses to the player’s dialog based not only on the line that is being responded to, but also previous conversations and actions the player has taken. This system would be similar to how in RPGs such as Dragon Age: Origins NPCs can say different things at the start of a conversation based on previous events (love interests opening up to you once you’ve reached high enough approval, Alistair’s angry outburst should you fail to save a certain Arl’s family, etc.). The big difference is that using Ruins’s system, the NPC would have the opportunity to decide what to say at every step in the conversation, not just at that opening line. This can lead to much more natural conversations during which everything the player has done previously is made relevant. In others words, the dialog builds on itself similar to the way it would in real human interaction.

Although I could probably write an entire essay on the system itself, I would like to dedicate this post to an explanation of the heart of the system: the decision algorithm. Susan’s decisions will be based on three variables that the game tracks behind the scenes. These three variables are Trust, Romance, and Friendship.

Trust is the primary variable upon which dialog decisions are made. It sits on a scale of 0 to 100, starting at 50. Nearly every dialog choice and major action will affect Trust in some way, although as its value increases/decreases toward one end, the impact of changes will start to lessen. The more Trust the player has earned, the more options will be presented to them as Susan opens up and reveals more. However, having a low Trust score won’t necessarily eliminate those options, although it will make it more difficult to gather information from Susan.

Romance and Friendship are percentage modifiers to Trust and both operate similarly. Only certain dialog options will affect them, but they will open up new dialog paths that couldn’t be attained by building Trust alone and can have a large impact on what endings are available to the player. With a high Romance or Friendship score, Susan will be more invested in the relationship with the player and more forthcoming with information.

So how does this all fit together? The system requires each line of dialog from an NPC to set a minimum threshold for Trust, Romance, and Friendship. If the player has cleared the threshold on all counts, I take the difference between the actual and minimum values and subtract that from the maximum value of each variable (100 for Trust and 1 for Romance/Friendship). Then I return the calculated value for Trust plus the multiplication of Trust and each modifier. This is demonstrated in the code sample below:

if(Trust >= npcDialog.MinTrust
    && Romance >= npcDialog.MinRomance
    && Friendship >= npcDialog.MinFriendship)
{
    short dTrust = 100 - (Trust - npcDialog.MinTrust);
    float dRomance = 1 - (Romance - npcDialog.MinRomance);
    float dFriendship = 1 - (Friendship - npcDialog.MinFriendship);

    return dTrust * (1 + dRomance + dFriendship);
}

These values will be calculated for each available dialog option and the greatest value to be returned will determine what Susan decides to go with (in the event of a tie I currently planning on selecting her response randomly out of the winners). Going along with the theme of organic game progression that I’ve mentioned earlier, the variables the algorithm is based on will be hidden from the player, who will have to depend on the context of the dialog to navigate their relationship with Susan. This is done to pull the player’s focus away from keeping track of scores and instead draw them to the dialog itself. I believe that doing so will discourage the player from trying to “game” the system and encourage more natural gameplay.

Although the algorithm presented is simple, I have found it provides the most desirable results out of all the ones I have tried, and the three variables it depends upon work well with content generation. Although I have yet to iron out all the wrinkles in the dialog system, I am confident in the foundation I have laid with the decision algorithm and supporting class library. As the central mechanic, the quality and presentation of Ruins’s dialog will set the precedent for the rest of the gameplay experience. With that in mind, I believe I am off to an excellent start.

In my last post, I wrote about how I want more games to take an organic approach to story progression, allowing the player to dictate how the plot unfolds through gameplay, rather than multiple choice decisions. I envisioned a game where, once a “mission” is initially assigned, the game takes direction from the player (rather than having it be the other way around) and responds appropriately to how the player interacts with his or her environment based on the constraints of the game. Ruins is the product of that vision and the primary focus of this blog (save any tangents I decide to take off on).

So what is Ruins? I’ve often floundered when attempting to explain it to people, and my descriptions have ranged from “a combination of Bastion, Metal Gear Solid, Zelda, and Dragon Age…but not really” to “a character-driven game that subverts how players receive instructions.” However, I think the times that I was best able to communicate what I want Ruins to be were when I focused less on the details of the mechanics and simple related the story I want it to tell. Because at its heart Ruins is, above all else, a story.

The idea has been germinating in my head for nearly a year and a half now, ever since I played Dragon Age: Origins. Originally, my only goal was to create a game that centered around the player’s relationship with a character, to refine the experience I’d had while interacting with party members in BioWare’s heroic fantasy. I wanted a game with characters so interesting and compelling that you want to just sit there and talk to them—and it would let you. At the same time, I wanted to tell a story that features consequences the player can care about and places the protagonist in a situation that doesn’t have an easy way out. Over time, I found ways to tie these elements together and have come up with a concept that I believe will provide an entertaining, meaningful experience for its players.

You play the game as an agent of a secretive government branch that specializes in xenological and otherwise superhuman sciences. You’re sent to assassinate Susan Bishop—a young woman who has mysteriously developed psychic abilities and is being used by the military for research that goes against your agency’s interests. Otherwise, you’re not told much, but by exploring your environment you can find clues that will help you piece together the context of your mission. Oh, and that girl you’re sent to kill? She’s your guide. But rather than telling you where to go and what to do—oftentimes, she knows even less about your situation than you do—she talks to you, helping you sort through puzzles and sharing information about the world as seen through her eyes. Because of this interaction rather than her role as your target, she is arguably the most important character in the game world.

How her relationship with the protagonist progresses is up to the player and has an impact on the outcome of the game, but there’s no HUD score or meter to tell you how you’re doing. Instead, you have to rely solely on dialog to determine where you stand, although your actions outside of dialog can also impact her opinion of you. In a similar vein, the game won’t hold you to your word. Just because you say you’re going to do something doesn’t mean you have to when it comes to it, and you’re given room to make decisions up until you carry out an action. In short, the game can be lied to—and it can lie to the player in turn.

Of course, this is all part of the greater goal of producing a game that allows the player to progress through and interact with it organically. I don’t expect to realize this goal completely—after all, as of now I’m the only developer working on this project and am doing so alongside both school and work. But I think it’s a step in the right direction.

Someday we will have games that don’t feel like games—at least, not in the traditional sense. Players won’t have to worry about silly things such as scores or quest completion or winning. I’m certain any players who want to keep track of such things will still have ample opportunities to do so, but I don’t think that’s where the industry is headed. At the very least, it’s not what Ruins is for. Ruins is for enjoying the experience and maybe going back and playing again to see how things could have unfolded differently. And if, while playing, the player feels at any moment that the game is contrived, or forcing them into an outcome they would otherwise have prevented, then I won’t have done my job right. But that’s getting ahead of myself. For now, I’m just focusing on telling a story the best way I know how. We’ll see what comes of it.

…As a side note, there is another indie game titled Ruins that you can download for free here. It’s a cool experience that I highly recommend you check out. While it hasn’t gotten enough attention for me to feel I must instantly change the name of my project (which I unfortunately became attached to before I found out about Cardboard Computer’s creation), I am currently reconsidering and would appreciate any suggestions for a title that doesn’t steal the thunder of another artist’s material. Thanks!

I love video games. I love any medium with the potential to give its audience an experience that helps them discover a piece of the nature of the universe, humanity, or themselves, whether it be music, film, or literature. But mostly, I love video games. Because unlike any of its predecessors, this medium in particular takes its audience out of their passive role and lets them experience first-hand whatever it is the creator wishes to express. Games have so much to offer the world, whether it be through interactive storytelling, education, or even “gamefication”, the ugly word used to describe applying traditional risk-reward game design techniques to solve real-world problems. It’s because I love games so much that I’m so eager to see them change—to grow into what I know they could be.

How did we get Skyrim from this?

Game design is still a relatively new field of study. PONG, one of the first video games, was released in 1972, and it was nothing more than a digital version of table tennis. Over time, however, game design began to become more complicated. Games moved away from competitive score-based applications and were given endings—they could be “beaten”. A clear set of expectations began to manifest between the player and developer: a player was given a set number of lives, or maybe a health bar, and if they depleted these resources they were faced with the consequence of starting over, first from the beginning and then from progressive save locations (or in rare cases, continuous saving with in-game consequences). Story presentation in games has gone from being nonexistent, to being told through cinematic cutscenes, to being chosen by the player through a nonlinear narrative. Meanwhile, that very first video game has been programmed by novice developers around the world a thousand times over.

So why does all this matter? If I’m looking to the future, why am I choosing to preface my point with a (very) brief overview of video games’ 40-year-old past? It’s because the short-hand video games are currently using to communicate between designer and player—the health bar, the save points, the quest log—are much the same as the mechanics that were first invented in the 80’s—polished, perhaps, but basically similar. And if we want the art of game design to achieve its full potential, we’re gonna have to start trying something new.

I’m not afraid of the dark. It’s when I have a light that I know shit’s going down.

Alan Wake is a prime example of modern linear game design. The entire game, I simply ran from point A to point B (with the occasional sidetracking to locate stray manuscript pieces) and it was fun. I rarely questioned where to go, which allowed me to concentrate on blasting the Taken and figuring out what happened to Wake’s wife. But for all its strengths, the age of its design decisions shone through brighter than a heavy duty lantern. Why? Well, for starters, the risk presented by the game was too predictable. In my 9 straight hours of gameplay (I played the game from episode 2 on during a 24-hour marathon), the times that Wake was most vulnerable, wandering around in the dark without a light or a firearm, were the times that I allowed myself to relax the most, because I knew that nothing was going to come out of the shadows and attack me. While one mission subverted this by denying you your handgun and requiring you to to run past enemies using only a flashlight, the game expressly told you this was going to happen before putting you in danger, and afterward it reverted back to the old formula like a child who gave up on art because her teacher scolded her for not coloring within the lines.

Granted, I plan to spend my life thinking about this stuff, so perhaps I’m just picking at subtleties that no one really notices. But I guarantee that if you paid attention, you would notice the cues dropped by game designers that directly impact the way you play. After all, what do you do as soon as you pick up a new weapon in Metroid? And when looting the treasure chest outside the entrance to the dungeon’s boss chamber, what do you expect to find? Most gamers I know would get pretty pissed if they found a worn tunic in place of their anticipated potions, and you know Nintendo’s not gonna give you a shiny new gun without bringing something to shoot at along with it. This is all part of the designer-player relationship I talked about earlier, and while these gameplay elements have clearly worked well enough to stand the test of time, they make games horridly predictable. You may not think of that as a bad thing, though, in which case I bet you’re the kind of person who likes figuring out the culprit in the first two minutes of Law & Order and then waiting until the end to see you were right. (I’ll give you a hint: it’s usually the seemingly unimportant person the detectives talk to at the beginning of their investigation.)

I wonder what I get to use this for?

Wouldn’t it be better if more games progressed organically? If they simply handed you the pieces and then you decided what to do with them? There are some that do—or try to, at least. Modern RPG franchises such as Mass Effect and The Elder Scrolls have made a name for themselves by featuring gameplay centered around choice—you decide where you go, who you talk to, and what quests you accept. But even they have their flaws. As much as I love the series, Mass Effect rejects organic game design by pigeon-holing the player’s major decisions into two categories, Paragon or Renegade, and limiting the content they have access to based on the points they rack up along this either-or moral binary. And even in Oblivion, where you can experience 300 hours of content without ever completing the main quest, you have to complete a preordained checklist of events before you can say you’ve beaten the game.

While these techniques may work for video games now, I believe we’re going to start seeing a shift in game design practices, where instead of telling players what to do, games let us take the lead—and then respond to whatever it is we’ve done. We’re already seeing this trend in how BioShock Infinite plans to run scripted events based on the player’s actions, or Skyrim’s dynamic quest generation. We’re on the right path; we just need to keep taking risks and not be afraid to leave some of the tried-and-true formulas behind in exchange for better, more innovative methods. And as a gaming community, we need to not lose faith in developers the minute they try something new and fail. After all, if we didn’t make mistakes, how would we learn from them?

We would be complaining just as much if they had decided not to make any changes to Origins at all.

Of course, I didn’t start this blog just so I could spout off my ideas to the internet and then sit back and wait for someone else to do something with them. So consider this an introduction to a series of posts that will track my own endeavor to progress the art of game design in the form of a game I like to call (and will continue to do so until someone gives me a better suggestion) Ruins. I’d tell you what it’s about but… well… that will probably take up an entirely new post. For now, just know that it will try to explore organic game progression and feature a killer cast of characters. Hopefully that’s enough to keep you coming back for more.