Has any game done this yet?

[WindupAtheist]

Happened to think of an interesting game mechanic while looking at another thread.  Since I've played a grand total of three MMO games in my life, however, I can't be entirely certain someone hasn't done it already:

A camp of orcs (or whatever) spawn at a random location in the world.  Every X hours that they remain unkilled, the spawn moves up a notch in terms of difficulty and area.  For every Z number of orcs killed, the spawn moves down one notch, disappearing entirely when destroyed at the lowest level.  Each day thereafter, the whole thing has an increasingly high chance of starting over elsewhere.

This serves to allow players to make a mark on the world, albeit in a limited way.

"Oh shit, better kill those orcs next to the river, or else by tomorrow they'll be swarming all over the road."

Or...

"Crap, the mine is full of orcs.  Round up the guild and let's go exterminate them."

[Llava]

To a very limited extent, Horizons did this.

[stray]

MEO (Middle Earth Online....Or LotR online...Whatever the hell it's called), supposedly makes some high use of instancing to do things like this. Kill an Orc camp, for example, and the next day you'll see a group of Dwarves in same spot cleaning up and setting up shop.

[Daydreamer]

Did anyone else picture Crushbone Orcs and tents appearing in the middle of Freeport when they read that description?

Sounds like a neat idea, but you would have to be careful about caps and randomization.  With caps to high, an area that is left alone for a while will become unkillable to all but a select few.  On the other hand, If the randomization was done improperly finding an appropriate level camp could be either too hard (all behind near-unkillables) or too easy (A level X camp appears at site Y every Z hours).  But if you laid it over a more hard-coded foundation it would be neat - for example Loch Modan in WoW could be hard-coded to be between level 10 and 20, and use Troggs, Ogres, and Kobolds, but their location and difficulty and the fluff behind them could change.  Invading the silver mines to the N at level 13-14 one day, and emerging from the excavation site at 19-20 the next, with a comensurate change in quests.

Of course assuming you can do all that, you still have to make the change meaningful.  A level 14 kobold can't be the same as a level 14 ogre which can't be the same as a level 14 foozle.  Without giving all the mobs more skills and AI than they already do this would be difficult, but not impossible.  RPS type damage triangles may work, but I dunno.  Still, a neat idea even if the difficulties make it purely theoretical.

[Daydreamer]

MEO (Middle Earth Online....Or LotR online...Whatever the hell it's called), supposedly makes some high use of instancing to do things like this. Kill an Orc camp, for example, and the next day you'll see a group of Dwarves in same spot cleaning up and setting up shop.

I'm not sure I'd trust Dev talk from MEO about MEO.  though I haven't been following them religiously, what I have seen seems to indicate a higher than usually level of pie-in-the-sky wishing that usual in an MMO, and in this industry that says quite a bit.

[stray]

The idea makes sense though. The only way to do it without pissing everyone off is instancing.....Which is pretty much the antithesis of world mechanics that he's hinting at.

[Samwise]

Strategy games have certainly been doing this for years, albeit on a much smaller scale.  "Shit, the Zerg have a hatchery growing at the southeast resource node... send a dropship down there quick or there'll be hundreds of 'em!"

[squirrel]

The mechanic is problematic at least in the current MMORPG mould, which could use changing no question. The thing is grp A will find the 'tough' camp and kill it repeatedly until it's 'weak' then move on. Group B unknowingly following behind is going to get essentially worn out table scraps from the other group. No matter how you work it - this will happen, unless as suggested above, everything is instanced, or you completely change the kill mobile bags of xp/loot paradigm.

[Stormwaltz]

We were going to do it in Ninth Domain. Unmolested spawns would increase in size and strength, and ultimately begin draining the nodes of crafting resources in surrounding areas. The resources would be pooled in their camps, becoming a reward for clearing them. With enough pooled resources, they mobs would "build" defensible outposts.

Which means zip since that project lost funding, of course.

[ahoythematey]

The mechanic is problematic at least in the current MMORPG mould, which could use changing no question. The thing is grp A will find the 'tough' camp and kill it repeatedly until it's 'weak' then move on. Group B unknowingly following behind is going to get essentially worn out table scraps from the other group. No matter how you work it - this will happen, unless as suggested above, everything is instanced, or you completely change the kill mobile bags of xp/loot paradigm.

Well...perhaps you divide the entire gameworld into territories/zones, and use control or and/or conquering of those as your means of advancement or achievement.  For example, divide the world into Risk-like areas, where certain regions are more desirable to capture and control for different reasons(coastlines easily defendable etc.).  I think it would ideally work for smaller-pop servers, though, something like a few hundred players max.  Though, I do think that would make all MMO's better.

[schild]

We were going to do it in Ninth Domain. Unmolested spawns would increase in size and strength, and ultimately begin draining the nodes of crafting resources in surrounding areas. The resources would be pooled in their camps, becoming a reward for clearing them. With enough pooled resources, they mobs would "build" defensible outposts.

Which means zip since that project lost funding, of course.

Tell me the alternative was NWN. I want another reason to be angry at that piece of software.

[WindupAtheist]

Yeah, this idea runs into problems when applied to the WoW/EQ model.  Lots of good ideas run into problems when applied to the WoW/EQ model, because that model is a goddamn abomination.  Not to harp on my favorite subject again, but...  The weakest possible character in UO had 55 health, while the strongest had 112.  A run of the mill suit of armor reduced damage taken by 40 or 50 percent, while the most uber reduced it by 70 percent.  They disrupted the reasonably level playing field by adding a ton of uberitems, but it was still better than a system where an endgame character was hundreds of times more powerful than a newbie.

As long as such is the case, your world is always going to have to be static and carefully controlled, with content segregated by level.  Otherwise you're gonna have catasses bitching about having to plow through grey shit on their way to town, and newbs bitching about getting raped by ubermonsters.  The field has to be level enough that your average critter can at least engage a powerful character long enough to take a couple whacks, without being able to blow out a newbie in one hit.

Levels were a simple and convenient way to track character advancement when the games were being played on a tabletop or a very limited old-timey computer.  Someone needs to realize that the concept is fucking outdated already.  Furthermore, I'd love it if a developer decided there was a happy medium between total progress-halting chaos and bland theme-park sterility.  As it is, we can't have anything in these games be a surprise, because someone might be inconvenienced by it.

EDIT:  As an aside, UO recently did *something* like this.  Britain was "invaded" yet again.  Monsters would start spawning near the dungeon Despise, and if they weren't killed at a sufficient rate, more would spawn closer to town.  If that spawn wasn't killed, another spawn point would appear even closer to the city.  Eventually you'd have monsters running around Brit bank killing spammers, and the players would have to pummel successive spawn points until the "head" of the invasion was pushed out of town.  The only problem was, there was no fucking end.  You could never put a permanent stop to the spawning, and the damn thing dragged on for months.  Eventually everyone just did their spamming in Luna, and Britain was full of monsters all the time.

[Krakrok]

UO originally had orc camps that would randomly spawn around the map. It was part of the ecosystem thing I think. If they weren't killed they would despawn and respawn somewhere else or maybe they just moved once a week. I don't remember. Killing them didn't increase or decrease their strength however.

[HRose]

We were going to do it in Ninth Domain. Unmolested spawns would increase in size and strength, and ultimately begin draining the nodes of crafting resources in surrounding areas. The resources would be pooled in their camps, becoming a reward for clearing them. With enough pooled resources, they mobs would "build" defensible outposts.

"Wish" did most of this back in beta.

[Calantus]

EDIT:  As an aside, UO recently did *something* like this.  Britain was "invaded" yet again.  Monsters would start spawning near the dungeon Despise, and if they weren't killed at a sufficient rate, more would spawn closer to town.  If that spawn wasn't killed, another spawn point would appear even closer to the city.  Eventually you'd have monsters running around Brit bank killing spammers, and the players would have to pummel successive spawn points until the "head" of the invasion was pushed out of town.  The only problem was, there was no fucking end.  You could never put a permanent stop to the spawning, and the damn thing dragged on for months.  Eventually everyone just did their spamming in Luna, and Britain was full of monsters all the time.

I love EA.

[Calandryll]

UO had something like this a few years ago. The programmer who created it called it "Evil in a can". Basically the system could dynamically modify spawn rates and numbers based on how much the players focuesed on killing the monsters. The original use was to create spawns that would generate monsters and as players killed them, the spawns would change and eventually spawn a boss monster. Killing the boss monster would end the spawn for a time.

We used a modified version of the system to run some of the orc and savage spawns in the scenario ongoing content which were pretty close to what you are talking about. Players could defeat and stop spawns of orcs while unchecked spawns would increase in number and difficulty.

I'm guessing they used another modified version of the same system in the recent event that WindupAtheist mentioned.

[WindupAtheist]

Those "scenarios" you guys were running, that started with the orc/savage war, were goddamn brilliant.  For those who don't know, each one introduced a chunk of new content, and could affect the shard in a permanent manner based on what the players did.  On a lot of shards, the city of Yew has been destroyed for several years thanks to a scenario that those shards happened to lose.  That was some fairly ballsy shit.  A shame they quit doing it.  I think EA decided it was too expensive or some such.

You think Blizzard or SOE could maybe sink a couple million into doing something like this?  Naaaah...

[Paelos]

Blizzard runs something called Elemental invasions where teams of elementals randomly spawn on certain maps during the week. They are denoted by portals near the elementals. Killing enough of the invaders spawns an elite boss elemental worth a hefty amount in loot. I've helped kill off one of these invasions before, and I can say it's a fun time.

Two things are bad about these invasions. One, they don't go anywhere. Say the players leave them alone for a while, they don't expand on their spawn territory. I think they should. Two, the bosses are too easy. My group of five took one down after a few attempts. A good tank, mage, healer, DPS, secondary healer combo would do the trick easily.

I do give them props for trying, and perhaps it's a step in the right direction for world events.

[Calantus]

Also, it is more profitable to farm the elementals and ignore the bosses as they have something like 10 times the droprate on essences as their non-invasion cousins but stop spawning when the boss dies...

[Raph]

Dundee did exactly this on a UO gray shard (I think it was probably Ackadia) long before we hired him for SWG--sometime in 99 maybe?

The SWG POIs were supposed to do something of the sort, but never did.

MUDs, of course, did this long long ago.

[Strazos]

Gemstone 3/4 has/had invasions on a sporatic basis.

Once, there was a worldwide event for a week; a War.

Suffice to say, it was worth it, even if I couldn't actually hit anything.

[Glazius]

Gemstone 3/4 has/had invasions on a sporatic basis.

Once, there was a worldwide event for a week; a War.

Suffice to say, it was worth it, even if I couldn't actually hit anything.

Oh yeah, when the undead boiled up. I remember that. The vanguard always clawed their way out first, so the less experienced hunters could at least fight them off and get to safety, and the dragging/healing/instaports/field hospitals that got set up in the places they couldn't go...

Heh. Right, that's what made it worth it. There were a few rooms in the game warded against undead and creatures of any type, and people could set up a no-fighting zone on the go if needed to get somebody back up. So you could actually rescue someone or get rescued.

In the days of the continuous MMOG world, I wonder if it's possible to establish such 'safe havens'.

--GF

[Calandryll]

Those "scenarios" you guys were running, that started with the orc/savage war, were goddamn brilliant.  For those who don't know, each one introduced a chunk of new content, and could affect the shard in a permanent manner based on what the players did.  On a lot of shards, the city of Yew has been destroyed for several years thanks to a scenario that those shards happened to lose.  That was some fairly ballsy shit.  A shame they quit doing it.  I think EA decided it was too expensive or some such.

You think Blizzard or SOE could maybe sink a couple million into doing something like this?  Naaaah...

The funny thing about the scenarios was they didn't really take that much time to create. We had 3 people on the scenario team and generally we could create an 8 week event in about 8 weeks (including testing) - which was faster than normal publishes. They were also really unobtrusive to the service. Over the course of about 10 months, we did 26 publishes and about 10 client patches and not one caused any kind of crash or had to be rolled back. I think we had two bugs get through during the first scenario that were considered "immediate" and both of those were fixed within 24 hours. After that we only had really minor stuff. The nice thing about doing the updates weekly was we could hotfix any minor issues the following week. Compared to other UO publishes, the scenario stuff was incredibly clean. We developed a pretty nifty system of working with QA and budgeting time in our schedules to create specific QA tools that they would need to test our stuff. I think we had 2 or 3 QA people assigned to the scenarios and considering the amount of content they had to test in a relatively short period of time I’m still amazed at how quickly they turned things around.

The scenarios ended after I joined the UXO team and I never really understood why. We had evidence that they were having a positive effect on retention and logins were way up during the months the events were running.

Anyway, we used that “evil-in-a-can” system as the basis for lots of the stuff in the scenarios. The first was when the Orcs and Savages were fighting for control of the Orc forts. Players could take a side and which creature they fought more would determine who ultimately won the fort. We also had Orc camps spawn outside the cities that were being invaded which would affect the strength of the city invasion spawns if players didn’t attack the camps. The second scenario had these magical generators that would spawn golems until players used a panel on the generator and solved a puzzle to shut it down. We also used the evil in a can system for the final battle in the Exodus dungeon – players had to defeat Exodus’ minions to restore order to the Gargoyle town. So players at that time actually got to turn what was a monster spawn area into a full fledged city, complete with Gargoyle shopkeepers and two new crafting systems. The last use was the Yew attack you mentioned. I can’t recall what players had to do to save the town and push back the swamp though.

[Hoax]

Conan MMG claims they will heavily feature player influenced dynamic spawning type stuff.

*knock on wood*

[Strazos]

Gemstone 3/4 has/had invasions on a sporatic basis.

Once, there was a worldwide event for a week; a War.

Suffice to say, it was worth it, even if I couldn't actually hit anything.

Oh yeah, when the undead boiled up. I remember that. The vanguard always clawed their way out first, so the less experienced hunters could at least fight them off and get to safety, and the dragging/healing/instaports/field hospitals that got set up in the places they couldn't go...

Heh. Right, that's what made it worth it. There were a few rooms in the game warded against undead and creatures of any type, and people could set up a no-fighting zone on the go if needed to get somebody back up. So you could actually rescue someone or get rescued.

In the days of the continuous MMOG world, I wonder if it's possible to establish such 'safe havens'.

--GF

No, what I am talking back is the Bregandian War....they even went so far as to have GM's run around as a few "key" characters. They would launch attacks out of thin air, had roving bands of high-level mobs, captured a few people and held them as prisoners, and even taunted us publically.

Creativity FTW?

[AOFanboi]

Conan MMG claims they will heavily feature player influenced dynamic spawning type stuff.

From the company that brought us "entirely underwater settings", "many of the quests will be unique in nature so, when a quest is finished, you nor anybody else will be able to complete the same exact quest again" and "each clan will have their own building" in the Anarchy Online launch.

Googling AO Previews For A Better Tomorrow.

[Sairon]

If used correctly I'm sure it could be fun as hell. I don't think it will be that fun if it's the main objective to lvl in a game though. Dynamicaly spawning camps and have the players search and destroy the ones apropiate to their lvls could mean a lot of in the long run boring searching.

[MahrinSkel]

As HRose said, Wish had a primitive version of this in beta, and the final target was pretty sophisticated.  The solution to the "table scraps" problem was mostly a matter of *scale*, Wish was going to have a very large world where there were always going to be areas that hadn't been visited recently.

Another piece of it was a sort of "species manager", there was going to be a maximum number of creatures of a given type and a maximum amount of power for the groups of them to divvy up (there were also going to be minimums).  This was all tied together with the cilmate and terrain type overlays, a given creature might have 80% affinity for mountainous terrain, 40% for hills, 10% for grasslands, and 0% for desert and swamps, and a similar set of affinities for different climates.  So although the Foozles might be most commonly found in a certain area, they'd show up in others as well.  And if the Foozles were intelligent, each "camp" would have an RTS-like AI managing it, and each city would have something similar (trying to keep players from taking it, or trying to take it back).

Artificial ecologies have been studied by academics for about 20 years now, there's a huge amount of prior art that wasn't available when UO's virtual ecology blew up in their face.  Truth is, an artificial ecology isn't even all that *hard*, anyone who has kept a tank of fish alive for a significant period has solved the problem.

--Dave

[HaemishM]

I think what really stops MMOG teams from this kind of virtual ecology is the one question all game features should be asked. Will this make the game any more fun for the majority of users?

Frankly, it's a good idea for worlds, but in MMOG's that are games, it will just mean that people will be inconvenienced, either by not finding mobs they can fight or not finding mobs that are worth fighting. And that's not fun. Virtual worlds are about circle-jerk academic theories, MMOG's are about games. Nothing wrong with either one, but one is not something a mainstream audience should ever be exposed to, because they CAN'T HANDLE THE TRUTH.

And yes, in the DikuMud fog MMOG's currently have, this idea is all bad.

[tazelbain]

It's taken this long for them to figure in-game economy, it'll take another 10 years to figure out in-game ecologies.
Just as designers learned to strip out the unfun aspects of economy, they'll eventuall learn to extract the fun parts of ecology and bring them in. And there'll be much rejoicing.

EDIT: 

[Stormwaltz]

Will this make the game any more fun for the majority of users?

As with most things, it can be, but it's not guaranteed to be. When you get down it it, you're just talking about a very sophisticated method of dynamically generating landscape encounters. I think many people would enjoy the experience of raiding, pillaging, and burning down mob forts. The question is, is the ecology running the forts designed robustly enough that eveyone can share and enjoy? Or will the spawns get camped into submission before anything cool happens?

It's a system I was very nervous about balancing, but goddamn the end result sounded like fun. Especially when we started talking about unmolested mob forts launching raids on player-owned cities.

Just to be clear, this was something from my last job at the (now defunct) STE, nothing to do with BioWare.

[Yegolev]

To a very limited extent, Horizons did this.

A lot of this talk sounds like Horizons, or what Horizons had the potential to do at least.  Blight anchors?  Not enough cowbell.  Seeing the word DIE scrawled on the ground in blood during world end?  Exactly the right amount of cowbell.

[Roac]

When you get down it it, you're just talking about a very sophisticated method of dynamically generating landscape encounters

Not just.  Take the random encounter bit, and add a few scenarios where they affect the game.  If orcs (whatever) spawn between town A and B, they affect supply lines.  What does that do?  Kick up prices in town on certain goods, or reduce supply (or both).  If there's a little bit of AI, you can have power vaccums filled by different types of mobs.  Orcs getting hunted to extinction?  Maybe there's a surge of trolls, or whatever else, in the area.  Maybe even take over their camps, and change loot drops.  If you have a mission system, tie spawns into that; different sides offer missions against one another, or Human (again, whatever) cities. 

It isn't just an encounter when it affects gameplay; that's dynamic content.   

[Stephen Zepp]

When you get down it it, you're just talking about a very sophisticated method of dynamically generating landscape encounters

Not just.  Take the random encounter bit, and add a few scenarios where they affect the game.  If orcs (whatever) spawn between town A and B, they affect supply lines.  What does that do?  Kick up prices in town on certain goods, or reduce supply (or both).  If there's a little bit of AI, you can have power vaccums filled by different types of mobs.  Orcs getting hunted to extinction?  Maybe there's a surge of trolls, or whatever else, in the area.  Maybe even take over their camps, and change loot drops.  If you have a mission system, tie spawns into that; different sides offer missions against one another, or Human (again, whatever) cities. 

It isn't just an encounter when it affects gameplay; that's dynamic content.   

Hehe..I have to admit it's interesting to see the turn-around on the perception of this idea from the 6 months ago or so when I proposed it across a few different threads. As Roac mentioned, the deepest and most interesting parts are not only the actual presence of the npc's themselves being dynamic, but most importantly how they affect the rest of the environment, from power vacuums to dynamic missions, to other semi-emergent behaviour based on how the player populations interact.

[Sairon]

When you get down it it, you're just talking about a very sophisticated method of dynamically generating landscape encounters

Not just.  Take the random encounter bit, and add a few scenarios where they affect the game.  If orcs (whatever) spawn between town A and B, they affect supply lines.  What does that do?  Kick up prices in town on certain goods, or reduce supply (or both).  If there's a little bit of AI, you can have power vaccums filled by different types of mobs.  Orcs getting hunted to extinction?  Maybe there's a surge of trolls, or whatever else, in the area.  Maybe even take over their camps, and change loot drops.  If you have a mission system, tie spawns into that; different sides offer missions against one another, or Human (again, whatever) cities. 

It isn't just an encounter when it affects gameplay; that's dynamic content.   

This sounds very cool in theory but I think it will pretty boring in practice. For example if orcs has spread and reduces supply to a city, which means prices sky rocket, I think something like that will only irritate players. If you're in the area and want to buy arrows for your bow, and the price has doubled because of orcs, then that might be cool the first time, but after that it's just a mechanic which means you can't get your precious arrows. Leting trolls kill off all the orcs, and the other way around if the players helps the orcs might suffer from the same fate. If people wants the loot tables of the orcs, then killing off the trolls will only be bothersome. The same way if for example orcs are prefered to kill for xp.

I think it will be pretty hard to create a system like this which will be cool beyond the first hour of experiencing it. Missons could make it more intresting but I'm yet to see dynamic missions being fun in the long run because in the end they're all just built up around the same base of ingredients.