Making the World through Instances

[Glazius]

Well, sort of.

Another game design idea inspired by Legend of Mana.

Go to the map guides on that page. Executive summary: Locations get more powerful the further from 'home' you place them, and the later in the game you place them. They also spread mana out to the map, and with certain mana levels in certain locations special events happen. How 'powerful' a town is determines how strong the stuff sold there is.

So for an MMORPG... populate an overworld map with fixed town locations, and a bunch of doors into instances. Some of them lead into wizard's towers, some are rope ladders into the trees, some are cave entrances, et cetera. Note that some of how this shakes down assumes "instanced vision" - that is, a town is built with generally the same geometry but will look different to different people. Someone who's hurt the town will see it in squalor; someone who's helped it will see everything bright and shiny. This also applies to indoor environs within the town - help a merchant and his shop is bustling, blow him off and it gets abandoned.

Within town you can get quests. Most of these quests give you a key item, even if it's ephemeral. Now here's the interesting part - you're not told where the quest is. You decide. Walk out to the world, pick an instance entrance, and the quest sets up there. (You get a menu for which item to use when you click on the entrance.) Sometimes there'll be limitations - for example, a fire mage isn't going to set up in a tree fort, or a snow beast go to ground in a lava cave. But in general you have the world to decide where the quest is.

In general, quests get harder but more rewarding as you go further out into the wilderness. Put the goblin warrens in a cave near town and the goblins are weak, since the town guards have cleared most of them out. Put them in a remote forest and they're tough hunter-gatherers, peppering you with bows from remote branches, but they've got a lot of interesting stuff from the surrounding woods. But it's a little more complicated than _just_ that.

If you put a quest close to a town, the town will generally get a little weaker (because there's a goblin warren RIGHT OUTSIDE the front gates), and there's a chance the quest will be a lot harder and weaken the town more - for example, if a painter asks you to find his sister and you choose somewhere near a town as the kidnapper's hideout, the kidnapper may turn out to be, not monsters, but a high-ranking citizen of that town with his own guards, and suddenly that town has at least a corrupt person in power, and a few dead guards if you don't just stealth in and sneak the girl out. But this may also lead to a quest later in the game where you go after the corrupt nobleman.

If you put a quest far out in the wilderness, there's a chance the quest may get a little easier - for example, the tree-goblins have kidnapped an NPC ranger and you get someone watching your back as you hack through the place. More than that, it may lead to unique rewards - if a wizard in a remote tower kidnapped the painter's sister, it might turn out (after you escort her through dangerous terrain and back to town) that he was after some piece of the painter's, and when you look at the finished product it's the most beautiful thing you've ever seen, and you get "Memory of Spring", a special AoE heal with a long cooldown that bursts the painting's imagery out into the world.

Perhaps also in the long run each quest could have an impact on the spawn table - if a lot of players put a fire sorceror in a specific tower, more fire-based monsters appear around the tower, and the terrain changes. If the goblin warrens get dumped into the same general cluster of caves, more goblins show up in the overworld.

That's about all I've nailed down for now. Ask questions, it helps me refine!

--GF

[Pococurante]

The game world is basically a lobby with doors to instances?  If I as the the equivalent of a red PK and my gloriously blue/Good buddy want to group together are we in the same instance of a bright shiny town or a muddy handful of squalid huts?

Provocative idea but I'm still unclear on the details.

[schild]

I actually detailed out a system by this. It would work if, well. No, let me explain it a little better. Ready?

Each instance is a contained story that is not necessarily in accordance with the rest of the instances.

Basically, make a mold for a game and use that entire MMORPG system to create  a game that could span centuries full of smaller games. Maybe each instance is 2-3 hours long. Maybe you can only take 3 other people into one instance and have to go solo in another. Never more than 5-7 though. Then it gets into raid territory and the hardcore MMORPGers don't deserve fun. Anyway, fixed storylines and fixed amount of content can go along way towards immersion.

That's all I really want to say about it right now.

[Glazius]

The game world is basically a lobby with doors to instances?  If I as the the equivalent of a red PK and my gloriously blue/Good buddy want to group together are we in the same instance of a bright shiny town or a muddy handful of squalid huts?

Provocative idea but I'm still unclear on the details.

When you join a group your world ripples and shifts to become the leader's world. One very interesting example that occurs to me - somewhere in the world there's a very insular society which rings its homeland with walls every now and again, and different patterns of action give different wall patterns. With enough people in the group you can keep swapping leadership and walk through the walls.

The core conceit of this gameplay is that you're a godling taking on mortal form to investigate the last days before the collapse of the world, so that's exactly the sort of thing I want to be possible.

The overworld is pretty much a lobby, with the occasional thing to be done outside to open up another instance - if, instead of getting the "key object" to the frost mage's lair from the town archmage you talk to the tinker, he tells you about a flame projector he needs parts for. You go out and grab some flame cores from various fire creatures, bring them back, and the tinker gives you a "key object" to a partially-melted version of the lair.

But like CoH does with its train missions and fairy rings now, it's possible for the instance just to reflect the real world. There could be abandoned forts in the wilderness, fully explorable without any instancing - but also "key objects" that could work on the fort's doors to shoot you into an instanced version of the fort, perhaps with an otherwise-solid piece of geometry missing and letting you down into the catacombs or whatever.

--GF

[Xilren's Twin]

I generally like the concept of using instancing to tell better, more player focused stories and your implementation would certainly help give each players more control over the uniqueness of their gameplay.

One thought I've had on the whole "if two players who have gone different path team up, who's instance becomes the shared reality?" is this.  If lets say you do a "rescue the princess" quest and save her, and you buddy goes the evil route and take the "assassinate the princess" side, if he wants to join you on a plot line that continue from said princess being alive, he would have to relinguish either some or all of the experience and items he earned from doing the original quest in the opposite direction.  Basically it amounts to going back in time and choosing another road, undoing the past. (insert whatever game fiction you want that explain why you can do this; that's just window dressing anyway).

Why punish the player in this manner?  Well; firstly, if the game isn't a typical EQ level scheme with massive exp curve and phat loot focused , the losses wouldn't nearly be as noticeable or disruptive.  Secondly, it starts giving teeth to game choices so you could have some true divergence between characters based on how far down certain paths they go, plus, it's a built in respec/reverting mechanism. 

Sounds very small group oriented and nichy; I'm ok with that 

Xilren

[Pococurante]

Sounds very small group oriented and nichy; I'm ok with that

Me too.  Though I suspect it wouldn't be all the nichy (nietsche?).  As Second Life is suggesting, virtual immersive entertainment is becoming more mainstream.  A lot of the things we hear devs say (and we sometimes parrot) is the MOGs must have XYZ and ABC to be widely successful.  But Second Life rarely has those elements and the elements it does have are the kind we're told can't sell.

I have a healthy skepticism of market studies.  All too often they seem crafted to assure the client of the assumptions they had before commissioning the study.

[Glazius]

I generally like the concept of using instancing to tell better, more player focused stories and your implementation would certainly help give each players more control over the uniqueness of their gameplay.

One thought I've had on the whole "if two players who have gone different path team up, who's instance becomes the shared reality?" is this.  If lets say you do a "rescue the princess" quest and save her, and you buddy goes the evil route and take the "assassinate the princess" side, if he wants to join you on a plot line that continue from said princess being alive, he would have to relinguish either some or all of the experience and items he earned from doing the original quest in the opposite direction.  Basically it amounts to going back in time and choosing another road, undoing the past. (insert whatever game fiction you want that explain why you can do this; that's just window dressing anyway).

Why punish the player in this manner?  Well; firstly, if the game isn't a typical EQ level scheme with massive exp curve and phat loot focused , the losses wouldn't nearly be as noticeable or disruptive.  Secondly, it starts giving teeth to game choices so you could have some true divergence between characters based on how far down certain paths they go, plus, it's a built in respec/reverting mechanism.

What I'm thinking of right now goes something like this.

After you complete/fail your quest, you can only spend a small amount of time logged in in the overworld before your quest becomes reality. During this time you can try to team with someone who has the same quest and opt to throw your lot in with theirs after their quest finishes - this gives you the same outcome they got. Someone with the same quest who hasn't even started it can take the same option - there's a small timer that activates in all other matching party quest objects after completion, letting them sync up. This gives everyone the rewards of quest completion (whatever they might be) and amplifies the effect of the quest on the overworld environment. But you don't have to accept the results of anyone's quest but your own. It's actually _intended_ that High Lady Moonsilver and Dark Lord Blackthorn could still team up despite living in two different worlds - in fact, there might be some things that _only_ they could do, by switching leadership. (Lord Blackthorn leading them both through the demolished walls of the royal archives and Lady Moonsilver taking over to pull information out of the now-intact scrolls.)

For the case of save the princess/kill the princess, no quest contingent on saving the princess is going to have a "key object" matching any quest that someone got who killed the princess.

This engine is going to have a rough sort of use-based skill system, so spending "time" and losing some skill points to "rebuy" a key object might be an interesting mechanic.

--GF

[Sairon]

Sounds intresting but there's something which might be a problem, atleast for me. If you're going to void instance progress then advancement, as stated by Xilren, needs to be small. However advancement is one of the aspects which makes RPGs fun for a lot of people. Also if you're changing your instance background all the time, then I think that would spoil some of the immersion power which this game design grants. And if the instance progress is defenite and unchangeable, then instance mustn't affect your character all that much, to avoid "fucking up" your character by bad decision.

EDIT:some grammar

[Glazius]

Sounds intresting but there's something which might be a problem, atleast for me. If you're going to void instance progress then advancement, as stated by Xilren, needs to be small. However advancement is one of the aspects which makes RPGs fun for a lot of people. Also if you're changing your instance background all the time, then I think that would spoil some of the immersion power which this game design grants. And if the instance progress is defenite and unchangeable, then instance mustn't affect your character all that much, to avoid "fucking up" your character by bad decision.

Advancement is going to be basically by-skill except in some exceedingly rare cases - endgame monsters and spells that slightly tweak your stats. A "forge" buff taints its caster with Creation and its target with Metal in addition to its effects. A "sere" blast taints its caster with Destruction and saps the Earth of its target, in addition to the pain factor. The reward for completing some instance isn't a big whack of XP or an uberweapon - it's a change in the world you see, with maybe an additional plot unlock or minor character power. If you want to do the instance over again and "unchange" your own world, you need to sacrifice some of your power (skill points) and whatever benefits the completion had.

This does not prevent you from doing the instance 100 extra times with 100 other people, but you'll only get your final reward once without making the sacrifice.

--GF

[tazelbain]

You make it sound as if you are actively building this.

It seems GW has the capacity to do what you guys are talking about, but they do much with it.  I assume its because content costs so much. And GW makes no attempt to maintain continuity, kinda annoying.

[Pococurante]

However advancement is one of the aspects which makes RPGs fun for a lot of people.

I'm glad you said that - because it crystalized in words why I think it's wrong.

People want a sense something permanent happened.  Or to turn the prism slightly, they want a sense of progress and accrual.  But not necessarily "advancement" - that just happened to be the low-lying fruit someone chose many years ago and has since become the bedrock of fantasy games.

If "you" can give me a sense that what I did changed my environment, my monkey sphere, permanently than I'm satisfied.  Once we put the challenge this way suddenly new doors open up to scratch that itch that goes beyond monty haul, levelquest, and adrenalin.

We got monty haul and levelquest because for the past ten years we've been fighting engineering and infrastructure adoption problems.  We're past that now.  Broadband is common, the "least common denominator" game PC can still run WoW credibly, and the graphics/network engineering problems are nearly fully solved.  Publishers/Devs now have pretty complete packaged platforms/engines at their disposal - this is very new.

All that's left is new hooks, new ways to psychologically engage the customer.

Take WoW items and how it relates to character customization.  And hold faction in the back of your mind for a second.  Why can't I slot standard items with enhancements, use a different color pigment to stain the material, choose an alternate gem to adorn the edge.  I certainly can from an engineering perspective - detail threshold zoom allows the architect the freedom to continue delivering client performance but when the view threshold is hit I can show those details.  Why is faction limited to simple discounts when it should open up entire trees of quests and certifications/awards.  Why can't there be diplomat skill trees and smuggler templates.

Rhetorical questions.  But for the next few years this is where the action will be.

Focusing on twitch combat versus stacking, keyboard versus controller - these are all beside the point and people looking to be challenged by focusing on these are the proverbial drunk under the street lamp.  What engages the customer imagination is not hand memory or command template.  It's how fluid the game opens up their imagination and engages their sense of time flow.

[Sairon]

Yea but is that permanent change cool when it happens in an instance? You're not really affecting the world, only your little world which others can't really see.

[Pococurante]

Depends - do I carry the change out of the instance.  If not it's not really a permanent change to the character...

Just to clarify, I'm not suggesting something that changes the game world but something that changes the way to world reacts to the character.  The former is Less Fun for other players of course while the latter makes no difference except giving other players a cue how to react to the individual.

CoH's medal system is an interesting if watered down implementation of this mechanic.

If the world/lobby is completely static you lose whatever cool factor is provided by the instances.

[Glazius]

Depends - do I carry the change out of the instance.  If not it's not really a permanent change to the character...

Just to clarify, I'm not suggesting something that changes the game world but something that changes the way to world reacts to the character.  The former is Less Fun for other players of course while the latter makes no difference except giving other players a cue how to react to the individual.

CoH's medal system is an interesting if watered down implementation of this mechanic.

If the world/lobby is completely static you lose whatever cool factor is provided by the instances.

In this case, your world is not static, though in an absolute sense about all that changes are some textures and the permeability of various walls.

Over a long period of time changes from the instances may make changes come over the landscape. The ideal case is that so many people decide to, say, raid the Fire Lord's Armory at a "good location" that over time the landscape near that location starts spawning more fire creatures and fiery terrain, making it more obvious where the "good locations" are.

--GF

[Glazius]

You make it sound as if you are actively building this.

Well, it's been in my head for... three years now? Something like that?

In addition to determining your _personal_ storyline through the instances, I was planning a method to determine the _world_ storyline - basically, since the world is still being reconstructed after the collapse, you (being a godling) can step into an open-PVP zone tentatively named the Divine Plane, where all monsters are scaled to be (more or less) equally powerful, and what gets killed on the Divine Plane affects the development of the storyline. Kill dragons but not rats and the dragon raid never happens, but instead some town gets overrun with a plague...

--GF

[Bokonon]

Interesting. I don't know if anyone remembers the old Lost Continents concept being developed by [I forget Who], but their design was similar to a lot of the suggestions above. It was going to be a largely instanced [30s Pulp Fiction] world. You would often re-enter certain instances, which would reflect your choices. You could join another group doing the same instances, but it would take on the leader's vision of the instance. Characters could keep their look, but upgrade their costume with items that would make them more powerful, or different types of damage, or different looking. There was some vague talk of slower, more strategic combat. An archnemesis would follow you through all your instances (like what CoH seems to be thinking about).

They were probably biting off more than they could chew.

It had me intrigued; only MMORPG I went fanboy on. So of course it got axed.

What can I say about F13?  It's really my favorite website in the entire universe!  I love the irreverent banter and sly wit these keyboard jockeys produce.  And I especially love the staff, they're AWESOME.

[pxib]

This isn't just a combinatorial explosion, it's a combinatorial minefield.

Even if the differences are only "some textures and the permeability of various walls", the potential exists that any individual in a group might become leader and that every individual might have a wildly different, uh, worldview. That information might only be a pile of binary switches, but the number of switches required for even a simple town could be immense. Making an interesting number of possibilities available isn't the same as making every possibility interesting. Without superhuman efforts on the designers' parts, players could accidentally box themselves into boring and broken worldviews by accepting another player's apparently positive solution to some apparently arbitrary quest.

I'm all for devoting heavy development time to plots and stories rather than graphics and skills, but allowing a meaningful amount of change at the level you're describing requires an unbelievably robust sort of plotting and storytelling. Every new potential interaction must be checked against the list of every previous choice, and every interesting interaction -- the sort which realistically changes the world -- adds at least one  new setting to that list.

This is avoided by making most changes uninteresting, but that shortcut has robbed  multiplayer games of their power of myth.

So bravo, you have slain the dragon of the singular nature of genuine change. In your game, anybody can make a real difference without denying another the chance to do so. Unfortunately, the treasure that dragon guards is just another dragon... and one of potentially infinite size.

[Glazius]

This isn't just a combinatorial explosion, it's a combinatorial minefield.

Even if the differences are only "some textures and the permeability of various walls", the potential exists that any individual in a group might become leader and that every individual might have a wildly different, uh, worldview. That information might only be a pile of binary switches, but the number of switches required for even a simple town could be immense. Making an interesting number of possibilities available isn't the same as making every possibility interesting. Without superhuman efforts on the designers' parts, players could accidentally box themselves into boring and broken worldviews by accepting another player's apparently positive solution to some apparently arbitrary quest.

First, there is no "apparently". The design world conceit is that your characters are part-divine, remember? When you look at a completed quest object, you know what it will do to reality.

Second, there isn't really a single switch for every quest from every person in every town. Quests affect a one-dimensional "mood score" for the people in town, and changes happen to a person's local environment based on their mood. In the painter example, rescusing his sister boosts his mood significantly, and boosts her mood even more so (so she actually shows up - in the original town the effect of her low mood is that she's invisible and her room is empty). If you boost the town's mood enough through fending off dangers that affect the whole town, the painter can start painting again, even if you never talk to him up to that point.

I'm all for devoting heavy development time to plots and stories rather than graphics and skills, but allowing a meaningful amount of change at the level you're describing requires an unbelievably robust sort of plotting and storytelling. Every new potential interaction must be checked against the list of every previous choice, and every interesting interaction -- the sort which realistically changes the world -- adds at least one  new setting to that list.

You say that like it's a bad thing. I only half joke.

But I seriously doubt even with this that the story will be within an order of magnitude of the complexity of the _code_ that runs current MMOGs.

--GF

[pxib]

I may have misunderstood your goals. Individual NPCs are interactive with the player, but oblivious to one another except as they relate to the "condition" of the town as a whole? The artist's sister for example, because she sits in her room all day, only interacts with NPCs by raising their mood on a sliding scale when quests they are interested in, related to her, are completed by the player? To invent an example: If the library is abandoned and ramshackle, that has no specific effect on other aspects of town life because the "Archlibrarian" and "Retired Wizard" who normally dwell there serve no purpose to the locals other than that they are happier (or unhappier) with those lovable old scamps around?

I agree that story complexity ought to be, objectively, less difficult even than coding a stable network architecture. As it stands, however, there are many people who have tackled the second... have written books about it, have published source code for workable examples, have become possessed of fat resumes and are ready to be hired for your project. I do not believe anybody has successfully tackled the first. Even tiny, single-player Infocom style text adventures trip over themselves when branching interactive plotlines have multiple characters.

I think we would be well served to have a proof of concept that wasn't also massively multiplayer, instanced or no.

[Glazius]

I may have misunderstood your goals. Individual NPCs are interactive with the player, but oblivious to one another except as they relate to the "condition" of the town as a whole? The artist's sister for example, because she sits in her room all day, only interacts with NPCs by raising their mood on a sliding scale when quests they are interested in, related to her, are completed by the player? To invent an example: If the library is abandoned and ramshackle, that has no specific effect on other aspects of town life because the "Archlibrarian" and "Retired Wizard" who normally dwell there serve no purpose to the locals other than that they are happier (or unhappier) with those lovable old scamps around?

That's pretty much the size of it, yes. Barring interaction by the player the world is largely static.

I say 'largely' because of course the devs can introduce in subsequent patches the 'major events of the apocalypse', like an ice wizard freezing the town over, or a plague. But even then these events are keyed to something the player will do, maybe even something so simple as a count of the 'world-affected' quest objects in the player's memories. Ontogeny recapitulating phylogeny and all that - a new player to the game will experience the world as it began, until through their actions they trigger the quest events that have affected the other players.

--GF