The Thousand-Hour Sim, Part 1

One of the ideas I mentioned yesterday was that there's only a few genres where a large number of players (10,000 or more) will be willing to spend multiple hundreds of hours playing the game. There's a lot of games were a good chunk of the audience will spend around 100 hours. Many single-player RPGs have that much content, and a lot of multiplayer games will keep players busy for 100 hours.

But take a game like WoW - there are millions of people that put in hundreds of hours every year, and probably millions that have put in over a thousand hours over the six years since the game has come out. Some of the friends that I worked with averaged six hours a day - usually spending a dozen hours on the weekends and a few during the week. That works out over 2000 hours a year. Even if they burned out in six months, that's a thousand hours right there.

Competitive games, like Starcraft and Counter-Strike, are another genre where thousands of people might spend thousands of hours.

What would you have to do to make a sim that could keep at least tens of thousands of players occupied for a thousand hours?

Summary of the FMMORPG Argument

There were eight factors I mentioned when I described Fantasy MMORPGs as the ultimate genre. They are:

1. environment variety
2. tangible, human-scale environment
3. non-trivial gameplay
4. opponent complexity
5. opponent personality
6. solo and group gameplay
7. inspiration
8. open end-game

The first factor means having a wide range of environments. The second suggests using an environment that can be rich. Gameplay needs to be complex, and getting to a really rich complexity means not just having a complex puzzle but introducing AI units into the game that introduce complexity, keeping the game fresh. The fifth factor, opponent personality, is like the second: it puts a face to the game world that players can relate to; an environment that produces stories. The sixth factor is about allowing players to play whenever they want, while also giving them team goals, where they can work towards goals larger than one player. Inspiration (and something I didn't mention, commitment) are ways of giving players long-term goals to keep them coming back. The final factor says that the game can't end after a hundred hours; allowing players to stay in the same game world for the entire duration of their play-time can keep them there.

The factors can be broken down into a few threads. I'll probably redo this list at some point. Obviously, this list isn't definitive; I'm building a model for game design and there are surely many other models just as suitable. Anyway, the points I'll cover today will be:

• Gameplay needs to be rich
• The game world needs to be rich
• The game has to enable stories and emotional moments
• It should allow the player to play, and make progress, whenever they want
• The game should provide large goals, requiring time and/or teamwork
• The game should have an anchor that commits players to the game

Some of these feel essential to me, others feel like caveats.

Positive Attributes

I think some of the attributes should influence the core game design. The game should be built around these principles. These are the things you should be trying to do when you design the game. The negative attributes, which I mention later, are a list of things you link at and think, is there something in this design that will drive away players? I think the negative attributes can be tweaked - that once the core game mechanics are nailed down, the next step is to go through a list of "gotchas" and make sure there's nothing there that would piss off players.

First I'll tackle the attributes core to game design.

Gameplay needs to be rich. This is the single most important attribute. Without rich gameplay, the game becomes rote quickly, and play-time becomes boring. I've talked about the essence of fun before, and my theses on boredom is that it happens when the player doesn't have to think about gameplay; if the time the player spends thinking about decisions is too small. As complex as the game might be, if the player has mastered the game rules then there's nothing more for him to learn.

Take a game like Halo. They introduce new mechanics throughout the game and, by the end, you've explored all of those mechanics. Once you've beaten a level, there's no reason to go back and play it again, except maybe to find a few hidden secrets or complete it faster. New sequels really just introduce a larger game world; they don't add new gameplay elements. Players spend 20-40 hours playing through the game, and then wait two years for the sequel, then spend another 20-40 hours.

Compare this to Grand Theft Auto. The worlds are about as large, but there's a richer variety of things to do. As a result, players can spend 100 hours in the world. There's only about 100 hours of "quests" in the game, though. If they added more quests, would you keep playing? Probably not, because without more mechanics, execution of those quests becomes easier and easier. As you play through the game, you pick up new weapons and learn new game skills - but you exhaust those skills by the time the game's done. That's what I mean by new mechanics: there'd have to not just be a new place to play (ie a larger world, such as in the expansion packs) but new ways to play.

The game world needs to be rich. I touched on this above. The world shouldn't just be large, but provide new environments. There are over sixty different zones in the world, and each one has many different sub-areas. Plus well over 100 different dungeons. The starter zones can be plowed through in under an hour, but by level 50, there's at least five hours of quests in each zone. The dungeons provide a couple hours of gameplay but can be replayed, too. Each of those zones and dungeons has its own creatures; usually dozens in each zone, and each creature type (like, say, furbolg) has several variants (such as warrior and shaman).

That means a lot of art assets, of course, but that's what "rich world" means! Players can easily spend ten hours in Stranglethorn Vale. Although the whole zone is jungle, the land ranges from the beach, to the river, to hills, to villages, to caves. Creatures include pirates, goblins, trolls, panthers, tigers, raptors, basilisks, and ... much more. Each little area has its own story (told through the quests) and there's a few greater storylines that take players through the zone.

Worlds are rich because of the variety of their assets and they stories they tell. Speaking of which...

The game has to enable strong emotional moments. Stories are a player retelling of emotional moments. The strongest emotions are fear, anger, and love. WoW doesn't really want to anger its playerbase, but anger does have a greater place competition-driven games (and therefore in the PvP aspect of WoW). Fear is the most common, but it's not often remembered as such. The most memorable moments for me in WoW were when I was running in fear from high-level enemy players, or in instances when our tank died and I was panicking trying to keep the group alive (I played a healer!), or close calls when I got attacked by several creatures and was running for my life. When these encounters ended well, that fear turned to extreme relief. I wanted to tell everyone I knew about what just happened, and the story got retold the next day at work, too.

Strong emotional moments make the game into a larger part of the players' lives. When we take those stories to work the next day, to the pub, into our dreams, its more than just a way to pass the time. Am I going to tell stories about that one time in FarmVille when I planted some corn? Ugh. We hate the boring coworkers that tell that kind of dumb story. NO. Yeah, sometimes those WoW stories are the same - but not always.

Players can feel love, or something similar, in these games, too. Probably a bit more common is respect, and dedication. Players in guilds might feel commitment to their guildmates, and want to show up on raid night not just because the raid will be fun but because they want to help their guild out. It's really cool to have a good tank in the group, or to have a great healer behind you, or an awesome DPS guy that is super-great about handling adds. I really respect those players and their skill with the game, and that generates stories that I want to tell, too. (But god I hope that raid is fun, too. See section one: gameplay should be rich.)

Curiosity can also be a strong motivator. Curiosity drives suspense novels and thriller flicks. It's usually not something that the designer would want to drag out for dozens of hours, but there are a lot of cases where a bit of curiosity keeps me playing for an hour, or to keep coming back to a quest line for an hour of play every so often. Some quest text can introduce a mystery, and either dole out new bits of story in each quest or give the player a chance to go exploring and find out the solution for themselves. This is a kind of meta-game; the game proper is combat, but if you've got a player hooked on a mystery they'll play through the combat to find out who done it! Yet another way of making gameplay richer, and in this case it works because of the emotions that the game makes players feel.

The final "core principal" is: The game should provide large goals. One of the things I've bitched about with browser games is that they don't feel like there's a larger goal. Travian actually has one, and it's a very appealing goal to me, but they did a crappy job of advertising that long goal. I played Lord of Ultima for about a week but quit mostly because I felt like I had no goals. Sure I can sit there building my city up, but who cares? Compare that to Nile Online, which I'm still playing: the goal is to get to a high enough level that you can "retire". When you do, you get a pyramid built and named after you! At that point, the game is "over" and one can leave the game, but at least it's a motivation to keep playing.

MMOs provide many large goals. To get to max level, to get into a raiding guild, to get to the hardest raid instance, to get awesome gear. Even just to complete the easiest end-game raid instances is a large goal, taking weeks of play to prepare and attempt. Single boss fights used to last a half hour or more though these days they tend to be 10-15 minutes. Each fight might take a few attempts to master, and there's around ten bosses in each raid. Throw in trash encounters (which can be tricky to master as well) and "raid night" is several hours of gameplay each week.

Accomplishing large goals tends to drive other gameplay. Preparing for raid night might mean running dailies to get cash to buy a new piece of armor, or to get enough faction to be allowed to buy a new piece of armor. Hopefully those dailies are fun and don't get too repetitive, one of the few places where I think WoW falls down. But consider this like curiosity: players will engage in hours of gameplay (hopefully fun by itself!) in order to achieve a meta-goal. They're not in combat to kill the creature, they're not killing the creature to gain rep, they're not gaining rep to buy new armor -- they're buying armor to make more progress on raid night. And they're doing that for bragging rights and to support their friends.

Non-Essential (Negative) Attributes

One can probably come up with a huge list of  rules saying "don't fuck up your game by doing X". I happened to extract two from my previous post; there's dozens more that I've thought of since (another post?) but I'll cover these two in some detail to explain what I mean.

Players should be allowed to play whenever they want. If your basic game mechanic is combat, then what would keep them from playing when they want? I heard EQ basically couldn't be played solo - which meant, if you wanted to play, you've got to wait around to find a group to join. One of the things that I think made WoW wildly successful is that it doesn't have this restriction. If you log in and want to play, there's always, always some solo task you could do. You can hop into a PvP battleground, go on a solo quest, do a daily, farm some crafting resources, or find a PUG instance to run.

This principle is better expressed in the negative: don't add anything to your game that would prevent players from playing when they want to. Not being able to play is frustrating, and one of the things in browser games that piss me off. I'd enjoy them a lot more if I could play at my own leisure. Having the game say "sorry, go away, we don't want you" makes me feel bad things towards the game and that's a great way to lose players before they hit the 1000 hour mark. And it kind of means that they can't hit 1000 hours. It's basically admitting that you don't even have 1000 hours of stuff for them to do!

I'm probably going to play Nile Online til I get into the top 100 (and then retire), but there's no way I'll put 1000 hours into it - it just won't let me. It would take me years of play to get that far, and I'm far more likely to find something else to do. If I can retire after six months, what's to keep me coming back again and again? Hence:

Players should feel committed to the game. One thing that kept many players subscribed to Ultima Online was their houses. They had a lot invested in their house, and they didn't want to give it up. This is a great example because a lot of those subscribed players didn't even play! They'd log in once or twice a week to prevent their house from falling apart, but they wouldn't actually play. Imagine how many more players they could have kept around if there was something else to do?


But really my point is that commitment itself is strong enough to keep players subscribed.


Summary


So I didn't even talk about sims at all in this post, and I've been hacking away at this post for hours. In part two, I'll explore how to apply these rules to the sim genre and see what I come up with.


Meanwhile, I'm gonna go pass out... zzzz

Gameplay Complexity and Opponents

Yesterday in my praise of the fantasy MMORPG as the ultimate genre, I touched on the notion that a game that involves human-like opponents has a lot of natural potential for rich gameplay. Today I want to expand on that a bit.

Game genres range from racing to shooters to RPGs to slot-machines to sims to platformers and on. Multiplayer games obviously pit you against other humans. Some of them have single-player variants, where the part of the enemy is played by AI that tries to act as close to human; ie, the opponents have the same range of actions and strategies as humans do. Others pit you against a puzzle, or simplified enemies with very constrained actions.

Puzzles - No Opponent

I'll start with the last. Puzzle games like Tetris don't have "opponents". It's just you against the puzzle. Simulations, including games like SimCity and Roller Coaster Tycoon, pit you against a complex puzzle. Simple puzzle games have simple mechanics; the complexity of the game is how the rules fit together. Some such games are like skill games - gameplay at the higher levels of Tetris require the player to have internalized the gameplay rules effectively to the point that it's like a sport, and where reflexes and the ability to think a few moves ahead are important.

Skill games are rarely complex. Coding their game rules is simple. Design can be difficult, however; the process of coming up with rules that are amenable to skill is tricky. Back during the golden decade of the FPS (the 1990s), many game companies tried to make competition-worthy FPS games, but most fared poorly. For a single-player game, something that's going to provide 15-40 hours of gameplay, the game mechanics don't need to be very rich. Elements that would lend themselves to longevity as a skill-based competition are hard to design in, and the economics of the game industry often mean that the people that would have experience with competition-style gameplay aren't the ones that are chosen to lead game design. It's one thing to say "we want a lively online community" and another to actually spend money and effort finding a designer that can do that well. There's a rich discussion there but I'll skip it for now.

Because my point is: the complexity of these games is usually the intricate ways in which a small set of game rules combine.

Could you turn a sim into something players would spend a thousand hours playing? How long could you play Roller Coaster Tycoon before you exhausted the whole game? Do you think you could convince a million players to each put a thousand hours into it? These games (on the PC) are lucky to sell a million copies. Most of those players might put less than a dozen hours in.

What would it take to make a sim or puzzle game that players spend that much time on? It feels silly to even think it's possible, since one hundred hours is a stretch for these games. Maybe there's some diehard fanatics that put that much time in (the Transport Tycoon community is one such example), but that's a tiny community. That's maybe a thousand players - not a million. I'd like to explore this idea further... and indeed I do in this followup post on the thousand-hour sim.

Platformers - Fake Opponents

Platformers have opponents, but they aren't anything like human players. Enemies such as the goombas in Mario follow very rigid rules. They don't even try to behave like human players. That's not the point; these are effectively puzzle games where part of the puzzle is an AI-controlled "humanoid" character with simplistic, predictable rules.

Usually there's not a lot of richness to these games. They're exploration and achievement games, with a little bit of skill thrown in. Gameplay is fun, but that's driven in part by the environments, not the opponents. The human (or sentient) appearance of the opponents just fills out the game world; makes it seem like a real place and less like an abstraction.

Recent Mario games (such as Mario Galaxy) do wind up being somewhat rich, but not too much. Most of the richness is in carefully timed execution of special moves. In other words, they are skill games. Puzzles are straightforward, the interactions of gameplay elements are rare, and the focus of the game is more on exploration than anything else.


Head-to-Head Action and Strategy Games

Whereas there's no real "opponent" in puzzle or sim games, and simplified opponents in platformers, other action games have real humans as the opponent. RTS games like Starcraft and FPS games like Counter-Strike fit this model. Even when you're playing against the AI, that AI is trying to act as human as possible.

Both you and your opponent have the same range of abilities. Chess is a simplified version of this model, where both players have exactly the same units at their disposal. FPS games are similar; often skill games (see above), but the richness in the game comes from the fact that the opponent is truly human and so not only has a huge bag of tricks but can also learn new tricks as time goes on.

RTS games rely heavily on the humanness of your opponent, but they often twist gameplay by allowing players to choose different "races." Starcraft is the standout in this model. Most Starcraft games are played by players of different races; Terran, Zerg, or Protoss. They are all basically doing the same thing - placing buildings, recruiting soldiers and ships, and sending them off to attack. Yet the units that they have are different, and their overall approach to winning is different as well.

The richness in FPS games comes primarily from the human opponent. The fine balance of rules can make the game more or less balanced; more or less random. Yet it's the humans that make them great.

RTS games, however, add much more. There's a lot more strategy to RTS games. Choices made early - not just single actions, but choices to develop in one way or another - define the game. Do you try to tech up? Rush? Expand early? Press to aerial units? Fake one build but go another way? Where FPS games often have rich tactical fights, RTS games also have strategic fights.

RPGs

RPGs are a bit like RTS games, in that you are facing human-like opponents with skills and actions similar to your own, but with a different strategic twist. In Starcraft you have Protoss vs Zerg, in WoW you'll have a Rogue against a Warlock.

Humans provide the richest opponents. Some are better than others; in RPGs, you might face someone that doesn't know their class very well at all. But most of those that engage in PvP know what they're doing. Competition ladders help great players find similar opponents, while allowing the crappy players a chance to win every know and again against other new (or... well, crappy) players.

RPGs add more to the single-player experience, however, by having a much longer stretch of new abilities. In an RTS, you'll become familiar with all of the units and buildings at your disposal over the 30 hours or so in the single-player campaign. In an MMORPG, there can be hundreds of hours of advancement before you've gotten to the top of the ability tree, learning new spells or abilities every few levels.

Instead of giving you a small set of tools with rich interactions, like chess, RPGs have a much larger set of tools, each with its own special purpose. The trick isn't to learn how the abilities interact, but usually which ability to use when; how to fight against certain types of opponents. The huge array of possible opponents is what makes the RPG rich.

Human-like Opponents with Different Tactical and Strategic Choices

My goal here was to explore the game design choices that produce a game that players could enjoy for hundreds of hours. Leaving aside the diehard fanatics that might stick to their favorite title long past when the majority of the market has gotten bored, we're left with three major avenues.

The first approach is to build a game that requires player skill, whether solo or competitive. Players spend their time getting better at those skills. Games like Counter-Strike and Tetris are the best examples, but even then they point out the weakness: although both have audiences larger than (say) Transport Tycoon, these are still relatively niche titles. Lots of people love them and (relatively) lots put in over a hundred hours playing the game, but there are bigger genres out there.

The second approach is to create simple mechanics with extremely complex interactions. Chess is the canonical example. Players have a chance to learn ever more complex interactions. Every time they play, they probably learn another new trick. The success of this genre depends on the size of the market, however. There can't be hundred of Chess-like games, because the complexity of interaction depends on the size of the market. Not everyone has the talent to be a grandmaster, and in order to even get grandmasters, you need enough people to push the curve that far.

The third approach is pitting player versus player. Whether you've got a skill-based game, one with simple rules but complex interactions, or complex rules, pitting humans against each other means that rich gameplay will evolve. You'll either need to do a whole lot of playtesting and balancing or be willing to tune the game after it ships, but with PvP, humans will figure out complex strategies.

Personally, I think the two richest game genres are the RTS and the RPG. The best, longest-lived games in both genres have PvP (sometimes simulated), complex gameplay interactions, and a measure of player skill. If you want to create a game that pulls in players for hundreds of paid hours, you'll need more than one of the approaches.

Designing an MMO part 1

Designing a New MMO, Part I: Get Rid of Classes!

Everyone wants to make an MMO. They're fun games, and with WoW pulling in over a billion dollars a year it looks like an insanely lucrative market.

Except, of course, for all those failures. Like Tabula Rasa, which cost a hundred million to make and only brought in a sixth of that in revenue. Unless you've got 84 million dollars you don't mind never seeing again, jumping in should be done with care.

I like thinking about MMO design. I think it's like talking politics. It's not like me and the rest of the crew at the water cooler are going to run for office. Ultimately, the only effect each of us has is one vote -- out of millions. Does it really matter what I think about politics? At most, I'm influencing a dozen people. And I haven't yet converted any of them to the One True and Proper Political Party, so what does it matter?

It doesn't matter. It's fun, though. Likewise, us scrubs talk MMO design. It's an entertaining exercise.

Usually the first topic to come up is something like "classes are lame" or "levels are boring". I think this is a fairly fundamental discussion.

But I'll skip it, because Lum did a much better job than me. Go read that.

(There's no guarantee that I'll ever write a part 2.)