Friday, September 01, 2006
RPG Design Seed Challenge
We've had two very lively discussions on RPG design and theory this last week in comments about Innovation in RPGs and the Evolution of the Computer RPGs articles. I'd say the comments are a whole lot more interesting than the original posts, but that's not saying much. :)
However, I thought I'd take some of those ideas and run with them, and present some "relatively" innovative RPG design seeds based on some of the ideas we discussed, and see what other people can come up with.
Some basic thoughts (not restrictions, just ideas for playing with):
* Might Doesn't Make Right: One of the problems is that most RPGs can be won by maximizing your combat capability. The world can always be saved by someone with a strong enough sword-arm, apparently. Though he might need to do some fed-ex style missions to get to that point... which often coincidentally also require a good sword arm. Borrowing from Ultima IV, why not require something separate from combat skill required to win the game?
* Conflicted Development: Continuing to borrow from Ultima IV, why not have the different elements required to win the game be indirectly in conflict with each other? This prevents the player from simply maxing out each element - they all interact with each other.
* NPCs matter: RPGs (particularly the MMOs) have started to evolve to the point where all NPCs are either enemies, or quest / information dispensors or vending machines for trading gold and equipment. While making a believable simulation of a human is impossibly difficult, it's definitely possible to make characters a bit more interesting and serve more of a gameplay purpose... such as the foils in Wizardry 7, or even at the very least giving them the more interesting trading options of Ultima Underworld.
So here are three proposed game design "seeds" to start things out.
Gang Boss
You start out as a low-level solo criminal in a world of low-level gangs and several competing high-level crime organizations. Your goal is to rise to the top, and unite them all into a single organization under your absolute authority. You can commit your own heists to improve your own skill and amass a personal fortune, but real power comes from gaining authority over others. This comes from amassing fear and love from those you come into contact with. People may owe you favors; others might be blackmailed; still others may be bought or won over through your personal charisma (or gang of thugs). But beware - a quick, overt rise to power will gain you jealousy and hatred by others who will stop at nothing to take you down before you become too powerful. And there are always a few law enforcement officers and politicians who just refuse to be controlled, and are too dangerous to "remove."
As an added bonus, if you create this game, you can also pretend to have enough money to be lawsuit-worthy and send a copy of it to Jack Thompson. That'll guarantee you enough free press coverage to make a small fortune.
Aggressive Negotiations
Five small nations stand at the brink of war, a situation masterminded by a powerful (but unknown) nobleman and his sorcerous allies who intend to take advantage of the situation and seize control of the entire land in the ensuing chaos. You begin as a simple hero questing for adventure, but as your fame and power amass you find yourself being used as a pawn by these variouys forces in their preparations for war. You can either throw your weight behind one nation and help make certain that they emerge victorious (and foil the shadowy mastermind's plan for personal domination), or you can try and use your influence to back all five nations down from escalating hostilities, exposing the mastermind and returning the land to a state of peace.
There are no simple, straightforward "quests" to do this. You have only limited influence over several factors effecting each nation - their readiness for battle, their strength, their threats, their relationships with each other and smaller factions (including yourself), and even sub-factions within the nation with different agendas. Many of these factors run counter to each other - what might endear you to one king might anger another and bring him closer to attempting a "first strike" against the nation you are favoring. Slaying a feared dragon might win you the love and gratitude of the nation it threatened, but it might also free up their army to finally attack their neighbor.
Obviously, some sort of interface to help the player make sense of the current climate and choose their next action would be critical.
And of course, you'd still have to have the big, final, climactic battle with the shadowy mastermind and his sorcerous minions at the end. It's only fair to let the player to lay the smack down on the guy at the end.
Beyond The Pale
Borrowing a little bit this time from the Call of Cthulhu "dice and paper" RPG. There are evil, horrible, frightening creatures threatening humanity - and you are one of the few who can find them and face them. Unfortunately, facing them often costs you part of your soul or your sanity. And some of the evils horrors are far, far too powerful to be faced directly, no matter what your skills, spells, and equipment. You can only defeat them indirectly, and sometimes only for a while. And there's something truly monstrous on the horizon. You have to prevent its coming, before succumbing to madness, death, or despair.
This last design seed could be really tricky to pull off - but the main point is that the player can NEVER be powerful enough to face certain threats directly, and may often be in "over his head" where fleeing or avoiding combat might be the only reasonable option - he'll have to use indirect means to destroy the foes. In addition to this, greater power against the bad guys might come at some cost to his ability to resist them in some ways - some additional weakness (sanity, soul, humanity, whatever) that runs indirect counterpoint to his improved offensive capability. Or maybe a series of these indirectly-coupled capabilities. Maybe by making an arrangement with a vampire, he regains some measure of sanity by defeating a greater menace, but is now indebted to a powerful and not-too-friendly being. Constantly indebting yourself to other beings is not a long-term road to success.
Your Turn!
Okay, none of these would be "simple" RPGs to implement, I admit. But the mechanics I described here could be abstracted to some degree to keep things simple. After all, the virtues in Ultima IV were just simple numbers that you could check with one NPC (Hawkwind?) to see how you were doing on them.
And yes, I know how funny it is to talk about "innovation" and then talk about borrowing elements from other games in the same light. But in my opinion, all innovation is derived from someplace. We all stand on the shoulders of giants. As the saying goes, the secret to creativity is to hide your sources.
So ... anyone else want to take a stab at it?
Labels: Game Design, Roleplaying Games
- posted by The Rampant Coyote @ 10:37 AM (Permalink)
Comments:
Links to this post:
<< Home
I wish I had the time and budget to make Agressive Negotiations. Nice one. That concept just sends my inner architect/designer buzzing like crazy.
I'd like to mention one trick of conflicted development pulled by Jeff Vogel in his Geneforge line - spoilers, though.
Throughout the games there are canisters that increase your stats or abilities by altering your genetic code, cumulatively increasing your lethality to truly startling levels. But they also affect the endgames available to you, as each canister gradually reduces your humanity until the only ending available is that of a indiscriminate killing machine.
I'd also add that if you're interested in innovative game mechanics, you might do well to look at the indie tabletop RPG scene, where there have been a lot of interesting developments over the last few years. Some of them would make for better inspiration than borrowing - implementing a good CRPG Nobilis, mindblowing as that would be, or Polaris seems all but impossible.
Dogs in the Vineyard and The Riddle of Steel are better candidates for mechanic adaptation. Without going into all the possibilities from each, here's a look at the experience systems. In DitV, the player gains experience only when they sacrifice something to stay in the conflict - this is a clumsy description of an elegant game mechanic, but I have no better, short of giving a full example of play in a post already overlong. In RoS, the player gains experience from pursuing their stated goals - killing dragons may be a goal, but so could forging an alliance. Or raising your orphaned niece. Having different experience systems available enables different playstyles - Bethesda-style skill-based experience systems do some of the same things, but this lets you tie experience to plot while keeping variety an option.
If you're interested chiefly in game design, I'd start with Nobilis. It's older than the rest, and has been fairly influential. The price is startling, but the content far more so.
Throughout the games there are canisters that increase your stats or abilities by altering your genetic code, cumulatively increasing your lethality to truly startling levels. But they also affect the endgames available to you, as each canister gradually reduces your humanity until the only ending available is that of a indiscriminate killing machine.
I'd also add that if you're interested in innovative game mechanics, you might do well to look at the indie tabletop RPG scene, where there have been a lot of interesting developments over the last few years. Some of them would make for better inspiration than borrowing - implementing a good CRPG Nobilis, mindblowing as that would be, or Polaris seems all but impossible.
Dogs in the Vineyard and The Riddle of Steel are better candidates for mechanic adaptation. Without going into all the possibilities from each, here's a look at the experience systems. In DitV, the player gains experience only when they sacrifice something to stay in the conflict - this is a clumsy description of an elegant game mechanic, but I have no better, short of giving a full example of play in a post already overlong. In RoS, the player gains experience from pursuing their stated goals - killing dragons may be a goal, but so could forging an alliance. Or raising your orphaned niece. Having different experience systems available enables different playstyles - Bethesda-style skill-based experience systems do some of the same things, but this lets you tie experience to plot while keeping variety an option.
If you're interested chiefly in game design, I'd start with Nobilis. It's older than the rest, and has been fairly influential. The price is startling, but the content far more so.
# posted by 
: September 01, 2006 9:24 PM
: September 01, 2006 9:24 PM
One of the problems with Aggressive Negotiations is that there are so many possible game-states that I don't see how to begin without abstracting enough to turn it into a strategy game with an RPG minigame - like the unlamented Birthright CRPG.
One idea on how to handle this would be, taking a page from Ars Magica, to treat it as a career: the player is active for an episode every year or two, and directs his activity in general terms in the meantime, like the careers in Darklands or Traveler.
The time in between buffers continuity a bit - NPCs can die or leave town in the interval, and will naturally be commenting on recent events rather than (highly unpredictable) PC doings as each episode begins. This also lets the PC decisively solve the problem at hand without always having to introduce a probem behind the problem.
One of the things that King of Dragon Pass did very well was to build a large library of events with multiple varying preconditions - some would only happen if you had a senile noble on your clan ring, others if there was a stronger leader off the ring, others if you were feuding with one neighbour but trading with another, or if you had a worshipper of the trickster god on your ring... the list seems endless, but reduces down to a manageable list of conditions.
The result was that random events happened frequently, but always felt tied into what was going on - what the player had been doing. It increased the sense that PC actions had both meaning and consequences.
Additionally, several multi-arc events could occur, giving the sense of other people's stories wandering through your own, as well as a very few plot-fixed events that would occur every game - though whether they were historical backdrop or important to your clan's wellbeing often depended on what, if anything, you did about them.
This seems like a way that you could implement something like Aggressive Negotiations - every faction (and many subfactions) will desire to undermine their opponents; what they are capable of depends on the relative strengths involved. Those relative strengths, in turn, depend on which plots the player thwarted last time around, and which they let by, supported, or didn't notice.
One idea on how to handle this would be, taking a page from Ars Magica, to treat it as a career: the player is active for an episode every year or two, and directs his activity in general terms in the meantime, like the careers in Darklands or Traveler.
The time in between buffers continuity a bit - NPCs can die or leave town in the interval, and will naturally be commenting on recent events rather than (highly unpredictable) PC doings as each episode begins. This also lets the PC decisively solve the problem at hand without always having to introduce a probem behind the problem.
One of the things that King of Dragon Pass did very well was to build a large library of events with multiple varying preconditions - some would only happen if you had a senile noble on your clan ring, others if there was a stronger leader off the ring, others if you were feuding with one neighbour but trading with another, or if you had a worshipper of the trickster god on your ring... the list seems endless, but reduces down to a manageable list of conditions.
The result was that random events happened frequently, but always felt tied into what was going on - what the player had been doing. It increased the sense that PC actions had both meaning and consequences.
Additionally, several multi-arc events could occur, giving the sense of other people's stories wandering through your own, as well as a very few plot-fixed events that would occur every game - though whether they were historical backdrop or important to your clan's wellbeing often depended on what, if anything, you did about them.
This seems like a way that you could implement something like Aggressive Negotiations - every faction (and many subfactions) will desire to undermine their opponents; what they are capable of depends on the relative strengths involved. Those relative strengths, in turn, depend on which plots the player thwarted last time around, and which they let by, supported, or didn't notice.
# posted by : September 02, 2006 1:07 AM
: September 02, 2006 1:07 AM
Michael -
Dude, I feel for you man. I'm in the same boat. I'd love to do all three of these. I don't think any of them need to be done with a high budget or anything - I'm sure you could do all three with something as simple as RPG Maker. But they would all three take a lot of time and testing.
Dude, I feel for you man. I'm in the same boat. I'd love to do all three of these. I don't think any of them need to be done with a high budget or anything - I'm sure you could do all three with something as simple as RPG Maker. But they would all three take a lot of time and testing.
HC -
Wow! Lots of information, thanks!
On the Geneforge thing, I think that works fine *IF* the player has a clue as to what is going on when he does it. The lack of better ending opportunities at the end shouldn't come as a complete surprise.
I have been starting to look at tabletop indie RPG scene - mainly through reviews and accounts by others. Unfortunately, my FLGS doesn't carry anything like that, so it's all special-order (or DriveThruRPG.com ) to try anything really different. Nobilis is one I've heard a lot of jokes and praise for.
One option for implementation of something along the lines of "Aggressive Negotiations" would be to implement the rival nations (or perhaps the rival factions) as actual AI characters in the game. You could go so far as the route Cliff Harris did with Democracy to implement them via Neural Networks, but that is certainly not necessary. The independent nations could have a set of moves (some repeatable, some one-time commitments) available to them on perhaps a per-week basis.
More interestingly, there could be a lag between when they commit certain actions, and when the actions reach fruition, allowing the player (or rival nations / factions) to intervene. You'd have to be careful as the implementor not to overcomplicate matters - the player should be able to identify (via rumors, spies, whatever) what's going on in the game and feel he has enough understanding to intervene if necessary.
Wow! Lots of information, thanks!
On the Geneforge thing, I think that works fine *IF* the player has a clue as to what is going on when he does it. The lack of better ending opportunities at the end shouldn't come as a complete surprise.
I have been starting to look at tabletop indie RPG scene - mainly through reviews and accounts by others. Unfortunately, my FLGS doesn't carry anything like that, so it's all special-order (or DriveThruRPG.com ) to try anything really different. Nobilis is one I've heard a lot of jokes and praise for.
One option for implementation of something along the lines of "Aggressive Negotiations" would be to implement the rival nations (or perhaps the rival factions) as actual AI characters in the game. You could go so far as the route Cliff Harris did with Democracy to implement them via Neural Networks, but that is certainly not necessary. The independent nations could have a set of moves (some repeatable, some one-time commitments) available to them on perhaps a per-week basis.
More interestingly, there could be a lag between when they commit certain actions, and when the actions reach fruition, allowing the player (or rival nations / factions) to intervene. You'd have to be careful as the implementor not to overcomplicate matters - the player should be able to identify (via rumors, spies, whatever) what's going on in the game and feel he has enough understanding to intervene if necessary.
I'm reminded of recent discussions with colleagues, related to adding a "layer of abstraction" atop a combat-based wargame, where elements from the upper game feed directly back into the main gameplay zone. We were thinking about a game of nobility and brinksmanship, where your courtly machinations provide bonusses or unlock shortcuts to the other players representing the fighting masses "outside". I don't really have the vocabulary to describe it properly, but I'd be tempted to call it "dual-sourced gameplay". Stated better, you have two games within one. Both interact through very specific parameters, each penalizing, benefiting or informing the other.
A real example is the Dystopia Half-life 2 mod, where players can play the role of a hacker and are permitted to enter a separate cyberspace arena to hack the target building's defenses, thus allowing the rest of their team to progress within the level. The fireteams cannot accomplish their objectives without the hacker, and vice versa.
There are certainly examples of a negotiation meta-game within some MMORPGs. These are sadly not formalized within many of these games' mechanics, yet inevitably emerge as a meta-game between the players. One example is Planetside, where members of two opposing factions will secretly collude in order to annihilate the third. I've personally witnessed secret meetings, far from the main combat zones, where highly ranked members of each faction will face each other...and talk. Very often this would result in a temporary alliance, and the main forces of the armistice would literally ignore each other and collapse together upon the forces of the excluded faction. Fun stuff, and the designers could have done very well to implement incentives for this (they never did, but that didn't stop us). EVE Online also does an excellent job of nudging these player-driven mechanics along into a formalized system of treaties and bounties.
...NPCs can die or leave town in the interval, ...rather than (highly unpredictable) PC doings as each episode begins. This also lets the PC decisively solve the problem at hand without always having to introduce a probem behind the problem.
Well said! Having NPCs would solve plenty of problems in the dual-sourced game, just as you describe. NPC participation feeds a predictive stream into the gameplay, serving as an anchor that players require in their dealings. The "problem within a problem" is precisely why we don't see more of the dual-sourcing, yet I feel that balance issues could definitely be solved using more actor NPCs within the actual gameplay.
Coyote also said it well when he mentioned that NPCs are relegated to merely being vendor-questdelivery-mob units. Adding NPCs into the fray, as it were, is precisely the sort of idea that will move the MMORPG genre forward. At least that's my opinion.
A real example is the Dystopia Half-life 2 mod, where players can play the role of a hacker and are permitted to enter a separate cyberspace arena to hack the target building's defenses, thus allowing the rest of their team to progress within the level. The fireteams cannot accomplish their objectives without the hacker, and vice versa.
There are certainly examples of a negotiation meta-game within some MMORPGs. These are sadly not formalized within many of these games' mechanics, yet inevitably emerge as a meta-game between the players. One example is Planetside, where members of two opposing factions will secretly collude in order to annihilate the third. I've personally witnessed secret meetings, far from the main combat zones, where highly ranked members of each faction will face each other...and talk. Very often this would result in a temporary alliance, and the main forces of the armistice would literally ignore each other and collapse together upon the forces of the excluded faction. Fun stuff, and the designers could have done very well to implement incentives for this (they never did, but that didn't stop us). EVE Online also does an excellent job of nudging these player-driven mechanics along into a formalized system of treaties and bounties.
...NPCs can die or leave town in the interval, ...rather than (highly unpredictable) PC doings as each episode begins. This also lets the PC decisively solve the problem at hand without always having to introduce a probem behind the problem.
Well said! Having NPCs would solve plenty of problems in the dual-sourced game, just as you describe. NPC participation feeds a predictive stream into the gameplay, serving as an anchor that players require in their dealings. The "problem within a problem" is precisely why we don't see more of the dual-sourcing, yet I feel that balance issues could definitely be solved using more actor NPCs within the actual gameplay.
Coyote also said it well when he mentioned that NPCs are relegated to merely being vendor-questdelivery-mob units. Adding NPCs into the fray, as it were, is precisely the sort of idea that will move the MMORPG genre forward. At least that's my opinion.
The fighting games with command classes have been moving toward dual-sourced gameplay; the accounts of Chromehounds, for one, suggest that gameplay for a commander is so different from ordinary gameplay that they are effectively two different games which influence each other.
I'm not sure you need much AI for the nations in Aggressive Negotiations - a wide list of available plots and suitable preconditions, plus a random number generator, might do well enough. Better, in some ways - it's easier to write clever plots than it is to program a computer to do so.
If you have enough of them, you can let the nations plot in the interval between adventures, and use that to generate the backstory for each episode. Again, KoDP had some success with dynamically generating sagas based on event-seeds (http://www.geocities.com/bernuetz/kodp/saga01.html); it shouldn't be too hard to have one to three sentence summaries for different plot outcomes woven into the word on the street. Plus, that way the PC hears about some kingdom's action, and then hears something (ideally, not *quite* right) about something they were in on, and so forth.
Another approach might be to generate the details of plots procedurally - an "in the library, with the knife, 'twas Miss Scarlet!" approach. There will always be an identifiable list of plausible targets, and an identifiable list of means to reach those targets.
This can continue to random variation - if one nation is bribing an outpost sergeant to open the gates, the investigation and thwarting of this ploy play out differently if said sergeant is living the high life on these funds, paying for his sick mother's treatment, or secretly a double (or triple) agent. I don't see how to make these variations both many and well-written,though.
I ordered Nobilis out of curiousity and on a whim; it proved, in almost every possible respect, to be not your normal RPG, from the concept to the physical instantiation. It was one of the nicest surprises of the year. Actually, I'm kind of curious - what were the jokes about Nobilis like?
I'm not sure you need much AI for the nations in Aggressive Negotiations - a wide list of available plots and suitable preconditions, plus a random number generator, might do well enough. Better, in some ways - it's easier to write clever plots than it is to program a computer to do so.
If you have enough of them, you can let the nations plot in the interval between adventures, and use that to generate the backstory for each episode. Again, KoDP had some success with dynamically generating sagas based on event-seeds (http://www.geocities.com/bernuetz/kodp/saga01.html); it shouldn't be too hard to have one to three sentence summaries for different plot outcomes woven into the word on the street. Plus, that way the PC hears about some kingdom's action, and then hears something (ideally, not *quite* right) about something they were in on, and so forth.
Another approach might be to generate the details of plots procedurally - an "in the library, with the knife, 'twas Miss Scarlet!" approach. There will always be an identifiable list of plausible targets, and an identifiable list of means to reach those targets.
This can continue to random variation - if one nation is bribing an outpost sergeant to open the gates, the investigation and thwarting of this ploy play out differently if said sergeant is living the high life on these funds, paying for his sick mother's treatment, or secretly a double (or triple) agent. I don't see how to make these variations both many and well-written,though.
I ordered Nobilis out of curiousity and on a whim; it proved, in almost every possible respect, to be not your normal RPG, from the concept to the physical instantiation. It was one of the nicest surprises of the year. Actually, I'm kind of curious - what were the jokes about Nobilis like?
# posted by : September 04, 2006 2:33 AM
: September 04, 2006 2:33 AM
Merely a command structure layer for a wargame wasn't *quite* what I was thinking about, but I can see how one would go there based on what I was saying (blame my poor writing). Brinksmanship, the dance of nobility, higher-order socialization, and non-combat questing are all sorts of things that can affect a battle before it's even begun! I used those wargame examples to illustrate the conceptual idea of dual-sourced, but as I mentioned, the specifics and gameplay would vary widely. I've never seen those types of things woven into a game before. Perhaps you can correct me. Anyhow...
From the A.I. side of things, you are absolutely correct in that it's not required. Yet, in MMORPGs that depend on inter-player interaction much as what has been described in Coyote's Aggressive Negotiations, I'd say NPC actors would augment the experience greatly. Particularly in terms of building popularity for the game (i.e. lack of players playing the game), and in terms of filling in the content when player activity is generally low (e.g. late at night). These are all shortcomings of MMORPGs that are struggling in the face of the WoW juggernaught. An inter-player mechanic needs players! On a side note, the A.I. henchman of Guildwars were the only thing that kept me playing the game for as long as I did. Finding a properly configured group is damned hard sometimes!
So yes indeed, your comments are very interesting, as I'd never really looked too deeply into KoDP. I will try it for sure. I can't comment on what the game is, or what it's about, but with a sling full of plot elements that are randomly mixed and matched doesn't *quite* appeal to me too much, unless the writing and ideas are really, really good. And even then, the narrative tends to suffer from a rather unwieldy clobbering. Does it become predictable from a player experience?
From the A.I. side of things, you are absolutely correct in that it's not required. Yet, in MMORPGs that depend on inter-player interaction much as what has been described in Coyote's Aggressive Negotiations, I'd say NPC actors would augment the experience greatly. Particularly in terms of building popularity for the game (i.e. lack of players playing the game), and in terms of filling in the content when player activity is generally low (e.g. late at night). These are all shortcomings of MMORPGs that are struggling in the face of the WoW juggernaught. An inter-player mechanic needs players! On a side note, the A.I. henchman of Guildwars were the only thing that kept me playing the game for as long as I did. Finding a properly configured group is damned hard sometimes!
So yes indeed, your comments are very interesting, as I'd never really looked too deeply into KoDP. I will try it for sure. I can't comment on what the game is, or what it's about, but with a sling full of plot elements that are randomly mixed and matched doesn't *quite* appeal to me too much, unless the writing and ideas are really, really good. And even then, the narrative tends to suffer from a rather unwieldy clobbering. Does it become predictable from a player experience?
Well, as I often say - if you give even a detailed game concept (which these weren't) to three different developers, you will end up with three totally different games.
The jokes I heard about Nobilis were concerning how nobody seems to understand exactly how to play it.
One of the keys with dual-sourced gameplay (or even multi-sourced) is what Noah Falstein referred to as "Emergent Complexity." Basically, from my perspective on what he's said, if you can take two different gameplay mechanics that are otherwise simple on their own, and interweave properly in that they both interact with each other, you can create a game thatcan be much, much more involving, deep, and FUN than simply creating those two games separately.
As to procedurally generated events - I'm a fan, though I would try and use them sparingly. I played lots of Daggerfall with procedurally-generated missions. While it was a lot of fun, it didn't take long before players got bored with the mad-libs style missions that followed very obvious formulas. Part of the trick to doing them right, I think, is to "show don't tell" in this respect. Throw in clues as to what the game is trying to make happen, and the player will fill in the blanks himself. Much as they do with the conversations in The Sims. The abstract "simlish" conversations are much more compelling and even more believable to players than trying to generate realistic conversations in a human language.
The jokes I heard about Nobilis were concerning how nobody seems to understand exactly how to play it.
One of the keys with dual-sourced gameplay (or even multi-sourced) is what Noah Falstein referred to as "Emergent Complexity." Basically, from my perspective on what he's said, if you can take two different gameplay mechanics that are otherwise simple on their own, and interweave properly in that they both interact with each other, you can create a game thatcan be much, much more involving, deep, and FUN than simply creating those two games separately.
As to procedurally generated events - I'm a fan, though I would try and use them sparingly. I played lots of Daggerfall with procedurally-generated missions. While it was a lot of fun, it didn't take long before players got bored with the mad-libs style missions that followed very obvious formulas. Part of the trick to doing them right, I think, is to "show don't tell" in this respect. Throw in clues as to what the game is trying to make happen, and the player will fill in the blanks himself. Much as they do with the conversations in The Sims. The abstract "simlish" conversations are much more compelling and even more believable to players than trying to generate realistic conversations in a human language.
Daggerfall was also in my mind, as I thought about the difficulty of good AI plots. I think things would be even harder in an MMORPG - good conversation is hard to fake, and critical to an actor. Eliza and similar one-trick ponies aside, can one do good AI conversation yet?
If you've played it as much as I have, you've seen most KoDP random events before. That doesn't mean that you know the right answer - most of them don't have "right" answers, though they always have several good roleplay answers.
The key, I think, to avoiding predictability is having a large number of events, and being able to add more with ease. Most of Galactic Civilization's mods are simpler versions of this kind of random event addition, and whether the further writing comes from staff or the players, new material would keep things from seeming repetitive.
The other thing is that, in KoDP, you're running an Iron Age clan. Things are supposed to be repetitive, within limits. The seasons come and go, and people live more or less as their ancestors did. Done right, what repetition there is gives a feeling of connection, rather than boredom.
No one knows how to play Nobilis? That's... appropriate. It is not so much complex as requiring a certain mindset from all involved.
Simlish is a great example... but how would you 'show, not tell' a Daggerfall-style quest? Because I think that's what you'd need to have a full range of plots.
Emergent complexity sounds like a useful design priciple.
If you've played it as much as I have, you've seen most KoDP random events before. That doesn't mean that you know the right answer - most of them don't have "right" answers, though they always have several good roleplay answers.
The key, I think, to avoiding predictability is having a large number of events, and being able to add more with ease. Most of Galactic Civilization's mods are simpler versions of this kind of random event addition, and whether the further writing comes from staff or the players, new material would keep things from seeming repetitive.
The other thing is that, in KoDP, you're running an Iron Age clan. Things are supposed to be repetitive, within limits. The seasons come and go, and people live more or less as their ancestors did. Done right, what repetition there is gives a feeling of connection, rather than boredom.
No one knows how to play Nobilis? That's... appropriate. It is not so much complex as requiring a certain mindset from all involved.
Simlish is a great example... but how would you 'show, not tell' a Daggerfall-style quest? Because I think that's what you'd need to have a full range of plots.
Emergent complexity sounds like a useful design priciple.
# posted by : September 04, 2006 3:22 PM
: September 04, 2006 3:22 PM
but how would you 'show, not tell' a Daggerfall-style quest? Because I think that's what you'd need to have a full range of plots.
I'm not quite sure. You'd need to let the player fill in the blanks somehow. I've thrashed a few ideas out with some colleagues, but never came up with a definitive answer. But thinking aloud (or visibly?), here are some thoughts.
You have a simple, procedurally generated quest:
Faction A asks the player to go to Dungeon X and retrieve Artifact Y. Oh, and Faction B will try and stop the player and take the item for themselves.
With Daggerfall, that's pretty much the end of it. I don't remember if you'd have faction B trying to kill you for it afterwards, but if it did, it might be some random attack as you left the dungeon (City of Heroes has that for a couple of missions, too).
So that's the "what's" of a quest, but not the "Why's". And that's why things look like a cardboard cutout to the player - it's very obvious that there are no "why's" to it, it's just a randomly generated quest. It's very clear there's no rhyme or reason to it.
Now, the next level deep might be to suggest some reasons, without going overboard and trying to construct a complex, believable story (because computers SUCK at creating believable stories or characters). But just drop a few hints that seem to imply an internal consistency:
#1 - Setup
Pre-generate parts of missions, or maybe have missions build on events generated in-game. Perhaps the event could be, "Artifact Y has been found." And people are suggesting what factions might be interested in it.
#2 - Quest Creation
Bring in some of the existing game state. Rather than randomly choosing the conflicting faction, choose one that is a current rival / enemy based on the game state. Choose objective of mission - if possible (not all the time) from something from the pool of pre-generated events.
#3 - Quest Execution
Suggest some backstory that may or may not be there. Maybe on the assassins from Faction B, the player might discover a note informing the assassins who they are hunting, where to ambush him (which happens to be exactly where he was ambushed), and perhaps some additional details. In fact, those details can be a self-fulfilling prophecy. Maybe they didn't exist until the moment the player reads the note.
#4 Quest Follow-Up
The results of the quest shouldn't be forgotten by anyone very quickly. In fact, the results should "create" some new rumors and dynamic changes in the world. For example, completion of the quest could have two results: Faction B hates Faction A even more, increasing their rivalry. Or Faction B has chosen to re-open diplomatic relations with faction A, out of fear of their newfound power. These changes instantly generate rumors as to what's going on, but short on details.
Yeah, these suggestions are vague and could quickly evolve to being unmanageable. But if something like this could be done (and I'm not offering... yet...), it might suggest a level of depth and complexity even beyond what really exists. And while things like the relationships between factions A and B might be no more than a single floating-point value, this might invite the player to read far more into it than is really there.
And, if the player is like me, I'll happily throw disbelief out the window if the game can pull it off, even if I *have* seen what's behind the curtain.
I'm not quite sure. You'd need to let the player fill in the blanks somehow. I've thrashed a few ideas out with some colleagues, but never came up with a definitive answer. But thinking aloud (or visibly?), here are some thoughts.
You have a simple, procedurally generated quest:
Faction A asks the player to go to Dungeon X and retrieve Artifact Y. Oh, and Faction B will try and stop the player and take the item for themselves.
With Daggerfall, that's pretty much the end of it. I don't remember if you'd have faction B trying to kill you for it afterwards, but if it did, it might be some random attack as you left the dungeon (City of Heroes has that for a couple of missions, too).
So that's the "what's" of a quest, but not the "Why's". And that's why things look like a cardboard cutout to the player - it's very obvious that there are no "why's" to it, it's just a randomly generated quest. It's very clear there's no rhyme or reason to it.
Now, the next level deep might be to suggest some reasons, without going overboard and trying to construct a complex, believable story (because computers SUCK at creating believable stories or characters). But just drop a few hints that seem to imply an internal consistency:
#1 - Setup
Pre-generate parts of missions, or maybe have missions build on events generated in-game. Perhaps the event could be, "Artifact Y has been found." And people are suggesting what factions might be interested in it.
#2 - Quest Creation
Bring in some of the existing game state. Rather than randomly choosing the conflicting faction, choose one that is a current rival / enemy based on the game state. Choose objective of mission - if possible (not all the time) from something from the pool of pre-generated events.
#3 - Quest Execution
Suggest some backstory that may or may not be there. Maybe on the assassins from Faction B, the player might discover a note informing the assassins who they are hunting, where to ambush him (which happens to be exactly where he was ambushed), and perhaps some additional details. In fact, those details can be a self-fulfilling prophecy. Maybe they didn't exist until the moment the player reads the note.
#4 Quest Follow-Up
The results of the quest shouldn't be forgotten by anyone very quickly. In fact, the results should "create" some new rumors and dynamic changes in the world. For example, completion of the quest could have two results: Faction B hates Faction A even more, increasing their rivalry. Or Faction B has chosen to re-open diplomatic relations with faction A, out of fear of their newfound power. These changes instantly generate rumors as to what's going on, but short on details.
Yeah, these suggestions are vague and could quickly evolve to being unmanageable. But if something like this could be done (and I'm not offering... yet...), it might suggest a level of depth and complexity even beyond what really exists. And while things like the relationships between factions A and B might be no more than a single floating-point value, this might invite the player to read far more into it than is really there.
And, if the player is like me, I'll happily throw disbelief out the window if the game can pull it off, even if I *have* seen what's behind the curtain.
Excellent food for thought, guys.
I don't have time to craft a worthy response, but if you haven't already go check out The Singular. It's a pre-alpha web-based adventure game, co-created by everyone's favourite joystiq blogger, m3mnoch (no affiliation).
Why this is relevant is because you can create quests for other players to enjoy. It's all text based, but I honestly don't see why pictures (or other types of data or meta-data) can't gradually be incorporated.
Post a Comment
I don't have time to craft a worthy response, but if you haven't already go check out The Singular. It's a pre-alpha web-based adventure game, co-created by everyone's favourite joystiq blogger, m3mnoch (no affiliation).
Why this is relevant is because you can create quests for other players to enjoy. It's all text based, but I honestly don't see why pictures (or other types of data or meta-data) can't gradually be incorporated.
Links to this post:
<< Home
