Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
November 1, 2014
arrowPress Releases
November 1, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Games have the potential to be infinite, so why are you still working on a throwaway project?
by Anton Temba on 04/08/13 10:15:00 am   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


The never-ending cycle of waiting for the next great thing.

Lets get this straight. Books, images, music and video are a single-use item. Once you've seen them, read them or listened to them, the discoveries has been made and you move on to something else.

Sure, you may revisit them for a second or even a third go, but by that time, they're already old. You know what happens, you know what it looks likes and how it feels. Its done.

A throwaway product.

Before I continue, I do infact recognize that its a throw-away product for a single person. It may have as many uses as there are the amount of people, which are born every day, making its potential lifespan infinite in that sense.

But thats not the point. Lets get a bit selfish here.

This article about you as a consumer, not the whole world.
Picture this the following:

So you've experienced some neat scripted game and its now over. It was fucking awesome.

You want more.

There is none at the moment.

Lets also mention that making games is a hard and slow process.


That sucks, doesn't it?

You're back to square one. You experienced something inspiring, but it ended and thats it. You wanted more, but the experience was cut off as soon as the content ended. Now you're left stranded to look for something else to enjoy.

Whats worse, is that it may take a long while before you get something that good again, but if its the same limited design, you'll consume it in a few hours like the last game and be back in this same state of emptiness.

Waiting till the next good game comes around.

Its an endless cycle. So what's the point of continuing this?

However, there is another way.


Something a computer program can that a movie can't.

Video games have something that movies, music and books don't. The potential to be infinite.

When I say infinite games, I don't mean those simple highscore-based arcade games, no.
I'm talking about games with the same seriousness and scale as those massive scripted ten plus hour adventures like Mass Effect or Skyrim, but much better and truly infinite.

If you're freaking out and saying this is impossible, then stop. It is. The main difference between those above examples and the infinite game, is that you play as yourself and not forcibly being someone you're really not.

From the perspective of gameplay design, it works much more fluently, is more consistent and doesn't require the artifical constraints you get often get in a scripted game. It just works.

Most of the time, unless your main goal is to specifically teach something or tell a story, which movies, visual novels and books are more fluent at, these games are meant to give you a specific experience. A thrill, an emotion or some particular feeling. An atmosphere.

The trick is to focus on that experience and build around that, rather than stuffing the player into a some arbitrary role, when the real goal of the game is to simply give an experience, like being a pilot, monster, soldier, hunter, spy, hero, adventurer, commander, god... you name it.

The whole aspect of putting a player into a specific character's role only shifts the focus from what is really important, which is about delivering an experience, not being someone you're not.


Were not really making the games we really want, are we now?

But alas, the games industry, both AAA and indies, seem oblivious to all this and keep doing the same thing over and over, despite deep down really wanting that infinite game. Talk about the definition of insanity, eh?

Whats mind boggling, is that developers spend years of their life to create games, only for them to have such a limited - and often rather shoe-horned - design that can be exhausted within a few hours, some times even in just mere minutes.

This is understandable for movies and books, since they're not interactive by nature and cannot generate fresh new content like a computer program is capable of doing, either through multiplayer, user content or procedural random generation.

But come on, whats with the obsession of shoe-horning the scripted movie framework on video games?

This is especially the case when the player is put in the role of a specific character, usually with the hopes that he or she will experience the story from a more closer angle.

There is a huge problem with this design. A big conflict.


Fitting a square peg in a round hole.

You put the player in full or even partial control over a character that has a previously established past, abilities and perhaps a personality of some description.

The problem is that the player can do anything he wants. Any story or personality the character has or the path that he takes will get perverted in one way or another.

Its no mere coincidence players are having fun doing absurd things that the character they play as would never ever do. Like tea-bagging corpses. Or going AFK during a conversation with an NPC in the middle of something important.

Typical player behaviour in a video game universe.

Why? Because you simply can. As the player you're in control, after all.

A game is an interactive system and its natural for this behavior to occur, so your attempt to tell a convincing scripted story and creating strong immersion in this way goes right out the window.

And yes, an obvious thing to do against this is to limit what the player can do, to the point that the pretty much runs on rails, but at that point your game has turned into a glorified movie.

Doing so cripples the game's ability to be infinite and it greatly limits its flexibility in terms of gameplay and what the player can experience within it. Thats just contradictory game design right there.

Trying to force a storytelling approach to a video game while giving the direct control over a character to the player is as shoe-horned as it can get. The more serious a game tries to take itself, the more cringe worthy it becomes.

So you're basically shooting yourself in the foot with a design that adopts the approach of movie and books.


So, back to the point of infinite games.

Like mentioned previously, there are three big things that make the whole infinite game idea possible; multiplayer, procedural random generation and user content. In addition, there are two more; persistency and emergent behaviour.

Using these five concepts, either all of them or even just a few of them, it is possible to create a video game that never gets old or lame unlike a movie, book or those traditional scripted video games do inevitably.

With all that said, I return to my initial question:

Why would you spend years creating something so complex as a video game with the goal to deliver an experience that limits itself with a scripted, finite design?

All that hard work you put in would get used up in mere hours, where instead you could create a game with a design that is infinite?

Not only you could create something that will have virtually infinite longevity for yourself to have fun with endlessly, but also provide a lasting product to your players that offers them same infinite entertainment, with much wider flexibility on how it can be played.

To infinity and beyond!

Talk about some seriously huge value for their money. This is what video games were meant for. This is what they're capable of.

Besides that, people are starved for infinite games, you can see signs of this in any hardcore gaming communities when you ask them about it. Theres a huge opportunity to be had here for success and a new form of games to enjoy.

Go out there and design your games with infinite gameplay in mind and a strong focus on the atmosphere and emotion its supposed to evoke in you. Its all a matter of design.

Related Jobs

Rensselaer Polytechnic Institute
Rensselaer Polytechnic Institute — Troy, New York, United States

Assistant Professor in Music and Media
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Next Games
Next Games — Helsinki, Finland

Senior Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer


Arseniy Shved
profile image
It was interesting to read this post.
Having said that, maybe it's just my philosophy, but I prefer finite games to infinite. Or finite anything, basically.

It's like this Kōan:
Two traveling monks reached a river where they met a young woman. Wary of the current, she asked if they could carry her across. One of the monks hesitated, but the other quickly picked her up onto his shoulders, transported her across the water, and put her down on the other bank. She thanked him and departed.
As the monks continued on their way, the one was brooding and preoccupied. Unable to hold his silence, he spoke out.
"Brother, our spiritual training teaches us to avoid any contact with women, but you picked that one up on your shoulders and carried her!"
"Brother," the second monk replied, "I set her down on the other side, while you are still carrying her."

When something is in the past, it does not mean that it has ended. I like to think about the game after the end credits. If said game gives me the reason to do so.
The original point of this Kōan was totally different, but it seemed like a good analogy ^__^.

I tend to think that something outstanding cannot be infinite - it immediately becomes usual.
Moreover, sadly, multiplayer (sorry BF3) and user content do not guarantee endless outstanding experience. Procedural random generation can be infinite, but to maintain the development of the experience (we do not want to be stuck on one thing for eternity, right?) new laws are to be added to said generation.

Anyway, thanks for this post - every time, when someone mentions eternal stuff I tend to spend many hours thinking about... basically everything, and it's been quite some time since I've done so. So thank you!

Anton Temba
profile image
I see what you mean. There is actually a way to combine that memorable experience from a finite adventure within an infinite game.

The basic idea it works the same way life works. You live in a persistent universe by living a daily life doing daily things, but then there are also major events that shake the world and give an adventure.

A good example in real life is a school year, a camping trip, vacation travel, work week, a local disaster (volcano, tornado, flood, storm, accident...), a mundane grocery shopping trip and so on.

Life is a persistent thing that keep going, meanwhile the examples above could be considered as individual, finite adventures of varying lenghts and importance that you take on within this universe. As they are finite, you get that instance of an memorable adventure, while you're in an infinite universe.

So my point is, an infinite game is made up from infinite amount of single, memorable adventures that have a finite goal. Thats how the framework works.

As for new content and the way to have it provide an interesting experience ultimately boils down to the robustness of game's framework.

The key is to have everything being interconnected, rich in detail and cross-compatible.

When you've got a content system that is very rich in detail, you can add a single new piece of content, but combined with all the existing content already in the game, you may have thousands of new possibilities, thanks to how the system allows every piece of content to interact with each other in unlimited ways.

An example of this: You have a caveman engineer game and you build structures and machines out of elements around you. Then you add a new piece of content: The wheel.

Now the game takes a completely new turn. The applications for a wheel are immense. You can now build cars, windmills, unicycles, looms and anything else where a wheel's amazing spinning properties may be used.

Miroslav Martinovic
profile image
I prefer combination of both. For example, Dishonored. Great story and campaign, but since Dunwall City Trials came out, I've been playing almost nothing else than Back Alley Brawl. I'd love the game to have a mode where you can just jump in to any of the levels and play a similar mode, where you just run around and fight infinitely spawning enemies. Or at least a mode where you could skip all the cutscenes, including the "loading" ones where Samuel rides you to the mission, etc.

RTSes are great in this, they've got campaigns (though not really good ones usually), and skirmishes.


I think the main incentive to make "finite" games is, firstly, game devs nowadays work with and think about the medium too much like "interactive movies", and secondly, finite content gets "consumed", so it makes space for another content to be produced and sold. With DLCs, the tendency to plan for this will probably get only stronger, unfortunately.

A side note to score attack: if your mechanics are fun enough, score attack is everything you need, despite, offcourse, not being everything you COULD have.

Michael Joseph
profile image
but how will they sell sequels and DLC?

There is the occasional Spore but there are more indie games working on the types of procedural universes you describe. Off the top of my head

Shores of Hazeron
Infiniverse (which I think was made during a 7 day rouge-like competition)
Ascii Sector

Perhaps one of the lessons here is that abstract visuals makes it TONS easier to program these games. And perhaps that is a due to a combination of factors namely- less work on graphics means more work gets done on the underlying system. Many games spend so much time on getting the visuals of their procedural creations to be on par with non procedural games that they never get around to working on the rest of the systems which are supposed to define the game.

Also the current reality is the people who play these diamonds in the rough are a relatively small amount. But perhaps that would change if more investment was being made in developing and promoting them. I still believe that "fun is fun" so if you can find something that is fun but maybe not packaged to be attractive to the mainstream, then "all you have to do" (TM) is work on that packaging and polish.

Anton Temba
profile image
"but how will they sell sequels and DLC?"

Same way they do now. Adding new content to an infinite game can DLC, meanwhile a sequel is either an improved game or just different. Infinite game does not mean ultimate game. You can't design a system that works suitably for every type of gameplay.

Robert Boyd
profile image
I'm not a fan of multiplayer & procedural random generation and user content tend to result in much lower quality content than if that content was hand crafted by a skilled designer.

But hey, if someone manages to make a game that's infinitely replayable while still maintaining the quality of games like Dark Souls & Persona 4 Golden, I'm there.

Michael Joseph
profile image
one could just as easily say
"If someone manages to make a game that is visually stunning while still maintaining the quality of games like Dwarf Fortress, Infiniverse and Shores of Hazeron, I'm there."

I only point out the obvious because saying it this way suggests a shifting of the blame and burden to the deep pocket folks who are making games like Bioshock Infinite (which is certainly not infinite.. it's like 16 hours)

Anton Temba
profile image
The quality of content depends entirely on how everything is set up. Where to actually use procedural content, where not to use it, how to establish standards for user content to follow and teaching them how to do it...

Its all a matter of design.

Jacob Germany
profile image
The first games I think of, outside of the indie space (or at all) when I think of procedural generation and user content are Civilization and Spore, respectively. The former creates a replayable experience better than the vast majority of scripted, one-shot games. The latter creates far more engaging silent characters than the vast majority of other silent protagonists one could name.

I am actually at a loss as to where random generation and user content lead to such a drop in quality.

Rob Graeber
profile image
Can't you have both? Like a lot of AAA games, a high quality hand-crafted single player and a multiplayer mode or something else to give the game some replayability.

Chris OKeefe
profile image
There are of course different opinions and preferences, but I think it's premature to say that procedural generation and user content is of a 'much lower quality.'

A few examples;

Skyrim. I think that, as long as you are diligent about the user-generated content you use, the game can be vastly and objectively improved over the original design through the use of mods. Many of the mods are of a quality that is comparable (and sometimes better than) the original content. The fact is that although the designers of a game are extremely talented, they do not hold a monopoly on talent and they have a limited amount of man-hours to dedicate to providing content and features. Also, they cannot always provide the best possible assets because it is in their interest to make the game scale well on different setups. Someone with a very good computer can (and do) run higher resolution textures, higher numbers of assets in a scene, higher polygon counts, etc. Objectively superior, if your computer can handle them.

Minecraft. I think Minecraft proved a few things that those of us who grew up playing with Lego and simplistic games kind of already knew deep down. Exploration and creativity go hand in hand, and even a world made out of little cubes can be beautiful and wondrous and mysterious and dangerous -- if you give the players the tools to interact with that world in a meaningful way.

It seems to me that games over the past two decades have been leaning more toward trying to wow players with astounding visuals but not allowing them to interact with that environment. It's almost more analogous to a theme park or an art exhibit than an experience. You are to see it and move on. The player has little to no relationship with that world. No matter how pretty it is, it cannot have the same emotional and psychological impact as finding something truly impressive in Minecraft, knowing that it is unique and special, and knowing that every brick of it is yours to manipulate.

Roguelikes in general. I've been impressed by certain roguelikes as of late. Dwarf Fortress is of course an old favorite and an excellent example of how much depth and essentially infinite gameplay can be derived from a (very) minimalist graphical engine (if you can call it that). Another roguelike called Cataclysm: Dark Days Ahead allows the player to play the role of a survivor in a post-apocalyptic New England, with all the day-to-day problems of survival surprisingly well simulated. Food and drink, medicine, vehicles, crafting and construction, weapons and fighting, and infinite exploration. There is some incredible emergent gameplay and it's actually pretty easy to get into for a game with that kind of depth, a depth that is made all the more appreciable thanks to the constant risk of permanent death.

I've always liked games that give players the tools to craft their own stories. I don't mind a brief narrative game; sometimes the experience can be pretty dramatic. But I've never played one that I connected to as much as a game that took the leash off and allowed me to explore and craft my own narrative. Those are infinite games, because as long as my brain is functioning, I can find new stories to tell in that world.

Ian Richard
profile image
I'm with most of the others here. I often buy a game BECAUSE of it's limited experience. Sometimes I feel like being a desperate soldier stranded behind enemy lines. Sometimes I want to play a smart-alec explorer and sometimes I just want to blow **** up.

Infinite games often fail for me because they don't give me that single focused experience. They give me a mechanical time-killer that never ends. I don't care about random drops... or procedural dungeons... I care about fun for as long as I'm playing.

As for being a developer... I don't EVER want to work on an infinite game.

- I've worked on software thats been in development 15 years and it's ugly. It's often messy and overly complicated due to changes over the years. A new project ever 3 years means that you can write it efficiently using your newly gained knowledge.

- I make games because I love to create. I don't want to be stuck on one project for the rest of my life.

- I like making money off each new project. I like hearing the opinions of players on what went right or wrong. I like learning from my mistakes so that I can do better next time.

I understand your point, but I personally prefer the "Throwaway" projects.

Michael Joseph
profile image
Fair enough. I don't think you have to worry about "infinite" games taking over any time soon :)

Anton Temba
profile image
Your mention of working on an ingame raises an important point.

Infinite games are not suitable for experimentation. If you're setting out to make an infinite game, a very clear vision is required and a project like this is often developed in the waterfall development model, simply because it needs a very robust, powerful framework to support a content system as interconnected as I mentioned in a reply a few posts above yours.

This means the design phase for an infinite game is the extremely important and also takes long for this very reason.

Also, as for limited experiences, an infinite game is basically an infinite generator of limited experiences and you can also choose which to go for and which to ignore. The cool thing about it, is that it all happens within a persistent and consistent universe, so the adventures you had fun with in the past can leave a mark that can be seen when you're on other adventures in the future.

Eido Gamer
profile image
Those interested in creating a persistent world -- or to use the author's terminology, an infinite world -- should watch the progress at of Chris Roberts' Star Citizen game.

He's not building a game. He's building a persistent world precisely along the lines the author suggests. Crowdfunding is now over $8.5M and growing (still).

Gustavo Martins
profile image
What if the only thing I desire is to tell a story?

You start with the assumption that movies/books etc are used up once, and that's it. Which, for some people, is completely wrong. There's movies I love that I watch 5, 6, 10 times, all for different reasons and searching different things. There's also games that I replay through several times, some times because I liked the story so much, sometimes because I liked the gameplay so much, sometimes because there was nothing else to do.

When you say that developers are wasting their time making games that aren't endless, it makes you sound slightly condescending. Just like any form of artistic and creative endeavour, people have different goals, and that's what makes it exciting. Even if only to serve the purpose of showing how good other genres are by being bad, they still serve a purpose.

My main issue with procedural everything is that for me, it loses it's meaning. It doesn't feel part of a
bigger thing. That's why my copy of Skyrim has sat unplayed in my HD for months, after I finished doing all the sidequests and main mission. However endless the "radiant" missions are, they just don't feel meaningful. Surely that can be remedied by having them have a bigger impact on how the world
changes, but so far it hasn't. If you can accomplish that, you're golden.

You also do not account for changes in the hardware. What would be the platform that would host your infinite lifetime game? The PC stands out as the best choice, but every new iteration, every new piece of hardware, you would have to do some tweaking in order to make sure it runs. What if in a new generation, your base code simply stops working properly? Will you recode the entire game from scratch? What if the only possible fix will destroy everything that the other players have already accomplished? Will you stop working on it and tell players to not update their machines if they wanna keep playing? What if the genre your game is made for falls out of fashion, and everyone stops playing. Will you switch off development and kill the game as it is?

To me, if you truly wish to make a game with "infinite lifespan", make a game that is great. Make it so great, people years from now will find a way to run it on their machines. Take classic movies, classic books, classic stories, they were not made to have an infinite lifespan, but they achieved that through greatness (or, as it is the case, infamy).

Now if you really want to implement some proceduralness to your games to spice things up, use some sort of butterfly effect approach: Make a basic story that the player has to follow, a standard adventure framework. But the catch is: every time he starts a new playthrough, have the game change a set of variables somewhere that alter the starting conditions, thereby leading to a different end while still maintaining the basics of good storytelling, leading to a good experience overall every time you play it. That's something I would love to see.

Michael Joseph
profile image
"That's why my copy of Skyrim has sat unplayed in my HD for months, after I finished doing all the sidequests and main mission. However endless the "radiant" missions are, they just don't feel meaningful"

I don't think that is necessarily a problem inherent to procedural worlds. It's a problem with heavy handed story games that drop you into a large world where all the design focus went into the story.

Procedural game worlds have to be first class worlds.

Chris OKeefe
profile image
Skyrim's radiant quests are not really a good example of procedural gameplay. All it is doing is selecting a pre-generated location and telling you to go kill something there. Or steal something. It feels flat not because procedural gameplay is necessarily flat. It feels flat because there's really not much to it.

It sounds to me from the way you are writing that you feel that 'infinite gameplay' is some kind of holy grail that hasn't been achieved. That's not the case. People have been playing Dwarf Fortress for years and they are still having new and novel experiences. Roguelikes as a genre are designed to be infinite and procedural, and some of them do so in very interesting and unique ways.

The key to infinite gameplay - in my opinion - is always going to be player agency. If you give the player the tools to interact with and manipulate their environment and make it their own, even a simplistic world such as that in roguelikes or Minecraft, can become living, breathing places. Their lack of visual fidelity is compensated for by their uniqueness, and their uniqueness is compounded by the stories we craft in those spaces.

Lately I've been playing a roguelike called Cataclysm: Dark Days Ahead. One interesting feature of the game is that you can, if you like, maintain a persistent world. One character's life may end with a bloody battle in a gun store. She hid behind a counter and kept firing at a zombie horde, but her ammo ran out and she was overwhelmed. After starting a new character you eventually stumble across the town she died in. What's left is the visual story left imprinted upon her death; blood and bodies piled up, shell casings everywhere, display cases knocked over and smashed, windows busted, and the character's corpse laying mangled behind the counter.

Although these kinds of narratives may not have the backing of paid voice actors, scene-specific scripting, and custom animations... they are unique, novel, and emergent. They are the direct consquence of player choices, good and bad. They are alive with potential, with possibilities for other outcomes. Nothing is certain. There is no inevitable quicktime event. No deus ex machina. No guaranteed survival. No quickloading. Just choices to make and the narratives that arise from those choices.

Yes it takes some imagination to go from ASCII to a detailed mental image, but perhaps the future will hold procedurally generated content on that scale that doesn't smack of copy/pasting.

Bart Stewart
profile image
I'm glad to see the interest this idea is getting -- it's something I've been thinking about for a while now.

My take on it, then and now, is that there's a market being underserved. Not everyone wants an infinite game -- for many gamers, if you can't win (and thus cause the game to end) then there's no point in playing. (Alternately: "that's not a real game!")

But that doesn't mean there aren't a meaningful number of people who would enjoy playing a game that was designed from the ground up to be playable -- not replayable, but *continuously playable* -- over years. Such a game would have the persistence of a MMO while being playable solo or with only a small group. The MMO aspect would be simulated by building highly dynamic and interconnected world-systems, along with NPCs with considerably more reactive and expressive power than the loot piñatas of today's MMORPGs.

I wrote up some ideas along those lines a while back here on Gamasutra:
ing_Worldquot_Game.php . And funnily enough, I was fiddling not five minutes ago with a one-page design doc for this Very Thing. So I think it's fair to say I am enthusiastically on board with the idea of a game that is intended to be so large and so dynamic, with no scripted content but a great variety of emergent content, that it could be played every day for ten years and offer new fun every time.

Of course there are practical difficulties to such a project, just like every project of any scope has. Monetization and how to create a lot of good-quality content are fair questions... but those are implementation-level questions, which can be solved.

What matters is the idea itself: whether a game can be designed that is attractive and engaging enough, and that generates its own content well enough, that some reasonable number of people would want to keep playing it for years.

I believe such games can be made.

On the other hand, once such a game does get made, maybe that's the last game a subset of gamers ever buy.... ;)

Nick Harris
profile image
I liked your blog post, then read:

...and made a long comment there, which apparently got swallowed into the digital void. So, I have posted it elsewhere on Gamasutra despite it also being pertinent to this discussion:

Bob Satori
profile image
Many players long for games that are not of limited use by design. But publishers? I don't think so.

Nor do I believe that any gamer or game-designer who feels that any scripted game ever was "awesome" will ever be satisfied with a fundamentally procedural design, particularly if they are attempting to recreate the scripted experience.

Good procedural game designs do exist and can certainly be awesome, but their creators are more likely the sort who have trouble understanding the appeal of scripted narrative "games."

Bart Stewart
profile image
>> Good procedural game designs do exist and can certainly be awesome, but their creators are more likely the sort who have trouble understanding the appeal of scripted narrative "games."

I think I would say that it is exactly the other way around: the people who naturally like to make highly scripted games find it hard to see how freeform, player-centric games could possibly be enjoyable to anyone. "If you don't tell players what they're supposed to do, how can you be sure they're having fun?"

Your point that traditional publishers won't care for infinite games is a good one, though. ;)