Balancing by "not balancing"

Run with me on this. I just finished reading a great article on the Elder Game blog about how balancing in an MMO is ultimately a fruitless task. Then I thought back to another post I had read a while ago about the idea of Self-balancing systems. I have to come down on the idea that you simply can't balance everything. A truly "self-balancing" system would be probably more complex than the game itself.

But I did have one thought on how to try to... offset some of the players' expectations. I while ago I was thinking about how some many players complain about the random nature of a lot of MMO abilities. they all do random damage, and many have a random chance to "crit". When you get blown up by some mage who hit the jackpot in the lottery on his numbers and insta-gibbed it's pretty frustrating. Same for that mage who won't have that happen again for a week or a month or never. So why not take more of that randomness out of the random number generator, and put it into the environment?

What I mean is, make the effectiveness of abilities dependent on where and when you're fighting. If it's night time, an archer isn't going to hit quite as often, and maybe will get a critical hit even less often. During the winter, a character using frost-based magic will be a little stronger, while during the summer a fire-based character will be stronger. If there's a storm where you're fighting, maybe no one can get a critical hit because the wind is just blowing everyone around.

Not only natural things, but perhaps add some more supernatural effects as well. Deep underneath the ground are magical "Ley Lines" that follow paths that no one knows (but perhaps could figure out), and near them perhaps magic is stronger, or perhaps just more random.

Over time, the player community will figure out all these factors, and players will begin to consider their surroundings: "Oh why did I stupidly try to shoot down that warrior in this nighttime windstorm instead of just running?"; "There should be a Ley Line over there, I'll wait to ambush the caravan there."

The idea is that with a world that creates enough dynamic effects on how powerful each class is, players will notice a little less how "overpowered" certain classes or abilities might seem. Obviously this doesn't remove the need for balancing, but it might just lighten the load a tiny bit.

Quests as a tool to encourage interaction

My experience with quests, and indeed most people's experience with quests in MMOs to date, was pretty bland for the most part. Sure, some incremental progress has been made beyond "kill ten rats", though those quests are still around, since they make killing ten rats feel slightly more productive. There's the occasional quest that's actually enjoyable; sometimes it's the unique mechanics, like driving around in a siege engine; sometimes it's the story, like the Wrathgate questline in WoW.

However, quests are all game stories, without exception I believe. They were all stories that are built into the game, were written by someone else, who tried to make them in such a way that you wanted to know them.

I have told stories about playing WoW many times (though usually to other people who play WoW), and I'm sure most people have a few good stories from playing an MMO. However, I'm willing to bet that none of those stories are a quest story. Oh, they might be a story that happened while doing a quest, but they won't be the quest itself. Why retell something that's been told to literally thousands of other players in exactly the same way? My favorite stories are all of times when I did something cool with other players. The time my guildies and I ran Stratholm deadside in 45 minutes in blue gear. The time myself and 14 other PvP-focused players running a premade defeated the strongest guild on our server in a hotly contested AB battle that ended 2000-1990 (vanilla WoW; the guild was in T2.5 and we were in PvP blues and epics).

You know what I conclude from that? The best quests are the ones that make you interact with other players, then get out of the way. I have virtually no memory of the story behind the quest to run Strath deadside in 45 minutes, but I remember many details of the run itself. I may not remember many of the details of the 2000-1990 AB fight, but I do remember the desperation and then excitement upon realizing we might be able to come back from a deficit and win.

I'd like to propose an example quest that could do this. Imagine a quest in town to escort an NPC to the next town, perhaps a 15-20 minute journey. The quest is only offered once, and when someone accepts, no one else is allowed to sign up, similar to some escort quests in WoW. However, imagine it's an open quest; you can optionally share it with as many people as you like. However, sharing it may reduce the reward slightly, the idea that the payment would be divided among several players.

Imagine, next, a corresponding quest in the nearby enemy faction hideout. At the same time that the escort quest is accepted in the nearby town, the hideout has an NPC start offering this quest to any single person interested to assassinate the NPC who is being escorted. This player can choose to share the quest with others, with the same sort of minor penalty. Alternatively, maybe a standard Wanted poster is up with a reward for any player who turns in his head.

The point here is to create a situation where players have a reason to attack each other beyond simple "ganking". With a structural reason in place, preparation can come into play. A player escorting an NPC with a bounty on it's head knows that there's a high chance they'll be attacked. This creates a more equal ambush situation, because the player is expecting it, though he still might be taken by surprise. Also, there is the interesting decision of whether or not to bring along help. The escort player may opt to not bring anyone, on account of the zone being under his faction's control. However, he knows that an enemy in his own zone might be able to kill the NPC before he can stop them. Similarly, the enemy player has to decide whether to bring along friends and increase the risk of detection by random players, or go it alone and risk simple failure, or risk being outnumbered by the escort.

I'm sure there are other many ways to vary this sort of a quest, including ways to make it more attractive to pursue. However, I think the general theme of structuring quests in a way designed to throw players together and at each other needs to be put to more extensive use.

A Game Built on Contingency

What if you had a game world that had a sort of structured dynamism? Sort of a mix between WoW, Eve, and D&D. I've been working on a sort of sketchup of an MMO, but I finally figured out how to put into words one of the main themes I'm thinking hard about for it: contingency.

In D&D, the Dungeon Master, the guy running the game, often has a primary quest for the players to do. However, often the players don't do what he expects. At this point, DMs do different things. Some improvise a new storyline, some find a clever way to coax the players back on the planned path. And some have several potential - contingent - storylines planned out that the players could follow, depending on their choices. This is the idea I would like look at: the idea of planning multiple outcomes in a game world, each one different, but all equally meaningful and interesting, and then letting the players on each server decide through their collective actions which outcome will take place.

~*~

Imagine a game world; a relatively small game world, say half the size of one of WoW's continents (and we'll base the general game mechanics off WoW to keep things simple). Maybe it only holds 1-2 thousand players; that's fine. Now, this world has several zones like WoW, but its focus is on PvP. The Horde and Alliance each have a single capital city, maybe 3 or 4 major towns, and a dozen outposts. All towns are potentially conquerable, including the capital cities; maybe the outposts can be too, but they're less strategically and economically important.

I'm told this is similar to WAR, except it's Order vs Chaos, or some such.

Now we'll put the economic system into a little perspective. Imagine that among the various resources, several are only accessible in certain zones and locations, and that each faction has control of some of those resources. Both players and NPCs can take advantage of these resources to various effect. However, there are no flightpaths, and the only mail that can be sent is text, no items. This means that resources have to be actually transported from site to site.

Some of you may be thinking of Eve right now.

Now, on top of this nice little conglomeration of features, lets add a little basic "dynamic content": natural disasters. Imagine landslides happening sometimes in mountain passes, blocking the road. Fortunately, there are other passes, though perhaps they only opened up when the landslide blocked the main one. In the forests, one day the road gets swallowed by the undergrowth or is blocked by a gigantic fallen tree, forcing players to find new ways through. Maybe a blizzard blows through from time to time, or a hurricane or tornado during a battle.

Sure, some of this may have been thought of before, but the point is to have these things happen often enough that players recognize that the world can change any time, often in inconvienent, but interesting ways. And most of all, these events affect things; they're not like WoW's weather. Another example: fire magic works less well in a downpour, while frost magic is amplified during the winter (to balance, seasons could last for a week or two; a game year might take a month in RL).

~*~

Now lets go back to that PvP stuff. Say the horde conquer some alliance town, cutting off access to a number of outposts, and putting that zone in the hands of the horde. That means the alliance is now excluded from that part of the world, right? Wrong. While the horde are busy rebuilding the burnt-down town with their own architecture, the alliance NPCs escaped during the battle, and hastily construct a hideout in a nearby forest or ravine. Now instead of offering quests to deal with local wildlife and attack horde towns, they offer quests to assassinate horde peons working and transporting supplies, and steal supplies. Meanwhile, the horde moves in some NPCs, some of whom give the horde players missions to escort various NPCs that the alliance has a mission to assassinate.

In other words, the destruction of the alliance town doesn't mean the alliance players can no longer play in that zone. On the contrary, they're encouraged to go to that zone and play more there than elsewhere. On top of that, they're given certain things to deal with the dominance of the horde in that zone. Perhaps they get a cloak that can be used to shadowmeld anywhere in that zone, to reduce the advantage of the winning faction.

What's that? The fact that the alliance didn't lose anything means that destroying the town was meaningless? Let's look at that in more detail. The horde now have access to more resources, more NPC support, and can act more openly in that zone. And what exactly is "meaningful" anyway? Excluding or killing the other faction? That's a rather 1-dimensional point of view. This is meaningful because the world changed. The town changed hands, the horde got new quests, and the alliance got different quests. Instead of being simply forced out, alliance players got new and different challenges, but also new tools to deal with them. And by PvPing, and winning, the horde literally changed the world; that's pretty meaningful.

Then if the alliance takes back the town, the horde retreat to their own hideout to formulate a counterattack of their own.

This is what I mean by contingency. From what I've heard of WAR, if you take a keep, it's taken, and the other side has to take it back. That's pretty simplistic. Heck, in WoW, if you destroy all the NPCs in an enemy town, aside from the fact that they'll respawn eventually, what you're doing is "denying service" for the other faction. Imagine instead that another nearby town immediately sends out an expendition of NPCs to clear you out of the town, and rez all the NPCs. Then you have the option of running away, or trying to defeat this focused and more dangerous challenge.

In other words, when the players act - in sufficient numbers - they force the game world to change. However, it doesn't have to change in simplistic or even predictable ways. Imagine setting up potential events where a powerful wizard happens to be traveling through the town at the time and summons a literal thunderstorm to repel your attempted invasion.

That may be a bit out there for now, but that doesn't mean it wouldn't work. Or make the town guards more alert and ready for a fight during the night, to discourage attacking a town when few players are online. But the important thing is to allow for one side or the other to win, and then change things based on that. And continue changing them.

Rather than building the world around static stories of quests and epic raids, build the world around a set of comprehensive ways for it to be changed. Create a structure for this to happen, automate it to a high degree (perhaps allowing for some minor intervention in extremely imbalanced cases), then after release, spend your development time on expanding the different possible contingencies of conquering and reconquering zones, on expanding the variety of quests pitting players against each other, on expanding the ways that travel paths change, on creating new random events to change the world on its own, in a natural way. In other words, continue expanding the potential contingencies and outcomes, to reduce the chances of the players encountering simplistic repetabove his headme. If the world is constantly changing in little ways all the time, it doesn't become boring. If I don't know what quest I'm going to get when I talk to that NPC the fourth time, if there's a chance it might be brand new, I will feel a small bit of excitement everytime I see a new exclamation mark above his head.

Themepark vs Sandbox? How about DM?

Raph Koster wrote about an interview that GameSetWatch had with some of the developers of Red 5 on their up and coming MMO, which I guess has the dubious aim of rivaling WoW (they've got something like $20M in venture capital backing them). The interview is mostly about Red 5's aim of creating a world that changes based on how the players act and behave. The popular metaphor they toss around was "save the village, and it'll stay saved". I think that's a rather bad metaphor for a good way of having a world where you can have an impact, but it does sort of get the idea across.

The feeling I get from Raph is that he's rather skeptical about just how far Red 5 is going to go, and about how effective they will be. And obviously the guys in the interview are extremely optimistic about what they're talking about.

I have to say, I like what they have to say, particularly the way they liken it to being DM in D&D rather than being a writer. Of course, Raph has his point in that at the end of the day, you have to find some way to actually implement your plan. He lists a couple of examples of attempts at creating processes that imitate dynamism:

"There’s a zillion things that have been tried along these lines.

• Spawners that grow in power and overrun stuff.
• Spawning spawners.
• Player cities and housing, which are persistent elements affecting the landscape.
• NPC factions with tilts based on user actions
• Keystone quests that affect regions of the game, spawns, or large-scale events
• Territory control games.

These things all create emergent narrative as users change things."

Some of those sound like some pretty interesting ideas. I wonder how many were really tried, and how successful they were. Being rather inexperienced with MMOs in general (I unfortunately, only played WoW, and have no intention of starting to play anything so time intensive right now), I really don't know what has been given a good chance at success. Especially since there are many games out there that have brilliant ideas, but poor execution, which often sentences those brilliant ideas to languish in purgatory until someone finds them.

The thing I'm noticing with these is that they seem... piecemeal. Maybe I'm being overcritical, but either way it's a good excuse to talk about it.

Several items on that list are simply slightly more abstract versions of processes that already exist. Spawning spawners, or growing spawners? More clever than simple spawners, but still predictable. NPC factions and territory control? Better, but basically just a tug-of-war, and king of the hill. Player housing is basically just player-generated content that no one else cares about.

Now, "keystone quests" that affect regions of the game, along with another example he gave of spawning complex quests that actually change the region around them. Those both sound more interesting, though they still have issues in my mind. One, quests that affect regions have the issue of not benefiting very many players. The spawning of more complex quests was not implemented; it sounds like that was because it was a bit too complex to implement on a reasonable budget. In other words, one doesn't have enough "payoff" in how many players get to play it, and the other has too much of a "cost" to implement.

But I still like those two better, because they try to combine more dynamic processes with narrative.

I would say that right now, WoW is probably the best at implementing narrative sorts of things in MMOs. Their quest system in WotLK is finely tuned, and combined with phasing, makes individual players feel special. And obviously their raiding game is top notch, also containing excellent "procedural" content.

The downside to their methods is that it's all based on illusion. The entire world is ultimately static. The least static part would probably be the Lake Wintergrasp zone, but the way it's set up is designed so that almost every battle that takes place there results in the attackers winning, creating a cycle of conquer-reconquer that results in an overall static situation.

The irony here is that against this procedurally static backdrop, there is the extremely dynamic social scene. Guilds forming and reforming, groups and raids running dungeons, friends being made and rivalries forming. In other words, the people playing this game are engaging in social activities that are largely unstable and very dynamic. Communities tend to spring up around every aspect of the game.

So, what if the game world only existed because of those dynamic communities? You'd probably get Eve. What if instead, the game world changed a little to reflect the activities of those dynamic communities?

Meaningful PvP

I guess I'll be starting out with an MMO post. Ironic, considering my previous blog was supposed to be about games, and I switched to this because I wanted to write more diverse posts.

Tobold's thought of the day is "Meaningful PvP". His initial thought is that meaningful means it has to hurt someone, otherwise it doesn't matter to people. This led to an interesting discussion in the comments, and I found myself agreeing with a couple people, including Garumoo's blog post. I ended up writing a ridiculously long set of comments in response, but I decided I'd repost them here, since I like the thoughts so much. I'll be expanding on them more in the future.


Skat, Soccer, Badminton, ludo, chess, strategy games,

Sure, they reset, Tobold, but I think the complaints of some people is that they feel like they're in one giant game of chess, and that someone's reseting the game every time a single piece is taken. That might be a better metaphor for that point of view.

I'm with the commenters pointing out that meaningful pvp doesn't have to be "if you lose, it hurts a lot". Obviously you have to have some penalty for failing on a small scale in any game; you even lose some time in WoW if you die. A good example of one of the times when it hurt a lot to fail in WoW was the suppression rooms in BWL in vanilla WoW. Anyone remember that? Having to start over an hour-long aoe grind if you fail on the boss was very painful. Obviously, they don't do that any more.

However, I feel that the MMO community is stuck in this either/or idea: that you either have losing hurt, or your pvp is meaningless. You just have to get more creative - and more convincing - in making pvp have an impact that is different from simply forcing the other player to lose.

Now, I don't know if it's possible to do that on a 1v1 scale, and frankly, I think that'd be a waste of time to try, since this is about massively multiplayer games.

I think Garumoo's blog post is brilliant, and is rather close to what I've been thinking about the past few days. What about making a game world where two factions fight for dominance, and it quite possible for one side to win and take over a zone (and potentially the whole world), but when they take a city or town or mine, the other faction modifies it's tactics and moves it's base to more stealthy locales? For an easy example, imagine Undercity getting conquered by the humans, and the forsaken take to deeper tunnels in the sewers, staging sneak attacks on the now conquered territory.

Furthermore, when a zone changes hands, change the zone! Change the town to follow the other faction's visual theme; change the quest givers and their quests; change the layout of the town. Thanks to phasing, the technology is probably ready. Plus, if you can see the effects of doing quests, even repeatable ones become a bit more exciting. Take the whole set of quests for the buildup to war against AQ in WoW, and make it small enough scale to fit just a town, and you have people actually seeing the effects of their contributions.

And don't forget the "losers", since they're the ones we're most worried about, right? Give them a secret hideout in the woods next to town, perhaps some secret passages into town, or to out of the zone. Perhaps the hideout contains a vendor who sells magical camoflaging cloaks that allow the user to do something like Nelf shadowmeld only in that zone. Instead of the original quests, build quests around the npc's interrupted plans. Have quests to go out and assassinate enemy npcs, or kill a couple enemy players.

On top of that, perhaps have a quest to kill an enemy player and steal the payload they were carrying for a quest in their new town. Enemy player collecting wood for building new guard towers? Kill him, loot just the wood off his body, and turn it in to build your new hideout. And there's another way of making the losers lose in a meaningful way, without making it ridiculously painful. And if you object to assassinating players, remember that those are quests to assassinate players of the winning faction. Set the rewards right, and as soon as a zone falls, the losing faction will have lots of other players pour in to take advantage of the new quests.

This does have the danger of becoming a world of a dozen different Lake Wintergrasps. In other words, a system where both sides simply switch places regularly, because it's the most rewarding. So you'd probably want to try to strike a balance in rewarding the changing of hands. However, if you set things up in such a way that zones take at least a few days to change hands, I think it would be substantially better than wintergrasp.

Ultimately, this is about the players shaping the world. Not in its entirety, but it'd be a great compromise between WoW and Eve, so to speak. Imo, Eve has been good at letting players shape the game such that the game feels real and dynamic. WoW has been phenomenal at creating the illusion that the game is real and dynamic.

I remember questing in the Grizzly Hills in WotLK, and seeing a herd of horses running around, actually acting like a herd, running in a group. But now I think, if I could, for instance, attack that herd, perhaps with a group of teammates, and drive it through an enemy town to cause distraction and damage, that would be a step up.

I read about Cataclysm, and think that the changes are cool, but then I suspect that the change is ultimately going to be a static change, and eventually everyone will get used to it, and that's a problem. I think about the zombie plague that happened for about a week some time prior to the release of WotLK, and thinking it was really cool to see the world in the grip of actual chaos. It was annoying, and difficult to play my normal way, true, but it was also exciting and more immersive. And when I died to a zombie, I didn't feel like I was hurting from that, because then I got to be a zombie for a while, which was fun in its own right, partly because I was now playing a slightly different game.

Apologies for the long comment, and also for the predominantly WoW-related references, but I figured it was the most likely to be recognized by the most people.