Game News Role-playing in Dragon Age

[Vault Dweller]

Tags: BioWare; Dragon Age

Here is another Dragon Age update for you, courtesy of David Gaider:
<br>
<br>
<blockquote>I'm not sure that I equate roleplaying with there needing to be every possible choice available. You will be able to do what is in context with the story, of course. It's not about running around the world and doing whatever comes to mind -- there are specific goals at hand.
<br>
<br>
As far as individual quests, there will generally be options as to how to complete them but there's never going to be every option. I remember in the beginning of Hordes there was an encounter with some kobolds holed up in a bar that had something like six different methods of completion -- and while it was an interesting experiment, that one tiny encounter required so much work it was almost mind-boggling and almost got cut because the complexity introduced too many bugs. So the idea, I think, is to strike a line somewhere in the middle.
<br>
...
<br>
Ah, yes, it was in SoU. I always mix those up. And there were only two consequences, if I remember right -- either you saved the kobolds or you didn't. There were just many ways of saving them or not saving them, probably more then were really necessary. In terms of the roleplaying it offered, it was very much in the neat-but-over-the-top category.
<br>
...
<br>
No, we have no rule that says there must be X options for every quest. We include whatever seems appropriate.
<br>
...
<br>
I'm no more interested in forcing such options if none seem necessary than I am in forcing a good and evil "path" onto every quest just to have it. Some quests are very simple and they only have one manner of completion, others have more -- others have a lot more. If you want to write a game where there is a design rule of 3 options minimum no matter what -- be my guest.
<br>
...
<br>
The main character does indeed do all the talking in DA, being the protaganist of the story. We've gone through this discussion on this forum before, and this has not changed -- the party is not a group of interchangeable main characters.</blockquote>It makes party members' dialogue skills kinda useless and pointless, but what do I know? Anyway, discuss.
<br>

 

[Zomg]

What was the context of this? Interview? Response to some forum shit? Written in blood, at scene of own murder?

 

[Amasius]

I remember in the beginning of Hordes there was an encounter with some kobolds holed up in a bar that had something like six different methods of completion...

...

Ah, yes, it was in SoU. I always mix those up. And there were only two consequences, if I remember right -- either you saved the kobolds or you didn't.

How... Biowarian.

 

[Vault Dweller]

Response to some forum shit.

 

[Edward_R_Murrow]

I have to agree that a lot of "unique" options could be difficult, but three (or more) is a much better number than two (or one) for outcomes/consequence sets across the board in a game trying to be a "superb" CRPG. Two outcomes feels very limited. Even a third "kill everybody" option helps out.

 

[Saxon1974]

Im ok with pretty much everything he said. I realize its not possible to have every single type of solution for every quest. If he is sincere is saying that there will be as many ways to solve something as they feel is appropriate that is ok with me. That is, if his definition of appropriate is the same as mine.

I would be happy with something like this

Guys steals your money purse, you can

1) attack with weapon or cast spell on him
2) Use some type of speech to intimidate or talk him into giving it back
3) call guards
4) do nothing

That sound reasonable to me, I hope it doesnt just become either you attack or you dont. Is that too many options to be reasonable in a large scale game? If it is, I think I would prefer a smaller game world with more detail.

Not sure how I feel about him saying the main character control all the dialog, for some reason that gives me a "linear" feeling.

 

[Fez]

As long as I don't get three option which lead to the same response. I had enough of that in the KOTOR games.

 

[Mithter Thibbs]

What I'm interested in is the degree to which "game world mechanics" plays a part in all this "choice and consequence" stuff.

One of the things I've noticed about the Biowarian approach to options is that they're explicitly implemented and scripted, if they're in there at all. Near as I can tell, it's because of the story-requirement attached to everything they include in their games.

Let's take a specific scenario: You're tasked with getting a vase from Nobleman Y.

Now, in a Bioware game, your typical approach to this task would be through dialogue. You can persuade the nobleman to give you the vase, you can bluff the nobleman, or you can go on a quest for the nobleman to earn the vase. Depending on which Biowarian implemented the quest, the vase might not even exist as such until it is given to you by the nobleman, and even if it did, odds are you wouldn't be allowed to kill the nobleman and just take the fool thing anyway. The point is, all of the methods of acquiring the vase are explicitly accounted for by the designers, and as such, all of the responses have to be accounted for as well. It makes for a good overt story, where each and every action of the player's is acknowledged by the game, but contrast that with a game with more game-world mechanics, where there are systems and subsystems set up across the game world to deal with certain possibilities and actions that the player might engage in.

Vases exist whether they're given to the player or no. Vases can be stolen. Estates can be broken into during the dead of night. Guards can be called if a theft goes not unnoticed. Guards will attempt to apprehend thieves. Magistrates can be bribed.

In games like that, players are afforded a great deal more freedom to roleplay the character as they see fit, and the game responds; perhaps not in an entirely plausible fashion -- whoever heard of heard of a mass murderer dropping a few thousand coins at a temple all all being forgiven jsut like that? -- and perhaps not with the story-related aspects of the player's actions being put so prominently into the spotlight, but it allows the player freedom to act.

At present, I see it as a hallmark of the Biowarian approach: everything the player can do must be accounted for explicitly by the developers. NPC's can't be killed, areas can't be visited prematurely, objects don't exist until the story is ready for them to exist. In games like Ultima and Fallout and TES, players are left to find their own way and to write their own story... at the expense of the game acknowledging that story. BioGames have a story that trumps player freedom, player roleplaying, even player game-playing at times.... but damn, it usually results in a pretty good story. Just not mine.

Which approach is better? I prefer the wide-open style personally. I can identify more closely with the character if I can make him go where I want him to go and accomplish the more mundane tasks which inevitably fill out any CRPG the way I want to accomplish them. It feels too much like I'm being held by the hand when every little quest, no matter how trivial it is, is dictated to me. Too many basic things have to be selectively disallowed at arbitrary times in order to work in the "explicitly scripted" approach, and that just gets me asking "why"? And not the good kind of why, where I'm asking what the villain could possibly gain from doing such and such to me. No, these are the kind of whys where Im' asking why I've been allowed to run from every single enemy up to this point, but where I'm suddenly as lame as a cripple wearing concrete shoes.... all because the story dictates that I not be allowed to run from this particular enemy. If that's the only reason for my suddenly broken leg, that's a piss poor reason, because it's got nothing to do with the story that necessitated that lame-ass game play device in the first place.

Often, it's not even about the actual options in the game I'm allowed or not allowed to engage in: it's about the perception of having those freedoms. Even if Bioware doesn't choose to try and implement a world-wide system to handle reactions to basic events, the perception of those unalterable story-events can be improved upon. Quest items, minimize their use or disguise them as something else when possible. Instead of giving me with a quest-item password to the next dungeon as an unsellable, undroppable scroll, simply set a flag somewhere in the innards of the engine that says I know the password, and will speak it where appropriate. No inexplicably cursed paper required, no possibility that the player will "lose the key", because they don't interact with that object to begin with.

Anyway, I can certainly see Mr. Gaider's point. If you're going to try and explicitly script each and every option allowed to players for each and every event on an individual basis, you're going to start running into problems really, really quickly. I just don't think that explicitly coding each and every action is the way to necessarily go about game making. It seems to take a lot of the "game" part out of the game. Can't sneak into the guarded compound at night if that area never has a night cycle. Can't break into the house if the lock requires a "plot key". Can't steal the vase if the vase only comes into existence the moment the nobleman puts the vase into your backpack. Feedback vs. Freedom.

 

[elander_]

"No, we have no rule that says there must be X options for every quest. We include whatever seems appropriate.
...
I'm no more interested in forcing such options if none seem necessary than I am in forcing a good and evil "path" onto every quest just to have it. Some quests are very simple and they only have one manner of completion, others have more -- others have a lot more. If you want to write a game where there is a design rule of 3 options minimum no matter what -- be my guest. "

So it's a Baldur's Gate clone then. That would not be too bad if character dialogs and companion are interesting.

Looks like consequence are only important to determine the end of the quest. It's a shame but it's also a lot better than having just one ending per quest and not being able to dodge a millimeter from the path.

It was cool if they had soft consequences per factions as in Daggerfall. Every action the player does pisses off somes factions and makes other factions happy. It doesn't amount to actualy changing much in the game but when some factions like the player too much they may start sending him clues and help for some quests or lowering the prices. It may also help the player with raising his rank in guilds. When they don't like the player some may refuse to talk to him and in extreme cases may send foes to attack the player during quests. It's not something drastic or catastrophic but it let the player know which toes is he steeping on.


Some examples:

In a quest for the KoW, Gorthwog will steal the stuff if the player takes too much time to get it. He will then have to negotiate it with Gorthwog.

If the player is a vampire and during a certain quest to kill a person who is raiding a city one of the vampire clans will reveal the location of the foe and say his clan will be grateful to the player if he kills this rogue vampire.

During a quest where the player is incriminated by an influential merchant,if he is a member of the thieves guild, they will help the player providing him a clue about an eye witness to the events.

Mess up with some characters like Elysana by not doing her quests the way she requests and she will send assassins to kill the player.



If you look at the structure of Daggerfall and compare it to the structure of Fallout quests you will notice that these quests are full of exceptions, twists, dialogs to deal with certain rumors for during quest, quest success, quest failure, reactions to player race and class, etc

This stuff is important and in Daggerfall we can actually play a thief for example without ever touching a sword and do most of the quests. This is what i call freedom.

 

[Araanor]

BioGames have a story that trumps player freedom, player roleplaying, even player game-playing at times.... but damn, it usually results in a pretty good story. Just not mine.

But I don't even find it to be a pretty good story, which makes it all the more a problem. Half the game is Bioware railroading, and the other half is the pervasive and meaningless combat. This all adds up to a pretty bland experience for me.

 

[Mefi]

As long as I don't get three option which lead to the same response. I had enough of that in the KOTOR games.

Or the same three options every time - Bethesda (c).

Varying the amount of them isn't actually that bad of an idea but, as ever, it depends on the actual implementation and whether or not it just gives time-pressed programmers a quick and easy way to force the plot.

Totally disagree with him that you can have too many options; providing that they each lead to individual outcomes.

 

[Spazmo]

One thing you can't disagree with him on is that it's silly to approach every encounter with a mentality like "okay, I need to provide options for Combat Boy, Speech Boy, Science Boy and Stealth Boy here. How the hell do I shoehorn in all those different skills into this encounter?" Sometimes it just plain does not make sense to provide paths for everybody. If you only have a Combat Boy path (or if only the Combat Boy path is properly supported), then that's kind of lazy. But I think a designer should create interesting encounters first (in terms of character and drama and so on) and then fit game mechanics around them as best they can, not the other way around.

On the other hand, I have to question Gaider's mention of this kobold encounter being so hard to implement. Now, a small disclaimer, I haven't even played the game, let alone seen what was involved in designing/implementing that encounter, but it sounds to me like whatever the outcome was, it didn't have a huge outcome on the game and shouldn't have been so difficult to implement.

It seems to me like what's needed in RPG design is a kind of more rigorous formalism for encounters. The analogy I would make is to object oriented programming. In OOP, most everything in your program is an object. As a programmer, you don't really need to know all of what's inside a given object to work with or around it. You really just need to know what goes in and what comes out. What's in the middle is kind of a black box, and the programmer just needs to make sure that the in end and the out end fit with the rest of the program and it will work, because the stuff in the middle is modular.

Encounters, I think, could be designed with a similar mentality. You send in some stuff (a PC, some item, some relevant NPCs) and you get some stuff back out (set this global flag, change this reputation marker, increase XP, give out phat loot). Whatever is in the middle is modular and won't mess up the rest of the game.

It seems to me what's going on now in RPG design is like old C code. Everything is in one big mess of scrambled code, it's all horribly interdependent and linked up and it's unmaintainable. Approaching the design from a more OOP approach would almost lead to quest definitions with preconditions and postconditions, and as long as those are respected, what's in the middle could be dinked around with at will.

Or something stupid like that.

 

[galsiah]

Encounters, I think, could be designed with a similar mentality. You send in some stuff (a PC, some item, some relevant NPCs) and you get some stuff back out (set this global flag, change this reputation marker, increase XP, give out phat loot). Whatever is in the middle is modular and won't mess up the rest of the game.

That's one way to go. I'd much prefer to have many connections between/within encounters though - just formalized into a general system, rather than as ad hoc decisions for each quest combination.
Designing six paths for one largely meaningless Kobold encounter probably is overkill - even if you have black-box encapsulation. The better way to do things would be to set up more general rules/connections so that a fair number of simple solutions to such situations were naturally present. You'd still need to do things manually for significant/unique/high-level quests/combinations, but you wouldn't need to put in extra work every time you want a few different approaches to a simple, commonplace, low-level problem.

With that kind of setup, you'd only need to prescript the outcome criteria - the various strategies for getting to those situations would occur naturally. If all you're aiming for is a few ways to kill/save Kobolds, it's a reasonable approach.

EDIT: essentially I'm not saying much different from you - simply that the "dinking around" often needn't be necessary once things are set up generally.

 

[Callaxes]

I don't mind Bioware games, the only ones I've played were the BG series and KotOR 2, they're not the best, but atleast they gave some moments of satisfaction.

But NOW seeing this stupid son of bicth trying to define what RPGs are and underrating the importance of C&C is just fucking stupid and arogant.

 

[OSK]

I don't mind Bioware games, the only ones I've played were the BG series and KotOR 2, they're not the best, but atleast they gave some moments of satisfaction.

KotoR 2 was Obsidian, not Bioware.

 

[Fez]

KOTOR2 was riddled with the "pick one of three options, but they all lead to the same result anyway" situations too.

 

[Callaxes]

KotoR 2 was Obsidian, not Bioware.

Right. The reason why I think Avellone and Feargus need to suffer brain trauma before embarking on the Alien CRPG

 

[Volourn]

"NPC's can't be killed, areas can't be visited prematurely, objects don't exist until the story is ready for them to exist."

Not always. Maybe not even half the time. It does occur, though.

 

[Mithter Thibbs]

Yes, sometimes, but not all the time. Or sometimes not. Sometimes, it almost feels like these things are left up to the discretion of the developer that implements them, and that's something that I think should be worked upon: consistency.

It's not that I really mind the Quest Item/Quest Key thing so much -- it's annoying at times and kinda goofy, splattering the game mechanics all over the place like that, but it's still a game, and it's not really that bad. I just wish there was more consistency to it all. If it's consistent, I can learn to ignore it. If it's inconsistent, I'm going to notice it every single time.

 

[aboyd]

On the other hand, I have to question Gaider's mention of this kobold encounter being so hard to implement. Now, a small disclaimer, I haven't even played the game, let alone seen what was involved in designing/implementing that encounter, but it sounds to me like whatever the outcome was, it didn't have a huge outcome on the game and shouldn't have been so difficult to implement.

I was building a small game with the NWN engine, and my goal was to have a much shorter game (5 hours) but loaded with options, heavy emphasis on replayability. Many, many ways to solve each quest. 9 dialogue responses for everything (the responses are numbered so you can key-press a reply, so I didn't want to go over the max of 9).

I have to tell you, even a small simple quest can become HORRENDOUS if you have many ways to solve it. The problem is watching the state of the quest with all the interactions. So if you have 8 ways to solve it, you don't just have 8 dialogue options. You also have to provide dialogue for all the other NPCs involved, for each stage of each of the 8 ways to solve it. I could of course cut back on the NPC interactions -- perhaps only 2 people could speak with any knowledge of the quest, thus there were only 2 * 8 = 16 dialogues to write. However, each approach to solving the quest is not binary -- not done or undone. There are midway points that might need dialogue too. So now a more realistic count is like this:

(2npcs * 4solutions) + (2npcs * 3solutions * 2midwaypoints) + (2npcs * 1solution * 3midwaypoints) = 26 dialogues

If you have an NPC who happens to be involved in multiple quest lines, now you're juggling dialogue for all of those, trying to watch state and offer the correct response for whatever point in whatever quest(s) the player is on. If 3 or 4 NPCs are expected to speak about the quest, it can get downright miserable.

On paper, having lots of branches in dialogue or quests is very doable if you map it out. A nice grid can help you track the state of everything. The problem is that you start to bog down. Changes become a hellish black hole of time loss as you have to track down everything involved in the change you wish to make. My neat little game never got off the ground because doing this stuff became very unfun.

I'm not 100% sure this is a rebuttal to you point, though. I'm siding with Gaider on the notion that multiple solutions get messy... but only within the NWN engine. I've not tried others. It's very possible that other engines handle this stuff in far superior ways, perhaps using your OOP approach, or at least offering a base of general solutions that don't require tracking state or dialogue. Perhaps it's Gaider's fault that he's not using such an engine. But at least with the current tools, I feel his pain a little bit.