Wednesday, August 29, 2007

The Idea of Rule Shifting

Most of the fun in a game comes from exploring the ways the rules and context interact.

In a platform game such as Megaman or Prince of Persia, the player explores how rules and context interact using simple rules and complex contexts. Ammo and health are unlikely to matter in ten minutes - you'll probably get boosted back to max between encounters.

Once the player starts getting used to the ways the rules interact with the contexts, these kinds of games will usually give the player an upgrade. Double-jump is a common example, as are immunity to spikes or a new special weapon. This changes the rules and lets the designer present all the contexts again, from a new perspective.

This is a tried and true method of making fun games.

Many games (such as Castlevania: Symphony of the Night and most Final Fantasies) add a "stepped rule change" as well. These games continually change the rules of the game in small doses by allowing you to change your statistics - upgrade your attack, your defense. This slightly changes the rules of how you interact with the context. Because of this, these games do the "big shifts" less often.

In a classic platformer, you're likely to get an upgrade that somewhat changes the nature of the game every 2-3 levels, usually after each boss fight. In games with a more statistical approach, you'll generally get these kinds of upgrades only two or three times over the course of the whole game. In Final Fantasy, you generally get a chocobo, then an airship, for example. Often you start with no magic, then get magic. These are changes that alter the nature of the game, but they happen very rarely compared to Megaman getting a new gun. This is because the small rule changes tide the player over.

But those stepped rule changes - usually just statistical changes - aren't enough to run the whole game, as many really boring RPGs attest. Similarly, if you hand out too many rule changes of the same basic type, they essentially become stepped rule changes. It's great to go from a wrench to a pistol, and fun to go from a pistol to a shotgun, but going from a machine gun to a fletchette machine gun to a plasma machine gun to an autocannon is not nearly so entrancing. It's basically just a statistical upgrade with a new mesh slapped in. As many boring FPS games can attest.

A good rule of thumb is this: if it changes the way the game is played, it's a real rule change. If it simply changes how easily the player deals with a specific kind of content, it's a statistical change.

But statistical changes shouldn't get the short shaft, because they are often linked to a second game! For example, getting better statistics in an RPG usually involves killing monsters for XP and gold. This puts the focus of the game on killing monsters, unlike a game with no such curve - in Megaman it is common to simply get past the monsters without bothering to engage them.

Similarly, having a long-term ammo supply means that there is a second game of preserving and hunting for ammo. Done correctly, this will even make a player choose a sub-optimal weapon simply to preserve ammo for his more optimal weapons. Guns such as the BFG can exist because they have very little ammo - if ammo for these weapons is easy to come across, the game breaks.

These secondary games follow the same basic idea: they are fun because their context and rules interact in interesting ways. The context of the primary game is basically the level, but the context of the secondary game is, to a large extent, the primary game!

In some ways this means it is difficult to "plan" the contexts. However, you can simply look at the salient elements of the level and the gross rules: if the level is big on sniping, obviously the secondary game is going to be cast in the context of a sniping level. If the level is about jumping, the secondary game's context is jumping.

This gets more complex if the context is largely player-generated. For example, in most Final Fantasy games you get to develop your characters in any way you want. This means that the secondary game (gaining new powers and statistics) is largely affected by the context of how the secondary game has been played so far! It stopped being a linear system and turned into a multi-dimensional glowing purple frog.

Anyway, that's definitely enough writing on this subject. :P

Comments? Questions? Monkeys?

Monday, August 27, 2007

The Importance of Attrition

So, like a few other people, I've been complaining about the gameplay of Bioshock. I compare it to System Shock 2 and wonder whether they did any playtesting at all.

But, on the surface, it's almost identical to SS2. Vita chambers and quantum entanglement units are basically the same thing - they resurrect you when you die. Similarly, both games are about door-to-door mayhem with guns and hacking and psychic powers. Actually, if anything, Bioshock should be better because the psychic powers don't suck.

I won't claim that SS2 got the gameplay right on purpose. After all, they didn't get anything else right on purpose. But they got a whole lot right on accident - from level design to storytelling and over to enemy design. Presumably, they accidentally made the gameplay fun, too. Amusingly, these are all the things that Bioshock did not achieve. Even working from SS2's blueprints, they were unable to duplicate SS2's high points, although they did have high points of their own that System Shock did not achieve.

The two games can't be directly compared without taking into account the fact that SS2 is a butt-ugly game with a UI that shows its age. It is almost ten years old, after all. BUT, if we were to imagine SS2 bumped up to Bioshock level graphics, I would not have a hard time choosing a favorite.

The reason the SS2 is so much more fun for me is simple: it's scary. Really scary.

But Bioshock is about as scary as peeling paint.

They have the same gameplay! Why is one scary, the other not?

As you might have guessed, I'm claiming it's because of attrition.

There is an issue commonly called "the Quicksave Dilemma". Basically, PC games are expected to have a quicksave-quickload functionality. But this puts a hell of a lot of pressure on the game designers, because whenever the player does poorly, they simply load the game and go through the same situation with foreknowledge of everything. Rinse and repeat, and you have a player that basically goes through the game "cheating" past your challenges. This reduces the depth of the gameplay by "short circuiting" the parts of the gameplay that are about resource management. Who needs resources when you can re-load until you get through without losing any health?

This is a huge problem in surprise-based games. If a big part of the gameplay is stumbling into an unknown situation and then having to fight your way out, quicksave will destroy the game entirely. These days, surprise-based games largely leave the resource management section of the gameplay to atrophy, boosting the moment-to-moment adrenaline-inducing play to compensate. You see this in Doom III, for example, and in Bioshock.

The vita chambers - and the quantum entanglement units - are basically an automatic quicksave-quickload, with the added bonus that all your enemies stay injured. So it induces all the problems that quicksaving induces.

There are a couple of ways of reducing this quicksave problem, but my favorite (and the best for horror games) is by using attrition.

No given situation will kill you, unless you do something completely silly. No given enemy is likely to even hurt you much, unless they're a big nasty. And not only that, the enemies give themselves away: you know precisely where they are and what they are, because they jabber to themselves. "Babies need ressssst... babies need sleep!"

But as you go from room to room, you'll continually be hurt a little, and that really starts to add up. The same principle applies for ammo or any other resources: they slowly get used up.

This isn't really something you quicksave to beat. It's not a lot of damage, and you can't usually do all that much better by quickloading.

Instead, the game drags your resources down despite the fact that you can quicksave and quickload. This isn't perfect: some people are so addicted that they will quickload every time they are hit. Especially in this day and age, where quicksaving and quickloading are really quick - when I originally played SS2, the "quick"save took about seven seconds. But it is a fairly good solution, especially if your quicksaving has other restrictions.

It is what System Shock did. Virtually every enemy save the wrench zombies was likely to hit you in the amount of time it took you to kill them, and they would all use up precious precious bullets. It was why you always got nervous when you heard the monkeys... the damn monkeys!

Bioshock did the Doom III method. Instead of dragging on your resources, they give you a huge amount of resources and cap your maximum carry. After any given fight, you could go back and scrounge your way to full capacity pretty easily, especially in late game. This means that any given fight has to be a significant threat. Every enemy needs to at least have the capacity to kill you.

This means that every fight is worth quicksaving through. It means that you're going to die a lot, and it's not going to matter. In Bioshock, dying is quickloading.

In Bioshock, there is no attrition. In System Shock, there is.

System Shock is scary. Bioshock isn't.


Also, System Shock 2 didn't resort to numbering their spells. "Immolate 2"? "Immolate 3"? "Immolate-aga"? :(

Bioshock: Final Review

Earlier, I posted a pre-review of Bioshock. Well, I finished it, so here is my final review.

The game's last level (and the cutscene before the last level) are the best parts of the game. The imagery is very good, and this is the first game - ever - that I suspect may have played over the player's heads. I especially liked that they didn't shy away from images that a committee would have usually voted out. Even had the rest of the game sucked, it would have been worth playing simply for the imagery.

It was nice to see a classic "failing utopia": felt like I was watching old sci-fi movies like Logan's Run or Lost Planet (I know, I know...). Except, you know, upgraded with fifteen million dollars of technology.

Apparently I got the good ending, although it felt... rushed. What is it with games these days? Is a dénouement too expensive? It makes me cry. Still, I guess I can't hold it against Bioshock, as no game does dénouements.

The game's central theme - whether they wanted it to be or not - was "betrayal". I can't remember a single plot event - not ONE - that didn't revolve around someone betraying someone or something (their values). It got to the point that the only plot twists I didn't expect were when someone died before they could betray more.

Still, all my complaints remain. Especially gameplay: the gameplay turned to complete crap once I realized that it was more effective to bludgeon big daddies to death with a wrench than to shoot them with explosives or AP rounds. After that, I simply killed everything with my wrench.

Gunmen? Wrench. Grenadiers? Wrench. Rocket-launching turrets? Wrench. Boss fights? Wrench.

It wasn't even "die and wrench again". After I started using the wrench, I never died again. So it wasn't simply Vita-chambers screwing the gameplay. There were a lot of gameplay flaws.

Man, the gameplay made me cry.

Still, gameplay isn't everything, and I rate Bioshock very highly.

Thursday, August 23, 2007

New Arcades

Ah, video arcades.

Video arcades work on a very specific principle: connectivity. In an arcade, you can play video games. You can even play video games with or against a small number of people (generally the games are either two or four player). That's connectivity, sure. But the real connectivity swirls around that.

There's asynchronous play - a high score table is the obvious example. You play with people (against people) without them even being there. The "line" is another example of connectivity: a dozen people crowded around the Street Fighter cabinet, lining up to fight the winner. While not playing - it happens a lot either because you're out of cash, not feeling that interested, or in a line - socializing is common. An arcade works on the principle of connecting you to other people while giving you an excuse: video games.

In America, the arcade went out of style in the eighties with the advent of home systems. The idea was that your children could come home and play games with their friends or family, instead of running off to the arcade to spend money with strangers far away from mommy's watchful eye. The crash of the home video game caused a revival in arcades for tolerably obvious reasons. Then the arcades crashed again as consoles revived...

In Japan, arcades remain popular even today, presumably because Japanese culture isn't really conducive to chummy family life. Legend has it that Space Invaders was so popular it caused a shortage of 100 yen coins, and even today weird new Japan-only arcade games are developed basically every year.

The arcade never died even in places where it seemed to. It simply mutated. There is always going to be a need for a place to connect while occasionally playing video games. In many places, the internet cafe has taken the spot that the arcade used to inhabit. Fundamentally, they're the same beast, except that the internet cafe uses computers that can play many games (and serve many purposes) instead of getting stuck with aging arcade cabinets. They also have a different atmosphere, of course.

Another example you might not guess is massively multiplayer games such as WoW. Fundamentally, these are games which use the exact same basic principles as arcades. You (and maybe a small group of friends) play games, often to compete asynchronously with others for status (rare equipment, level, rank...)

And, ask anyone, a MMOG is all about waiting. You'd think that a game you're running on your computer would let you play all the time, but that's not the case: you wait for travel, for spawns, for friends, for bids, for hundreds of things.

I bet if you compare play-to-wait times, you'll find that an MMOG has almost the exact same ratio as a popular arcade. And almost the same player reactions: some players will socialize, others will play secondary games, others will watch...

So, yes, fundamentally a MMOG is simply an arcade.

Can you think of anything else?

How about new cell phones?

Yeah, cell phones are fundamentally about chatting with your friends - frequently asynchronously using texting. But chances are, if you are the sort of person who texts, you also play with your phone. Not just games: music, photos, web surfing, all of these serve the same basic purpose as games, and are "asynchronous" in that you can share the results with people after you finish, rather than having to have them around while you do it. "Hey, look at this cool pic..."

Cell phones are evolving into arcades. Or, at least, a weird partial dispersed arcade...

I don't think it'll be long before phones evolve to fill this niche more perfectly. I can't predict exactly how this will happen, but I think it'll largely revolve around easier interfaces. Maybe cell phones that have touchscreens a'la the DS?

Can you think of anything else that is an arcade? Or maybe you have some thoughts on the future of cell phones?

Wednesday, August 22, 2007

Bioshock: Early Review

I haven't finished the game, yet: I think I'm 70-80% through it. Keeping that in mind... and keeping in mind the fact that System Shock II (SS2) is in my top three...

"Does it deserve a perfect score?"

Well, putting aside the fact that there shouldn't be such a thing as a perfect score - or even a score - no, it doesn't. So far.

It is a very good game. The presentation is incredible, the gameplay is relatively deep, the balance feels pretty good...

But I can't ignore the flaws. Here are the ones I noticed so far, in ascending importance:

The level design is... mediocre. I mean, it looks good, but if you compare these levels to SS2, you'll see that they are inferior. SS2 used corridors, large chambers, and lots and lots of side rooms. This game uses all of these things as well, except it's linear. You don't walk down a hall and duck into all the side rooms. You move from point A to point B. Occasionally there will be a major side route to pick up a specific item, but it's not the same as having a dozen suites lining the corridor. In my opinion, this makes the game world feel very transparent.

The enemy is design is... mediocre. So far, their best enemy design is the big daddy, and in all honesty, he's not cool enough to hold the entire game up on his own. All the other enemies are zombies with various kinds of weapons. There's a really boring teleporting zombie, too. Ooh.

The game is not fucking scary. At all. I was a little worried initially, but then I fought my first big daddy and got my ass killed. There is absolutely zero penalty for death, which means the tension went away. I got in the habit of resigning myself to multiple deaths per big daddy - that's not the way a scary game works.

Lastly, the writing is mediocre. While the method of presenting the writing (journals) is excellent, and the voice acting is excellent, the writing itself is horrifyingly flabby. It slobbers its way through whatever the local plot is, painfully landing on every point that needs to be made, and never wandering. The first one or two journals I find in an area are interesting, but the next fifteen keep hitting the same plot at a positively slothlike pace, making sure nothing goes unexplained.

The actual radio communications are also rather clumsily timed. It's too pat, too unrealistic.

Compare this to SS2, where the journals were fragmentary snippets on a dozen different plot lines. Nothing was fully explained, nothing was labored upon, the writing was lean and sharp. SHODAN's radio communications went on and on in a realistic manner - it gave the impression she was really paying attention to you, unlike the characters in Bioshock, who apparently only pay attention to you when you're about to do something momentous.

I know that a lot of people are going to be irritated by my review, but so far I haven't found the game to be worth a perfect. It's good, definitely play it. The fiction is interesting (although a rather painful farce), the world is well-defined... but I was a little disappointed.

How do you like it?

Monday, August 20, 2007


I am known for my party animal nature in much the same way a wombat is known for his berserker rages. That is to say, I don't do parties, and alcohol just makes the headaches start earlier and end worse.

Still, if I had passed up the Bioshock "Appreciation Party", I probably would have been stabbed to death by feral game design students, so I went. I didn't stay terribly long, as the post time indicates. As I mentioned: headaches.

Anyhow, it was a really nice party. It doesn't take "swankiest", that goes to Intel. But it certainly wins in "models dressed in blood and cancer pretending to be mannequins".

I got myself injected with apple tequila, shamelessly stole a glowing swizzle stick, admired the giant big daddy, avoided the consoles set up with the game (I'm'a gonna play it tomorrow), and convinced someone to send me pictures. It's hard to overstate the ambiance of the place: the whole club was redone to be Rapturous. Pictures will show that, I'm sure.

The big thing for me was a TARDIS in the corner. Yehaw! Apparently the club has one by default. Now that's one hell of a cool club.

Then I went home. I'm a party animal!

Edit: here are some pictures, although they aren't the ones I want. Those "mannequins" aren't.

"Virtual" Unequal Information

Yeah, okay, I'm really getting ivory tower in this one. It's so minute...

Last post I talked about unequal information. Basically, I said that differences in what information various characters, factions, and players have is what brings context to a game and to a story. And context is something that computers are extremely poor at.

However, meta-context is very important as well. For example, in a tabletop RPG, players think - contextually - that they are uncovering a story with the help of a GM. If they begin to think that the GM is generating the story on the fly (rather than just adjusting it somewhat on the fly), this can change the context and render all their information worthless.

Sure, it can be argued that players are always just exploring the GM's mind. But there is a contextual difference: a scripted story exists on its own, largely separate from the GM, even if it was written by him. It's a different context, a world of "solid" information instead of "fluid" information.

Of course, players often like playing a generative game just as much. It's not that one is better than the other (although an (at least partially) scripted story is more efficient at luring players in), it's just that they have very different contexts, and the players will collect information in a very different way.

Context can completely change everything, even if viewed the same way. For example, the big reveal in Fight Club changed the entire movie, and watching it again with foreknowledge of that reveal makes it a completely different movie.

The context differences between characters, players, factions, and so forth are what drive games and stories. If you play Crackdown to leap from rooftop to rooftop, you're going to have a completely different game than if you play to kill bosses. You're in the same city, but experiencing it in another way.

Hey, I played killing bosses by kicking them in the face lots, and my office mate played by shooting them and grenading them. We had very different experiences.

Computers are really poor at handling context... but I wonder if that's just because nobody has really thought about it in those terms before? Is it possible to build a "context engine" that can manage this kind of thing? Not just to manage the player's context, but also the character contexts within the game?

Thursday, August 16, 2007

Unequal Information

I've been talking about how computers aren't very good at context. It's really one of the biggest reasons that generated content is uninspiring: the computer doesn't know how to make a layered, subtle situation. Even the most basic first person shooter has remarkably complex contexts when you look at it in comparison to what a computer can do.

A huge part of context is unequal information. Not only between characters, but between the audience (player) and the characters or world. Sometimes the player might have more information, sometimes less. Both are common.

For example, in an RPG it is common to cut to the evil villains doing some nefarious thing - burning down a village or plotting to capture someone or something. The party members blithely waltz into the next town anyway. Although you know what's coming, the party members do not, creating a tension.

An opposite example is equally easy: in most decent FPS games, the main character has some kind of past... but you don't know much about it. It's revealed over the course of play, bit by bit, as it affects the world. The character knows things you don't know, and acts on those things.

For example, what is the Spartan's background? How about XIII? How about JC Denton? They often act in unusual ways or get in unusual situations due to their background...

Hell, even the old Super Mario Bros had this kind of information imbalance. It's a key part of making a story, and almost no story goes without it.

But there are no methods of plot generation which think in this way... at best, they think in terms of character goals. While a character goal is (often) hidden information, it's only a pale sliver of information imbalance. There are no algorithms that manage information imbalance.

I wonder if one could be built.

Tuesday, August 14, 2007

Component Plots

When you talk about "generative" or "adaptive" plots, you run into a lot of problems with implementation.

The common approach is to carefully script out every possible plot. I call this the "big atom" method - the "atomic" elements of your plot are chunks of plot that are often several levels long. This produces a massive "branching" problem, where the more influence you allow a player, the more scripting you have to do. Lots of games use this method, including the much-vaunted KOTOR and Deus Ex series. And they have it polished: they know a lot of tricks to keep the branching down and reuse script.

But the fact is that the method is limited because you have to specifically script out every element of the plot.

An approach a lot of people try is a "small atom" approach, where they create tiny, generic plot elements that are stapled together over the course of the game. This isn't used in real games, however, because context is a huge part of games and we don't really have an algorithm for that figured out. At best, the "small atom" approach creates a meandering, disconnected plot. Normally, it creates gibberish.

There is really no "medium atom" approach in common use. The closest we have is map generators in things like Diablo, where individual chunks of a dungeon are created, but they are arranged into a map somewhat randomly. This works decently for hack-and-slash games, but if you try to implement it for a plot, you end up with the worst elements of big and small atom approaches: although any given segment of the plot makes sense, over the long haul the plot is a meandering mess. And you have to carefully script every element. Although rare, this method has been used occasionally, usually by people like Chris Crawford.

Sure, there are steps you can take to make these systems a bit more palatable. For example, you can use a "big atom" approach for the base plot, but then use "medium atom" elements to determine how the player guides the "big atom" elements. You can even subdivide that to allow the "medium atom" elements to be guided by "small atom" events... Siboot uses small atoms with medium atoms to give some context.

Alternately, you can be vague, generic, or carefully insert the reasoning between plot elements, allowing players to think there is some mysterious force behind things or letting them come up with their own reasoning. This works best in games where you can hear the player thinking and then come up with actual reasons similar to (but not exactly equal to) the player's reasoning. A computer game could be built with this in mind, but it has yet to be done.

However, these are all dodges. The real problem with all of these approaches is that a computer doesn't know anything about context, and programming it to know something about context is, as of yet, unproven theory. The larger atoms basically let the writers create more context, to supplant the computer's lack of context. Of course, what causes the branching and the pain is the fact that context changes as the plot progresses differently... so the writers have to write more variants, and you end up with a huge amount script.

As far as I can tell, there are three ways to implement context that aren't simply changing the size of the "atomic" script or pretending to have a magic algorithm that "knows" context.

1) Player-generated context. If you supply tiny story atoms to players, they will be happy to assemble them into fully functional plots that make sense. Assuming some kind of massive player base, you would see something akin to the Sims' photo albums.

2) Level-generated context. Similar to having AI navigate a level by embedding pathing and waypoints, you can have a "map" of a plot and fill it in with chunks that are context-suited. If you know that there is going to be a lovecraftian demon as your end boss, you can focus on adding horrifying plot elements rather than adventuresome plot elements, and even have a pseudorandom beginning, middle, and end. This is similar to having "intelligent" atoms, but far more centralized.

3) "Long atoms". If you want context, why not build the context into the atom? If you have a lovecraftian demon as an end boss, you can make that plot element "cast back" a series of hints and foreshadowing, even call for suitable atoms to attach to those nodes. Similarly, if you accomplish a subplot early on in the game, it can "cast forward" the repercussions, injecting elements into the play at later moments without needing to "connect" to whatever elements are currently active. This is basically the reverse of level-generated context.


OLPC and The Only Computer You Have

So, I read this review of a beta2 OLPC laptop. Putting aside the actual language, the content of the review is definitely from a twelve year old. The big complaint is "it's slow, and needs rebooting every few hours".

I'm thinking back to my childhood... my old Apple II wouldn't run for more than a few hours. It would overheat and start randomly scrambling RAM. Rebooting didn't help, because rebooting didn't magically cool the chips.

This isn't a complaint about modern kids being spoiled. It's about what you're willing to put up with when you don't have any alternatives. You don't even really notice the limits and problems, because you aren't aware of better tools.

Frankly, I don't care how slow or balky the OLPC is. The only thing I care about is that it can be distributed easily and is durable enough to last a long time. If it runs out of power every hour, if it crashes every ten minutes... that's a big deal to us, but that's because we're all spoiled brats.

Of course, the beta4 is much faster and more reliable than the beta2, so I guess it's mostly moot.

Monday, August 13, 2007


I finally picked up a 360. One of the games I got was DOA4, because I liked DOA2 and DOA3.

I can't say that I'm an expert at DOA4, but I have now played it for many hours, and I have some impressions.

It looks like they were trying to get DOA4 to have a better high-end game than its predecessors: the priorities are a bit more complex and there's a host of new "split second" timing possibilities. They also tightened up the timing on counters and loosened the timing on throws, making the first a bit less dominant and the second a bit more (to the point where you can throw people while they're punching you in the face, which is very irritating to low-level fighters).

They also changed the AI. I find it harder, but that's probably just because I'm used to the old AI. I've always been better against humans. :P

They also weakened the floor game - it used to be that stomping on people and kicking them while they were down was a fun part of the low-skill combat. Although the moves still theoretically exist, they don't seem to be any use even in low-skill combats, which I find irritating. Even if a technique is dominated at higher skill levels, if it is fun for low-level fights, it's a nice ramp. Also, it's a bit easier to get fucked up in a corner, which is interesting at high levels (it's 3D, so keeping your position away from corners is far more possible than in, say, Street Fighter) but, again, irritating at low levels.

They also extended hit ranges a bit, which completely throws off my sense of distance, especially since these hit ranges often extend beyond the fist. That's just bad practice.

The whole game feels a bit more "turbo", which pisses me off. While the speed of a game doesn't affect my ability against other humans much, it makes the AI harder.

Anyway, the end result is that DOA4 is much less accessible than DOA2&3. My office mate tried to pick it up. While he's not an awesome fighting game savant, he did beat Art of Fighting. But DOA4 simply irritated him, because all the low-level play had been sacrificed.


Well, on the upside, there is a massive clarity in which characters have what strengths and weaknesses, and what characters they are strong or weak against. That's fun, and does allow for some level of handicap by carefully choosing a character strong or weak against the enemy.

But why don't they have a handicap proper? Sheesh, it's not hard to program in a damage reduction handicap.

Sunday, August 12, 2007


Ah, game criticism and academic discussion is advancing pretty quick. It's good to see.

But I always feel like these discussions are a little hollow. Because games are basically a set of rules and content, most discussions revolve around one of these two concepts. We're seeing more talk about balance, about reward patterns, about graphic style rather than graphics... we even see some commentary about the psychological "value" and "art" of games such as Ico.

But 99% of the time, they're talking about the game. I don't really think that you should focus on the game, because the experience is what the game is actually about.

A game can have shitty gameplay and still be very good. A game can have shitty art and still be very good. A game can have shitty balance and still be very good. It can be on tight rails and be good. Hell, it can have no bad guys at all and still be very good. This is because what experiences games provide varies from game to game.

I'm not talking about categories of play, or the four different kinds of player, or whatever other academic nonsense somebody has come up with today. Categorizing these kinds of things is worse than useless, because it implies boundaries where there shouldn't be any.

But it is worth remembering that your game is not a collection of rules and art. It's an experience that many different kinds of people are going to live through.

Thursday, August 09, 2007

Generative History and Ecologies of Sentient Beings

These days, games are getting larger and larger. Some games have many square miles of terrain and city. Sometimes, this is generated by hand, sometimes, it is largely generated algorithmically.

The problem with these large settings is that they ultimately end up feeling very empty. If there are people in them, they say the same boring non-things that the people in other places say. Also, if generated algorithmically, there's nothing particularly entrancing about the topography of the area, either.

Creating vast amounts of content has become one of the major challenges in modern game design, and most teams try to restrict themselves to well-known kinds of content - graphics, sound, levels, quests. These are expensive, but tools exist to help you create them effectively.

I think that, before too long, we'll have tools for creating other kinds of content, and one of the kinds of content I want to see more of is generative history/ecologies of sentient beings.

Whenever you see ecologies implemented in a game, it means that deer eat grass and wolves eat deer and adventurers eat wolves. It's simply a method for providing content which makes sense and adapts slightly. The adaptation quickly makes wolves either extinct or ubiquitous, but at least it's adaptation.

There are two problems with this approach. First, ecologies are really long-term things, whereas adventures are really short-term things. No adventure should feature the main character exterminating every wolf in the forest. Unless this is a MMOG, that's just insane.

Second, animals aren't very interesting. The difference between encounter wolves in forest A and bears in forest B is pretty minimal, and there's really not much punch to simply changing the number of critters you encounter. Also, unless artificially forced, there's no real balance: a low-level character can easily be attacked by a thirty-strong pack of wolves if your simulation is a bit more realistic.

What's more exciting are sentient beings and the fruits of their labor. These are also a bit more balanced, because the larger a group of sentient beings is, the less likely they are to simply kill off travelers.

The funny thing is that, on the most basic level, sentient beings are just as easy to simulate as flora and fauna. Sentient beings pop up wherever there are resources, just like animals and plants. They are also resources to other sentient beings, just like animals and plants.

Easy example: all people require food.

There is a gold deposit. The gold deposit has value, but miners require food.

There is a verdant valley nearby. People settle there because there is food. They are farmers.

Now miners show up at the gold deposit, and begin to send gold to the farmers in exchange for food.

This is much the same as "Grass grows where it is wet, deer eat grass, wolves eat deer". Except you're building groups of people.

Obviously, you'll want stronger rules that allow for economies, tech, trading, nations, conquest, and so forth. But the AI can be pretty simplistic. You can also have just one race, or multiple sentient races, or whatever.

Walking around in this world might not be any more entertaining, though. It would be busier, but is the difference between a fishing village and a mining town of much interest to the player?

Well, you can certainly add some easy things: the resources a place has to offer is based on what they do and who they trade with. But that's kind of shallow and ultimately uninteresting.

It's better to introduce history.

See, what makes things interesting isn't what's there now, but what used to be there. People aren't simply married. They are married because... they fell in love on a cruise? Their parents arranged it to unite their houses? There was an accidental pregnancy? It's the history of their marriage that makes it more interesting.

Similarly, a city bears its history on every street. This building used to be an opera house, but it went out of business back when this district got hit by that typhoon... this corner is the one where all the coffee shops are, because everyone walks by on their way to and from work...

History generally follows a few basic rules. An aspect can be introduced, grow stronger, or grow weaker. As an aspect changes, it causes at least one other change.

For example, a fishing village is introduced to the gold miners. The gold miners want food, and give gold for food. Gold gets stronger. Because gold gets stronger, something happens.

Maybe food gets stronger: people get into the industry because there is money in it. Maybe food gets weaker: the river is overtaxed and there aren't enough fish. Maybe a new facet is involved: textiles or banking.

This can be random, because any result can be explained with only minimal effort and nobody is going to be looking very closely. But it provides a lot of hooks.

For example, when you talk to someone, they'll be involved in one of the facets, and they'll know the history of the facet. "Well, my family's always been in the textiles industry, but the market is falling off because the gold mines are dwindling... so I'm thinking about being a fisherman. Or maybe going on an adventure."

Cities would grow organically. The docks are all fisherman's docks, but then the city grows and they are replaced largely with commercial freighter docks. You can still see the history of the city, in the layout and remaining fishermen's boats... a layout that is completely different from a city that has commercial docks without having a major fishing industry first. How fast the industry grew will also change the way the docks expanded: more rapid growth means more aggressive, larger, often slipshod construction.

With the right algorithms, this could theoretically automatically build the entire level for you. But in practice, it would still be human built... however, the humans would be able to say, "Okay, this area is mainly a commercial docks area, but there are a lot of old fishermen still around, and the commercial docks are really just kind of tossed up..."

You can keep expanding, of course: international politics, technological development, factions...

It's also easy to do with individuals rather than cities, or scale in the reverse and do it with nations or even planets. The philosophy is the same: you evolve an ecology of sentient beings and have them remember how their ecology/economy has changed over time. NPC 30322 will have something interesting to say, and city 192 will have a unique and sensical situation. You can even fill in-world books with ACTUAL STUFF instead of empty pages.

I don't suggest this is the best way to do things. However, I think it might be a good way to "flesh out" your world. You might say that city 192 is being invaded by Martians and has no standing army. The ecological simulation fills out the rest.

Wednesday, August 08, 2007

Gaming Weekend

Sometime next month, my old college holds Gaming Weekend. They also hold gaming Fridays every week. And we're not talking computer games: everything at these events is a board, card, or tabletop game, with some LARPs thrown in for kicks. (I was upset when Guitar Hero made an appearance, even though it is a good game.)

This rare dedication to live gaming is, I think, why WPI has one of the best game design curriculums. Of course, that's not really saying much, but at least it's saying SOMETHING. MIT is also big into games, to the point where Singapore pays them loads of cash to teach Singaporian students how to develop games. Harvard is just getting into the act with a bang-up new game club...

But the big thing about these places is that they don't simply focus on computer gaming. Sure, they're all into computer games, but they have other kinds of games they specialize in as well. MIT tends to specialize in the totally hardcore games like assassins and go, whereas WPI tends to do a lot of LARPs and economic games.

Whenever I see these things, I think "This is important."

Because you can't learn how to make computer games by playing computer games.

You can learn how to make computer games by making computer games, but computer games are extremely hard to make.

On the other hand, game design in general can be picked up pretty rapidly via live games. Not only are they faster to design and test, but while you play them you are learning game design.

"While you play them?"

When you're sitting around a table with four or five other people playing a game, the rules of the game are not the only rules in effect. For example, while playing Bang! it is always assumed that John and I are on the same side, even though we have no way of knowing that. It's certainly not a rule written up in the game. It's not even a "real" rule, in that half the time it turns out to be wrong and we open fire. However, it does change how the game progresses.

Live games are full of this kind of silliness - these "meta rules". Boyfriends aren't likely to try to make their girlfriends lose. The guy everyone hates is going to be slammed regardless of how well or poorly he was doing. The house rules say you can't use Lucre cards. You know Bob's weaknesses because you know Bob.

Because these rules change with every play (even with repeated plays involving the same people) they offer players a huge, ever-shifting variety of rules. Each play-through is functionally a new, tweaked version of the game.

This is not something which translates well into even multiplayer video games. While some of it translates ("Hey, Vega's cheap! Don't play him!"), it's not nearly as deep or wide. It's a puddle instead of a river.

Also, I think that these short games (but not too short) are better for learning from than longer games. For example, weekend long LARPs and RPG campaigns are fun, but the final experience is not as focused or repetitive, so you learn less.

This is why I think that clubs like the assassin's guild and the Sci Fi club (really the "board games and BSG club") are critical to the success of the computer game sections of these schools. Most of the people I've met from MIT who are involved in games were part of the assassin's guild. Most of the good/popular designers from WPI are either Sci-Fi Society members or Game Design Club members - usually both.

So play them live games! Over and over and over.

Saturday, August 04, 2007

It Was a Very Boston Day

This is a very dumb anecdotal post.

Today was very hot. So I went down to Cambridge to see what there was to see. If that decision seems strange, it's because it was hot: my brain wasn't working.

The sights of Cambridge were pretty much normal, save for the shimmering in the air. But there was something new. The most phenomenal smell. Like rotting vomit. It was the most amazing thing, and so prevalent all across the square I thought it might be coming from my coffee cup. To test this theory I walked to the river, believing that if the smell cleared up, it was probably safe to drink the coffee. Again, it was hot: my brain wasn't working.

Western crosses the river twice on the way to where I live, meaning that it actually takes you to the same side of the river you started on. Instead of hiking the depressingly urban length of Western, I decided to follow the river. After all, it would take me back to Western in the fullness of time. At this point, the heat had learned self preservation and was largely trying to keep me off of Western so I wouldn't catch an air-conditioned bus and recover my sanity.

I actually am not sure how long the walk was. All I know for sure is that my iced coffee turned into hot coffee before the trek was complete. But it was evidently a good day for being outside: everywhere I looked, people were draped on the grass or squatting on benches, crowded into tiny but overpopulated parties or onto kayaks. I realized that the heat might be getting to me when I noticed that, instead of ogling the various fleshy bits that people like to irradiate, I was carefully spying on book covers. Looking for titles that would spark my interest.

A hula party and one wheat field maze later, the river came back to Western and I limped towards home. Again, I forewent the bus, because the heat still had its claws in my head.

But it could not prevent me from searching for liquids, my coffee having long since boiled away.

Did you know that the Target eatery charges the same amount for a hot dog as it does for a hot dog and drink? It's same with sandwiches.

Madness! Maaaadness!

So then I got home and wrote a novelette about the various theoretical approaches to succeeding in life. Because I'm such a good example, you see.

And, of course, every personal post on every blog ever has to have a cat in it. So, yesterday, I was in a place that had two cats in it. EXCITING!

This is now officially a cat blog. Until tomorrow. Until half an hour ago, as it's 12:30.

I apologize for this post.

It's hot. My brain doesn't work.

Wednesday, August 01, 2007

"True" Simulation

I have found that, once I tackle a subject for a few years, there is suddenly no "best" rule set. An example that is coming up more frequently these days: social simulation.

Most people come into the field thinking that they're going to try to discover some underlying truth, some supreme rule set which will be able to do all sorts of social things. Most people, when they first try, come up with some ultra-generic system of addressing some kind of "social atom", like "I'll track how much they like any given thing and how much those things like any given thing and then use a logic chain to determine what the reaction to any attempt to affect any given thing is..." and so forth and so on. The end result, when applied, is labyrinthian and never works.

This is true of other fields as well, of course. Newbies tackling any system tend to come in with a simplistic approach that will solve all the problems of the world.

But there is no "best" approach, at least in games and simulations. There are variety of approaches, but the "best" approach is whatever simulates exactly what you want to simulate.

For example:

If I approach a game based around a social mechanic, I don't think to myself, "what's a good way to simulate social mechanics?" Instead, I think, "What's a good way to make the player feel the pressure and rewards of obligations?" or "What will make the player feel the contrast between the rash relationships of youth and the staid relationships of older people?"

See, the theme of the game isn't "relationships". That's like saying the theme of a game is "blue" or "culture". Instead, the theme is something very, very specific. I generally use a comparison of some variety, because it gives me a basis for the game's core challenge.

Understanding the theme, you can then decide on the rules of the simulation. You don't need to solve strong AI for this, and you never did. Trying to put strong AI into a game is like saying that you're going to build a go-kart, but it's going to have a cold fusion reactor in it. Just build the freakin' go-kart.