Slashdot Mirror
Can We Create Fun Games Automatically?
togelius writes "What makes games fun? Some (e.g. Raph Koster) claim that fun is learning — fun games are those which are easy to learn, but hard to master, with a long and smooth learning curve. I think we can create fun game rules automatically through measuring their learnability. In a recent experiment, we do this using evolutionary computation, and create some simple Pacman-like new games completely without human intervention! Perhaps this has a future in game design? The academic paper (PDF) is available as well."
You want your creativity to be automated?
You desserve what you'll get, welcome to your dump...
Hope the rest of the world will leave you there, for once.
Smile, don't click...
Can We Create Fun Games Automatically?
Sure we can, depending on your definition of the words 'Fun', 'Game' and 'Automatically'.
:P
Site & blog: http://www.mayaposch.com
The more I play games, both video games, board games or pen and paper RPGs, the more I see the obvious patterns that exist beneath them.
I stopped playing new boardgames as all these become obvious after a few games, and if you tend to like one, old games already implement them perfectly. You basically have 3 (arguably 4) components in any board game : randomness (go play dices if you like it), tactical planning (go play chess), bluffing (go play poker) and, arguably, negotiation that can be seen as a merge between tactics and planning but that often use a whole different range of social skills.
Video games have also some recurring ingredients. I played less of them so I fail to see them more clearly, but some of them are obvious :
- a sentiment of progression. Whether artificial (through leveling in RPG games) or real (from FPS where you get better at shooting, rocket jumping, etc...)
- hidden content of the game, that the player has to find or guess. It is usually some content voluntarily put there by the game developer (quests, levels, maps) some hidden game logic that one must understand (AIs behavior, puzzles, research trees). In the most interesting games (in my humble opinion) there is also content that is almost emergent. The creator only loosely coded some rules and it is the player's actions that create his own problems to solve. It often happens in strategic or development games, where you discover that a design you chose had some vulnerabilities and that by correcting this, you create a whole bunch of new problems.
That one last part is the most difficult to reproduce automatically, in my opinion. But a lot of successful games don't have any such emergent content, so I guess that automated games generation can prove quite fruitful !
The Wise adapts himself to the world. The Fool adapts the world to himself. Therefore, all progress depends on the Fool.
Can we get research grant funding automatically ?
I believe the answer is yes.
1. Choose a 25 year old topic (for example, a Pacmangame), reinvent it using lots of buzzwords such as swarm, hive, collective, competitive, but secretly just program a system using some generic rules, and a gradient descent algorithm that will force those generic rules to conform to the behaviour we wanted in the first place. Then publish a PDF (why oh why by the way is PDF proprietary format ANY better than Microsoft's proprietary format ?), and spam it across tech news sites.
2. Make some wild claim that this is the dawning of the age of Aquarius (or similar).
3. ???
4. Profit !
Good point. Different levels of "fun" and satisfaction.
Someone wrote about putting Age of Empires 2 on showroom PCs, and all the female customers went ga-ga over this game. They would then build mini cities and so on ... all without fighting. He said they wouldn't give a second look at AoE 3, or The Sims 2 ... they just wanted to play AoE 2.
Someone wrote about his entire family playing mostly older games (including all Mario games), and mostly avoiding newer, copy-protected games.
It amazes me reading these posts.
PDF has been opened. Admittedly the standards body which supports it is ISO, but I don't think anyone bribed them to approve it.
Seeing movies produced by following the "formula", do you want automated games? Do you even want a "formula" for "fun" game design?
Maybe its possible, but this starts to sound like automated art.
Think Deeply.
If programming and design could be done automatically, we wouldn't still have programmers. We can't even manage to automate creating simple apps. How could we possibly automate creating entire new games, which means new art, new rules, new everything.
On top of that, everyone finds something different in a game to be 'fun'. Some love challenge, some love adventure, some love collecting things... Attempts to make games that have everything anyone could love are usually pathetic flops.
"If you make people think they're thinking, they'll love you; But if you really make them think, they'll hate you." - DM
Walking into a computer lab at school, spying a mystified user staring at a screen. Investigating further, it turned out he was confused by the fact that
Make Game
Racing Game
2 tracks
In a programming IDE did not yield anything.
I record my sleeptalking
Sure, if you define "fun" as "a smooth learning curve", then you can make fun games automatically.
But not all of the fun is in the learning. Some fun is in tweaking humor. Some fun is in triggering a person's likes and dislikes (Nethack, ponies). Some fun is created by changing the venue (is it a space game? a historical shoot-em-up? A politics game?
Yes, there are underlying patterns to a lot of games. But simply limiting our definition of "fun" to "learning" and "follows the pattern" reminds me of the automatic novel generations in Orwell's 1984.
I don't think that this headline defined the problem well. Yes, some parts of fun can be automatically generated. But no, to make a fun game, it has to be interesting to a human, not just to a turing machine. And for that, you really need other humans to make the games, or you don't have the depth required for real "fun".
Correct Horse Battery Staple: 72 bits of entropy. Enter "Correct H" into google. When it generates the phrase, that's
They should have sent a slashdotter. They never read the story...
It pays to be obvious, especially if you have a reputation for being subtle.
fun games are those which are easy to learn, but hard to master, with a long and smooth learning curve.
Best example for this is Chess. Easy to learn but takes many years to master.
Umm... Mac OS X? Reads PDF quite well, with no external software installed whatsoever.
Windows? Bah!
One factor which I've noticed tends to create addictive puzzlers is quite simplicity of form. The resulting purity of function lends itself perfectly to entrancing, mesmerizing marathon sessions of blocks dropping, diamond spinning or whatnot, always seeking "one more combo!" as the points rack up on top of the screen. Tetris, Lumines, Bejeweled, the list goes on. Keep the concept simple, the list of controls short and the rules easy to learn. If I looked up the amount of time I spent trying to line up that four-block line in a perfect spot for maximum points, I'm pretty sure the number would be terrifyingly high.
Automatically? Most game dev studios can't even make fun games manually!
Yes, that was Stanislaw Lem; one of his Trurl and Klaupacius stories from The Cyberiad. "The First Sally (A), or Trurl's Electronic Bard"
In photography, you set up the boundary conditions, take a TON of pictures, then select the best ones from the ones you have. The best photographers have the best eye for selecting the remarkable ones out of the pack. This would shift game programming from an art like classical sculpture - where you have to plan far, far ahead, and don't get second chances - to an art like photography where it is more about creative curation than creative engineering. Evoluationary development of games wouldn't eliminate the creativity of the process or the product, it would change the creativity of the process and the product.
"Can We Create Fun Games Automatically?"
Sure if you want a boring game. So lets see what are the common aspects that help a game along and can an AI game writing system create the environment that usually aids in creating a good game?
Here is what is required based on the one truth mentioned in the story regarding 'easy to play hard to master' I will use online FPS games as the example because that is mostly where my head is at. But my comments also apply to most other games and for single player with a little tweaking.
The basics of play should be simple, allowing some success even with the basic weapons and basic movement skills. This is very important. To win requires to master both weapons and movements. For other games it is less acquired but allows advancing to a degree and not a dead stop in the game. Avoid at all costs where it requires going back significantly in order to acquire what is needed. Fortunately fps games rarely require this even in single player unless it is a long level and it must be started from the beginning. This is greatly assisted and avoided if saved games has a variable rollback feature.
Along with this, is the ability to replay a segment where you see your mistakes and even have them pointed out to you by the game visually or spoken. ie: You missed an item(s). You should have used this weapon for that distance or situation etc. Live tip manual system of sorts.
The environment must supply a few rudimentary visual and emotional rewards. In other words. There must be a certain level of eye candy that attracts the attention of the player. Actions of the player generate some reward such as a visual or sound and object reward beyond simple item reward. This should be removable for the skilled player. IE: Player ability to remove fancy explosions that affect frame rates and dial down sound effects that can be annoying over time.
Direct player to player interactions. The more a player can interact with friends the better. This includes such things as a text or voice channel. It is important in a game to be able to express your and hear the emotions of others.
Puzzles should be scalable and non linear. Basically to be able to set level of difficulty and with multiple outcomes and solutions. Do I need to explain why? Replay value is one reason.
Optional: A system that matches up equally skilled players and or a handicap system.
Interesting immersive world. This is the creative component that no machine can generate and why we still require real coders and artists bums in seats. So many games these days do not offer enough variation from game version to version. I am convinced this is due to a creative decision to keep a familiarity and not simply lack of creativity. There is room to maintain a series and still vary the content for the hardcore audience. After all, the only reason you do this is because of the repeat customer and certainly not for the new blood. For example: BF2142 did not fail, but it was also not a huge hit, simply because it was significantly buggy as to limit it's appeal, especially for the repeat customer. So a conclusion can be drawn that familiarity is not the be all, end all. The internet factor killed off the game after initial sales as word spread of major unfixed bugs after each patch. Call it the frustration factor.
The frustration factor. This is a big issue for any player. The cause is varied. I mentioned bugs that are game killers. From crashes to unexpected problems that prevent proper completion of some task or level segment. It goes well into the required skill to complete an action or solve a puzzle as well. Testers with various skill sets must be used before the game is released to avoid these issues. Beta testers tend to be in house and are to familiar or just as bad, drawn from the trusted experts in the gaming communities. The cause and effect is to many knowledge based problems ocurr that turn off the beginner. It is like writing a document and not including the acronym definitions properly or not at
Fun for me in games usually means there's always something to do or press. The old 2D games were more like this, but a large reason why I hate modern 3D games, is that there's often lots of sprawling around without really doing much (partially related to the 3D world, but it can be solved with difficulty).
I like the idea of how the article mentions that the algorithm biases towards games which can't just be won randomly. The board game is Go is the ultimate example of this I guess, where there are many *levels* of mastery.
But one has to be careful with this approach. If in a 3D game there's a small opening in (say) a castle wall, and miles around of plain grass, it's pretty easy to solve for a human player, despite the huge searchscape and 'narrow' solution that a computer would find tricky (which would apparently potentially rate as a good 'puzzle').
At the very least, developing models for other human factors such as reaction time, subtlety of graphic elements, and the challenge of pressing certain key combinations, would also be needed before final game automation could be achieved.
Why OpalCalc is the best Windows calc
90% of games that came out wouldn't suck
juenger1701
I propose an interactive website that posts a bunch of text, then people can create an account and add other bits of text to it. Some people will find those new bits so interesting or so ugly they will add more bits of text. Then people who posts lots of text will get credits that they can spend to hide other people's text that they don't like.
Then we'll pretend it's all factual and news and call it Slashdot!
Evolution SHMUP: http://www.kloonigames.com/blog/games/evolution-shmup
This was an experimental game, where the theory was to see how long a player would remain in a given game - as people continued playing, the system adapted gradually in order to maximize the fun value (in this case, the amount of time spent on a single game.)
This experiment has a smaller search space than the article, but isn't generating any "successful" games. This may be caused by the environment(i.e. the evolution scope is too narrow and thus isn't generating a variety of enemies), but the same problem can easily apply to the article in question.