Sign in to follow this  
Vensild

Designing a living environment

Recommended Posts

I am currently writing my master thesis in computer science. The subject is "what makes a computer rpg appear living to the player(s)". So my focus is not how to make a good game, or how to make a game that'll make a lot of money. I am asking for some help concerning litterature about this subject, because most of what I have found is more technical, thus as discussing skills, combat systems, how to handle 1000 players at the same time, 3d-engines etc. Not that this things are not interesting or not important. What I mean with "appear living" is a bit vague. It is of course an internal expirience for the player, and therefore difficult to define precisely. But it concerns the way you, as a player, can interact with the surronding environment, and how it reacts to you. Is it possible for you to ask npcs about what ever you want, with relevance to the game world? Do npcs remember previous interactions with you? Do you get a feeling of this world is functioning, and can you be a part of it? So with "appear living" I mean something close to "believable on its premises". So i would really appreciate some help in my search. I currently looking at the following books "Designing virtual worlds", by Richard Bartle "Swords and circuitry: A designers guide to computer role-playing games", by Neal Hallford And of course a lot of articles. An example: Mu’s unbelievably long and disjointed ramblings about RPG design: http://mu.ranter.net/theory/printversion.html I am also looking at litterature about HCI and at litterature about storytelling in books and movies. Any help or questions are appreciated! Regards Vensild p.s. I am sorry if my English is not correct, but its been a while since I have written in anything else than my native language.

Share this post


Link to post
Share on other sites
Your English is fine.

I think what you're getting at here is one of the main concepts behind Primogen - creating a functioning simulation in which the game takes place.

It's the background action of an environment (the underlying simulation) that makes it seem lifelike - things like the stations in the X space trading games operating amoungst themselves are a perfect example.

You should however support your answer (in your thesis) by highlighting what make a CRPG seem 'dead' - NPCs that never move from their post, static prices for goods, no people in a city going about their business etc. etc.

CRPG's are only just starting to implement these simulations. For the most part, they're irrelevant to the game story so there's been no priority to expend resources developing them. Now however, graphics are approaching photorealistic (at least as far as realtime perception is concerned) and further means of increasing immersion are required - the background activity is one area where things are being developed seriously - both in terms of interaction (destructible environments, exploitable physics) and set behaviours.


Share this post


Link to post
Share on other sites
this might not answere your question to the T, but its something that popped into my head while reading
Is that in creating a living world is to create a beleavable world from a Artistic point of view

Look at the movies, Labrynth and Dark Crystal by Jim henson. look at the old starwars films and see how they took a earthly set and created a beleavable world with its own life and culture, from a artsitic standpoint, And I think that following there path would help create a video game with a living atmostphere and not a stagnent video game. (Even cronicles of riddick are a good example, and matrix of what a living new world would be like)

As Far as the NPC's would go, that all depense on the AI you program, and what your thinking of would break the bounds of traditional AI and give it more life. Like creating the AI to have personality and the ability to give responses to a players question where it draws apun his own memory and what kind of personality he has, is he the type to give a responce, would he lie, would he just walk away and ignore you, etc, plus when there not talking what basic behavors do they have, will they start fights, steal, get drunk, go to work, give to the poor or beg themselves

But beyond people and visual effects, Government and society also are a key factor. what kind of government do they have and what kind of society do they have, if its a fantasy setting weather sci fi or high fantasy like lord of the rings there usualy is a government, if not if its tribal than so be it. Star Wars has its own government, in fact it has many, in the old star wars it was the Empire had rule.

Share this post


Link to post
Share on other sites
Quote:
Original post by _winterdyne_
Your English is fine.

You should however support your answer (in your thesis) by highlighting what make a CRPG seem 'dead' - NPCs that never move from their post, static prices for goods, no people in a city going about their business etc. etc.


The result of my thesis shall be a model, method or set of demands for evaluating CRPG worlds/environments in respect to how living/belevable they are, so The above example schould hopefully be labeled as a "dead" feature.

Quote:
Original post by _winterdyne_
CRPG's are only just starting to implement these simulations. For the most part, they're irrelevant to the game story so there's been no priority to expend resources developing them. Now however, graphics are approaching photorealistic (at least as far as realtime perception is concerned) and further means of increasing immersion are required - the background activity is one area where things are being developed seriously - both in terms of interaction (destructible environments, exploitable physics) and set behaviours.


I think that complexity is partly responsible for this sad development in CRPGs. Most game produces think it is easier to do technical tweaking than describe and code the complexity of behavior (political/economical etc). They go for the safe bet: better graphics, bigger weapons and more number crunching. The result is less focus on game mechanics execpt from combat. I hope the ongoing development against living backgrounds etc is continuing to the level where a player cant nessicerialy tell the difference between other players and NPCs. It will of course take a while before this wont be revealed the momont you start to communicate.
I think emmersion is a keyword in the design and development of good gaming expiriences in complex and living worlds be it sci-fi, real-world or fantasy rpg.

Quote:
Original post by Manic_Gamer
As Far as the NPC's would go, that all depense on the AI you program, and what your thinking of would break the bounds of traditional AI and give it more life. Like creating the AI to have personality and the ability to give responses to a players question where it draws apun his own memory and what kind of personality he has, is he the type to give a responce, would he lie, would he just walk away and ignore you, etc, plus when there not talking what basic behavors do they have, will they start fights, steal, get drunk, go to work, give to the poor or beg themselves

But beyond people and visual effects, Government and society also are a key factor. what kind of government do they have and what kind of society do they have, if its a fantasy setting weather sci fi or high fantasy like lord of the rings there usualy is a government, if not if its tribal than so be it. Star Wars has its own government, in fact it has many, in the old star wars it was the Empire had rule.


The AI part you describe are being tried in the game Seed, which is developed by Runestone Gamedevelopments. I have´nt had a look at it yet, but I have talked a bit about this subject with them. And to this is a prerequsite for a living environment that the NPCs you interact with/encouter have behavior based upon background/current situation/personality etc
Government or another system for behavior control/rules is of course also neccisary (<-- cant spell that word, sorry).
If an NPC is going to act and communicate in a meaningful way in a world it needs (imo) the 6 following features:
1) motivation, goals
2) knowledge of the mechanics in the world
3) memory
4) the ability to act
5) the ability to communicate about all objects (and there relations) it has knowledge about
6) basic needs that needs to be fulfilled, like hunger, sleep, powerlust etc

I'll look through my notes and find a more precise definition of this.

Thank you for your input!

Regards Vensild


Share this post


Link to post
Share on other sites
Guest Anonymous Poster
I'll suggest you take a look at some theory about realism in literature. For literature, there is a much wider body of sensible theory to relate to than for games.

For example, realism in fiction is partly related to the presence of incidental detail. Eg mentioning background things or events "just because they are there", rather than because they relate directly to the unfolding plot of the novel or short story. This notion translates directly into game environments. If in a a game you go to a supermarket, it increases immersiveness if there is stuff that makes it look right: carts, customers, wares, etc. This background clutter might all be largely irrelevant to the gameplay aspect in the supermarket - say if you are supposed to shoot somebody or talk to somebody.

Apart from the presence of indicental detail, also consider the logic cohesiveness of the world. It is more consistent if the person you have to find/kill is the supermarket manager (who has a reason to be there) than a cirkus clown or Nazi general (who doesn't).

I'm sorry I don't have specific titles to recommend, but at least it's a direction for your literature searches.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
Given your comment about a world being "believable on its premises", Id suggest that the term you are most often going to come across in relevant literature will be "suspension of disbelief".

Searching on that term might find more information about design and realism rather than technical documents.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
Quote:
Original post by Vensild
I am currently writing my master thesis in computer science.
The subject is "what makes a computer rpg appear living to the player(s)". So my focus is not how to make a good game, or how to make a game that'll make a lot of money.


.



One thing that is lacking in many games is an appearance of having an effect on the world. This can be as simple as having the results of the players actions persisting (like cutting down a tree and it being still chopped down the next time you log on). Of course any action must take an appropriate amount of effort (some people will chop down all you forests just for fun if you let them -- especially if you make it easy). World plotlines with quests and events and outcomes (based on player actions) that change aspects of the world (scoping down to small local areas) can give a feel of accomplishment or even just existing.

The mechanisms to do something like this demands a much more dynamic game world than what we have seen.
Trees as distinct objects would be relatively simple. A hierarchy of social/economic/political systems with dynamic/interactive (and self balancing) aspects is much much harder -- especially with 1000+ anarchists doing their best to destroy/warp the reality you are trying to create.

A largely self balancing game world will need systems of realistic repercussions, self correcting mechanisms and 'corrective' measures that game masters might employ (there always will need to be some manual guidance...)



Share this post


Link to post
Share on other sites
Quote:
Original post by Anonymous Poster
...
The mechanisms to do something like this demands a much more dynamic game world than what we have seen.
Trees as distinct objects would be relatively simple. A hierarchy of social/economic/political systems with dynamic/interactive (and self balancing) aspects is much much harder -- especially with 1000+ anarchists doing their best to destroy/warp the reality you are trying to create.

A largely self balancing game world will need systems of realistic repercussions, self correcting mechanisms and 'corrective' measures that game masters might employ (there always will need to be some manual guidance...)


The self-corrective mechanisms are the issue - it's these that are unrealistic, generally through necessity. Consider a forest of trees, even using accelerated game time, it may still take a long time for it to spawn new saplings, and mature them to harvestable trees. This may result in the 'spread' of a forest area once lumberjacking stops. This almost certainly will disrupt game balance to some extent (fine if you're running a social experiment, not so fine if the objective is for everyone to have enough fun to pay you money each month). Eventually we're left with limiting forest growth to a particular area, which in effect becomes a standard respawn cycle.

Truly dynamic geography is not a particularly great idea for an open-ended game - there's too great a chance of the game state losing its balance. Multiplayer games where there's a possibility of deliberate sabotage of the equilibrium make this problem worse. Manual management of the system to counter spurious changes to gamestate is an expensive solution.

So, the system has to self regulate, meaning we have to make some tradeoffs between realism and ease of management. It's actually easier to have players leave 'footprints' in game records by using dynamic tasks that are filtered through the regulatory system - and to have that system in part based on a 'floating' economic model - we allow the balance of the game world to shift between two or more (not very extreme) limits, and use a heuristic to drop nasty events on the better-off, and helpful events on the worse-off, acting as 'nudges' to the balance of the game. Nothing too obvious like 'you cannot mine any more gold', but something else like 'an earthquake destroys your gold mine at X'. The precise nature of the event doesn't matter, as long as it disrupts things enough in a particular direction.


Share this post


Link to post
Share on other sites
I believe that the most important thing that makes characters live in any medium is drama. Drama has become a little misunderstood throughout the ages but the Greeks were the ones to have really researched into this.

They asked: "What makes one play emotionally engaging and the other not?"

An important comparison was made between two plays.

1. the protagonist would go through the most extreme situations.
-His house would collapse
-his wife would leave him
-he would lose his job
-his country would be invaded

2. the protagonist would go through mild situations
-he would lose some money on the market
-he would complain to the city guard
-he hit his toe on a rock
-he tries not to show he's hurt

And it appeared that people were much more moved by the second version of events. It is for more reasons than you might assume.

One reason is believability and I'm sure you thought of that.
Another is drama, which is 'reaction to change'.

He loses money and blames someone else and he tries not to show he's hurt. It's there that you see character, someone's reaction to things. In the first example, he never has time to react to it all and it just becomes a pile of misery.

I think that game storylines have often suffered from their creators not understanding this principle. That's why there's so many 'safe the earth' extreme storylines that have little effect.

The player, to a degree, gets to make his own drama. So what's more interesting to focus on is her companions and those she inflicts situations upon.
If she steals from a tavern owner, will that make him less trusting of his customers? If she steals repeatedly, will he hire a guard or quit his business?
When she steals, will her companions try to stop her? Will they leave the group out of distaste? Will they alert the guard or won't they out of respect of their former friendship? Or will they say they won't say anything so they can get a guard right away and get you to jail where you belong?


Actually I think I know one of the most interesting situations where you can make a computer RPG appear living.

There is a border that refugees want to cross, but only some are allowed through. Maybe the land has been under repeated attack, maybe there's a disease or a ruthless king or dictator. This can be set in any setting, fantasy/scifi/modern.

At that spot, you have a huge queue that moves on once in a while when the guards decide they let people through or not. The scene might start with the player (optionally with companions) waiting in line, not far from the guards. This gives time to set the scene and catch the mood of the queued people as well as the guards. Then the player is not admitted on a trivial or unfair reason. The player can then leave, try again getting rid of 'those ugly boots' or whatever reason they gave him or instigate a border run with the other rejectees. Or help them accept their fate. Or...

It's a situation where people are forced to interact. The moving bustle of the queue makes it a dynamic lifelike situation (even when the graphics is only Ascii). And it's a situation with various solutions and much chance for drama.

I hope this helps you.

Share this post


Link to post
Share on other sites
Dunam, your comments really made my day. That border atmosphere setting is truly a center of interaction and emotional impact, good example.

I completely agree that situations do not require megadramatic, cataclysmic events to captivate interest. Simple, everyday events with deep emotional ties can provide a much richer, more living environment.

Not that the huge events can't do this as well, but when everything is a chain of world-shaking events, they lose flavor.

Large scale events need to be built up to with careful plot building. Killing off a character is not the only way to provoke emotions from a player.

Just some thoughts.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
Quote:
Original post by Humble Hobo
Dunam, your comments really made my day. That border atmosphere setting is truly a center of interaction and emotional impact, good example.

I completely agree that situations do not require megadramatic, cataclysmic events to captivate interest. Simple, everyday events with deep emotional ties can provide a much richer, more living environment.

Not that the huge events can't do this as well, but when everything is a chain of world-shaking events, they lose flavor.

Large scale events need to be built up to with careful plot building. Killing off a character is not the only way to provoke emotions from a player.

Just some thoughts.




The problem is then how many of these events can be staged (especially if they are intricate and fitted manually) so that the player doesnt do the same one every other day (and even if there are many, eventually running out of them).

They already cant afford enough GMs to even help queued people within 30 minutes , so either we have a static set of choreographed 'events' or some system that generates them 'on the fly' (possibly assisted by tools to make a few GMs more effective in some coordinating role).

Previous threads have talked about automatic quest generators and I had added comments about integrating with a larger world plot mechanism to continually change the mix of elements driving the situations. Another mechanism allowed for modular quest templates so that over time new ones could be added (including some supplied by players). Another component would be a server engine that supported expanding/reusing parts/areas of the world to stage new situations
as they are needed (filling in props to match the current quest template and then fading it out afterwards to then be reused for another).



Share this post


Link to post
Share on other sites
When talking about quests in games, I think there are three viable ways to go:

1. Fully scripted quest (I liked the quests in fallout a lot)
2. Reusable quests (missions where the action is fun and logical enough that it would require a repeat, such as escort missions or 'reduce wolf population' etc.)
3. GM controlled quests (time intensive for GM's)

Any other type I haven't seen done well yet (but let me know if you think I've missed something and which game I should check out).


BTW the best way to make a play seem lifelike in my experience has by having some people act rude or selfish. Nothing creates a sense of reality more where every character is selfish. Even though I do not think the world is like that, people tend to take selfish people as 'realistic'. Try it.

Share this post


Link to post
Share on other sites
I personnaly think that you could use a sort of special unlockable profession to make the world spin normally...

Let me explain.

People who usually play Paladin, as a sort of White Knight compulsive disorder, desperately needing to help widows and orphans, would positively JUMP on this sort of things.

To better exemplify this, I would quote Terry Pratchett's "Monks of Balancing", who are supposed to work in secret, and simply move small brass weights over the world, just to make sure it spins properly and doesn't tumble. To anyone exterior, it may appear useless, but it seems to work, since the Discworld hasn't tipped over yet...

Now, if you had a class, or maybe a sort of secret society using players' accomplished GM-generated missions to balance the world, I think it would allow for some leeway in the believability of the thing. Simply knowing you are on a mission to help the world balance out gets the player to accept stupid missions he would simply disregard if he wasn't already taking part in something greater.

Let's take the exemple of the "help reduce the wolf nuisance in Koom Valley" mission. No one would seriously consider a mission, but it could be accomplished through different ways... You could skin some wolves, or more rabbits the wolves feed on, or maybe set fire to preset areas, so that the taming areas, or raising areas are off for a while, and the respawn rate is lower, or anything in accordance to the different types of classes you may have. You could also plant some poisonous seeds in the areas known to home rabbits, if you were a ranger or a herbalist of sorts...

But you could also have some different societies... Maybe have the "secret monks of History", who make sure things happen. Or not, depending on what things are deemed to pertain to the REAL History... In terms of gameplay, it would simply mean generating PVP assassination missions, as counter-measures for PVE assassination missions. You simply give some backstory in the lines of "Mr. Bloodaxesonnephewson is on a rampage against the marauders living in Dangerdoom cave. But the marauders are supposed to attack a caravan in some days, and this caravan is supposed to bring weapons to that country, which will bring a stable peace to the area for a while. Should you fail to stop Mr. Bloodaxesonnephewson before it's too late, war would destroy the country for at least two years."

On the other hand, you don't have to actually make the things HAPPEN, since there is an almost constant respawn of baddies everywhere, including said cave. You can always use a corny explanation of what happened depending on your results in said counter-mission...

And the nice benefit is that you can always use said GM-generated quests to rid the world of those pesky griefers, or abusive gremlins. You can even create a mission to solve the abusive crafters problem, if you include complete looting of the preys...

Share this post


Link to post
Share on other sites
Quote:
Original post by Fournicolas
Now, if you had a class, or maybe a sort of secret society using players' accomplished GM-generated missions to balance the world, I think it would allow for some leeway in the believability of the thing. Simply knowing you are on a mission to help the world balance out gets the player to accept stupid missions he would simply disregard if he wasn't already taking part in something greater.


I really like how you expanded on the wolf population thing, to me it was just an excuse to get you 'killing wolfs'.

However I think the thing above would not have the player 'accept stupid missions'. He'll accept them for the XP. He'll accept them for the reward. But I doubt many players would find such a mission fun to do, since it's essentially just woving around the world which is ussually the most boring type of quest, especially when it's a repeat quest.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this