Who says you have to have the same thing (or things) spawn in the same place every time? I mean, other than the people who make a living creating theme parks.
Imagine, if you will, a patch of flat grassy meadow. What lives there? Some bunnies? Let's call this meadow bunny spawn "State 0."
Well, in virtual-world-land, what would happen to this bucolic scene? Exactly - some players would come along and murder the lot of them. And then they would come back, in 2-5 minutes, depending on the spawn timer.
But, imagine a more dynamic, programmatic, dare I say "clockwork" world in which the spawned NPCs have (very) basic needs, and can affect one another quite definitively.
The forest adjacent to the meadow has a bear living in it. Bears like meat, so it moves to the meadow and eats the bunnies. This goes on for quite some time, until the bunnies are gone. Now we are at "State 1."
What happens next? Well, the orcs in a nearby village require bear hides as part of their "basic needs" and happen upon the bear. Exit bear, and now orcs are in this meadow (State 2).
Who hates orcs? How about the elves of the forest? Orcs in the meadow mean elven defenders attacking the meadow. Now we have elves here. (State 3).
The elves of course have no need for the meadow and move away. (State 4).
Bunnies, being the opportunistic jerks they are, move in. We're back to... State 0.
The circle of life, my friends, in a spawn cycle. We start a simple cycle, and let it run, with players entering the scene at any given point in the cycle. This illusion of ecologic succession matches player expectations of a living world, and can have real consequences.
Need bunny fur, or are you a druid that requires herbivores nearby to cast a cool rain spell? Well, depending on when you enter this meadow, the different spawn state will inform your next decisions. I mean, you are the player - the actor - the one who gets to break these static cycles and bend and mold them to your will.
If you need bunnies and you are at "bear" stage, then you can kill the bear and camp for the orcs. Killing them breaks the cycle of elves (they would likely walk in, mill around, then move on quickly), leading to empty field... and bunnies.
This isn't rocket science, I realize. It would take some magnificent design chops to build such a living world and make it fun. But imagine trying to manipulate more than just a hotbar full of spells - but actually trying to manipulate the environment itself to suit your needs.
And now imagine player competition within cooperation. You need bunnies? I need bears. How do we resolve this? Can we hold things in between State 0 and State 1 - having both the bear and the bunnies at the same time? How fun would it be to see a druid in a field casting damage shields on bunnies while a ranger is shooting his bow at them as fast as possible to progress the spawn cycle? Would this be fun at all? An interesting design challenge.
Virtual worlds are already moving to more programmatic and dynamic ecologies, but we're stuck still in the belief that the ecology and the gameplay should have a major seperation - the scenery and the game should not be one and the same. Can we bring them together? Should we?
Tuesday, March 17, 2009
The Circle of Life in a Spawn Cycle
Subscribe to:
Post Comments (Atom)
21 comments:
Can we/should we?
I'm supposed to be working instead of displacement-activity tweeting, so I'll keep this short: hell yes. Please.
You bastard, you dared to say "clockwork."
I'm starting to get jealous.
The field would never get past bunnies.
In order for this idea to work you need a big world where fields can go unmolested for a bit of time.
But you can't increase the size of the world because players will complain of travel times.
I like the idea but... I don't know how it's survive first contact with players.
Hmmm.
http://www.raphkoster.com/2006/06/03/uos-resource-system/
Fascinating, Anonymous. I've never seen that article before, but it doesn't surprise me that I thought of something Raph Koster has thought of first.
It's rough being a working dev sometimes, because reading his article I can see some key differences between this and the design I came up with, which you will have to take my word on right now :).
Of course I will bow to the prior art argument.
Pete, this is where this becomes interesting. If the hunter needs orcs, then he will do what he can to ensure bears enter that meadow. By giving players conflicting goals as well as convergent goals, you create all sorts of deal-making and competition.
Also, if we consider many meadows, many forests, and many orc villages, then it would be impossible for players to break all the cycles unless they were too slow to cycle through - in which case, the designer's job would be to up the speed, or introduce new factors.
Doesn't Ryzom have some basic predator/prey/habitat modelling? I remember hearing a little about it in a podcast about the Ryzom Ring building tool.
For me pretty much the first thing I always think about when considering a new system is the barrier to entry for existing MMO players.
With that in mind, I think a system like this could work and has a lot of interesting upside, but the quest / incentive structure should be created (at least initially in terms of player progression) to promote active observation and usage of this system. For example, a quest that said, "go kill whatever is in this area", and when you turned it in the NPC gives you another quest to the same area, with a prompt like "Oh Noes, now look what you did! 'Whatever is spawned there now' are everywhere. This is your fault now, you clean it up."
The reason I think this sort of introduction to the ecology of the game is important is because of the perceived "reduction of convenience". That is, if a WoW player stepped into this world, got a quest to kill Orcs, and then went to the area where they -might- be, he'd be annoyed if they weren't there. In these one-of encounters with this New Ecology, the player perceives that the game is actually *less* convenient, and may have less fun because of that. (This of course assumes that the player has any notion of the level of convenience in these sorts of games).
I think a lot of the cool things that you get out of something like this could be aimed towards guiding players through this dynamic environment - Tracking, hunting, player-player communication to keep up with the trends and movement of mobs seems like it has a lot of space.
The other big question is how to incentivize players returning to an area they have visited before; the quest series is a ghetto way to do this but it seems like a more comprehensive answer would be more appropriate...
A similar example of this can be seen in EQ live on the moon in the zone Hollowshade Moor seen here http://everquest.allakhazam.com/db/quest.html?quest=1609
which I will quote:
The Hollowshade War is the ongoing tribal battle for control of the Hollowshade Moor by the inhabitants of the zone: the owlbears in the north, the sonic wolves in the east, and the grimlings in the south. Occasionally one side will spontaneously attack another, but you can cause one side to attack another by going to any of these locations (owlbear cave in the north, sonic wolf cave in the east, grimling village in the south) and killing the named NPC there and its two guardians, causing another side to attack. Sometimes you need to kill one or more placeholders to get the named NPC to spawn. If all three sides still exist in the zone, which side attacks will be random.
I think DragonRealms did something similar and gave each mob type some basic AI in order to have them move from area to area but the whole thing took to much processing power on the server side and it couldn't be run.
Anonymous (#2), a ring war cycle is not exactly what I am suggesting here, but it does have the essential features of a "static" battle that players can stick their fingers in and stir up. This is precisely the sort of ecology I am proposing, but I am giving much more freedom to the NPCs than in the example you presented.
Razor, I've heard of something like this, but then I look at successful virtual ecosystems in Second Life, and believe that the key here is figuring out how to keep variables down (but interactions up!), and to learn what needs to run free to provide the illusion of a living system, while abstracting as much as you can to keep things from setting the servers on fire.
The biggest challenge I actually see with a system like this (which is conceptually very compelling) is actually player feedback. The question in mind my is how do you let the player know that the bear phase is after the bunny phase, or that the cycle will eventually return to the bunny phase at all. It's not impossible to train the players to understand these things or have additional 'gamey' elements which communicate to the player about the cycles they are observing, but without proper feedback I would imagine that player interpretation of the system would fall more in line of 'oh, there are randomly bunnies around here, so I should run around until they randomly show up' rather than an understanding of the deeper (and inherently more interesting) mechanics which you detail.
Whoever can create a 'living online world' which is both cerebrally interesting as well as compelling as a 'toy world' (with appropriate feedback mechanisms) to the inhabitants will definitely have a winner on their hands.
Interesting idea. Now you have me thinking of stuff I wrote over three years ago as a starry-eyed member of the pre-release Vanguard community.
I'd add the horrors of EQ2's ring events on to this. Kill 3-4 groups, 12-16 mobs, that spawns a heroic group of 3-4 mobs. The that can spawn a named. Thing is it was a CHANCE of spawning a named. A named poxy goblin in everfrost for a Heritage Quest took 4 hours once, and some other guy snagged it. Long term EQ2 players can tell you a 100 horror stories about that spawn pattern. Claymore quest line the wardern snake in PoA. Varsoons and the creator, and that skele in the catacooms below Qeynos.
Ecologies sound great... but having to quest around them can be a real pain. I've just rocked up to hunt 5 random drop bunny skins. Bears are just spawning... and I know there are 5 other players all after the same thing. Do I really want to get involved in that bun fight yet again?
It is was guaranteed spawn, then maybe it wouldn't be so painful. But had some truely had some bad EQ2 experiences with cycling mob spawns.
I put up a brief summary of my original writings up on my blog, if you're inclined to mosey on over.
Stumbled onto this article today. There was a reference above to Raph Koster's article on UO's resource system. He had two followup posts, so I thought I'd put them here if anyone's interested in the full sequence...
UO's Resource System Part 1UO's Resource System Part 2UO's Resource System Part 3
Thanks for that, Per Ludus. Looks like a lot of great ideas are covered in those articles as well.
It's a fascinating concept to think of a world that runs under these kinds of rules
I wonder if one couldn't set up a sort of "demo world" that just ran it's own ecology without human intervention to see how it went. Then later, introduce a new "creature" that was modeled after human players. Tweak their rule sets a bit to make them act abnormally greedy (hording materials), or vicious (killing everything).
It strikes me that what you need to balance out players is to have a few more layers of critters that only come into the picture when the players run amok. For example, thieves or greater monsters (elites) that are drawn to players who gain a high reputation as a killer, or who collect a lot of money or any other resource.
It seems like human players could coexist in this kind of world, but at some point their sheer numbers would overwhelm the ecosystem and cause it to collapse - not unlike the real world actually.
Actually, there is one MMO that does a very open ecology of sorts and that is Wurm Online It allows for players to harvest wood, mine, craft, build, and so on as part of a living ecology. If you cut down a tree, it's gone. If you dig in the ground you can make caves and mine, but have to figure out where to dump the diggings, which in turn changes the terrain. It's no problem to clear the forest and build a town somewhere, given enough time and people to do the work.
The crafting system is incredibly deep, which is both cool at first, but incredibly burdensome if you're trying to "level up" your crafting.
Just read the links for Koster's design ideas on the Maslow model and the funny idea of a character carrying alot of meat near a hungry Dragon
occurred to me
Post a Comment