Monday, November 06, 2006
Designing A Computer RPG Rule System
Shamus Young has posted his thoughts (twice) on why the D20 (e.g. Dungeons and Dragons 3.0 / 3.5 ) system is bad for computer roleplaying games. You can read his very insightful articles here and here. And I actually do not disagree with him at all. I argue in favor of having the game rules being part of the player interaction in "The Rules of Roleplaying Games", but I also agree that the D20 system isn't optimal for computer games.
Whether or not it's optimal for tabletop gaming is subject a matter that is constantly debated.
Ultimately, there is no such thing as a perfect RPG system. At least, I don't think there is. RPG rule systems are abstractions that allow us to simulate... well, storytelling. The stories may be stupid, simple, and full of action and combat, but that's still the goal. The players play to get the thrill of seeing the story unfold with their characters, under their control, participating and influencing the development.
As with all abstractions, they have to be simplified and optimized to their purpose. "Pen & Paper" RPG systems are generally designed for human moderation and turn-based action. Naturally, taking that to computerized, real-time action system isn't exactly playing to the strengths of the medium.
For example: When running a Dice & Paper game, I expect combats to be few and to dominate game time. Roleplaying opportunities with non-player characters will take very little time at all. But when running a game through Neverwinter Nights, conversations took FOREVER due to the weaknesses of the text chat system. Even pre-created dialogs took a huge chunk out of development time. Combat, on the other hand, usually only took a few seconds, even for an epic boss-encounter. So I learned that an RP encounter could really "pad out" an encounter, but I couldn't plan to many.
Some things to consider when designing an RPG system - well, any game system, but I'm keeping it focused on RPGs for now:
1. Medium
Computers can crunch numbers and complex formulae with ease. Humans - not so much. Things that are such a big deal in a turn-based system (like who gets to go first) are almost unnoticeable in a real-time system. A primarily 3D, first-person view will have very different demands than an omniscient top-down view, particularly when it comes to controlling other members of the party.
Another thing that figures in here is the platform. While theoretically an interface can be considered an abstract layer that should be independent of mechanics, the reality is that whether or not the user is using a controller or a mouse is often an influence.
2. Audience
Who is your primary audience for this game? Not that they will be the only ones playing your game, but if you try to please everyone all of the time, you'll end up with something that nobody hates, but nobody really likes, either.
The nice thing about computer games is to some degree you can allow the player to tailor the system to their own preferences. But you can only go so far in this direction before you hit the law of diminishing returns, and it becomes more trouble than it is worth.
3. Genre / Style
So you've got Indiana Jones, who battles tons of machinegun-toting Nazis before breakfast while seeking artifacts of clearly supernatural nature. Then you've got Phillip Marlowe, tough-as-nails private eye from "The Big Sleep" (AWESOME movie, BTW). Marlowe has a healthy respect for guns, if not for the person holding them. ("Such a lot of guns around town and so few brains! You know, you're the second guy I've met today that seems to think a gat in the hand means the world by the tail.")
Both characters hail from a vaguely similar historical era, but vastly different worlds. Could the same game system accomodate both? Possibly, but not likely. The mechanics of the game dictate the flow, flavor, and style of the game. A swashbuckling-style game demands a game system that rewards risk-taking, while a gritty "realistic" genre would require a very different set of game mechanics. We've tried to run gritty, realistic games using the "Hero" (Champions) rule system for pen-and-paper, and it just doesn't work.
4. Scope And Scale
I think of scope and scale as breadth and depth to the game - though in game development, when you talk about scope, you are talking about everything that makes the game "bigger" in terms of development cost. But in an RPG context, I think of scale as how much content becomes available later in the game and how much the game changes from the early game. Scope is how much variety there is in game from the get-go.
This has an enormous impact on your game mechanics. For a higher-scale game, you need to keep unlocking aspects of the game as the player progresses. For a higher-scope game, you need a higher variety of player options (and content) from the get-go.
For example, Oblivion had a very large scope, but a much more limited scale. You could finish the entire storyline by level 2 or so, and the monster variety "topped out" at around level 20 or so. Most of the world was available to you after the initial dungeon, and the game even scaled the challenge everywhere based on your level.
The Final Fantasy games have a huge scale, but much more limited scope. The first half of the games are extremely linear, and there isn't much to do beyond combats and pursuing the storyline. But the character powers - and their opponents - change radically during the course of the game. By the end, your characters are ludicrously powerful compared to how they started, but there's still not much more they can do other than fighting and pursuing the main storyline (and a few optional side-quests).
Dungeons and Dragons 3.5 is unbelievably high-scope and medium-scale, even without the zillions of supplements provided by first- and third-parties.
Putting It All Together
The thing that complicates matters even further is that these four factors do not get along harmoniously with each other. Even a single factor, like audience, might be a little conflicted. For example, you may be creating a game for a more casual audience that cannot abide a complex, detailed system - but they find themselves bored by too much simplicity after several hours of gameplay.
That's why game design is more art than science.
(Vaguely) related musings...
* The Rules of Roleplaying Games
* You Can't Design Fun On Paper
* The Evolution of Computer RPGs
* Prototyping Means Sucking Less Sooner
Labels: Game Design, Roleplaying Games
- posted by The Rampant Coyote @ 6:39 AM (Permalink)
Comments:
Links to this post:
<< Home
I guess I'm having a tough time with this and your other post in really understanding what the argument really is, and what, if anything, is the solution to it.
I'm not as much of an ardent RPG fan as you are, but I do enjoy RPG systems in general, and I've always been a fan of the D&D system (still have a 1979 copy of the Player's Handbook and DM Guide on my bookshelf). SO I enjoy the discussion, but I'm not really sure where this is leading.
You started by discussing your impression of the rules of RPG games by stressing the importance of character attributes and randomness. In your second blog you don't disagree that the d20 system is bad for computer RPGs, but then you discuss other considerations like scope and audience.
I guess what I'm wondering is, what is really the source of the argument? What really are we referring to when we use the term "d20 system"? Is the problem with the use of stats, or is the problem more with the turn-based aspect of the system? Perhaps what I'm trying to say is that I'm really not absorbing what the true source of the problem is.
I'm not as much of an ardent RPG fan as you are, but I do enjoy RPG systems in general, and I've always been a fan of the D&D system (still have a 1979 copy of the Player's Handbook and DM Guide on my bookshelf). SO I enjoy the discussion, but I'm not really sure where this is leading.
You started by discussing your impression of the rules of RPG games by stressing the importance of character attributes and randomness. In your second blog you don't disagree that the d20 system is bad for computer RPGs, but then you discuss other considerations like scope and audience.
I guess what I'm wondering is, what is really the source of the argument? What really are we referring to when we use the term "d20 system"? Is the problem with the use of stats, or is the problem more with the turn-based aspect of the system? Perhaps what I'm trying to say is that I'm really not absorbing what the true source of the problem is.
# posted by 
: November 07, 2006 9:39 AM
: November 07, 2006 9:39 AM
Fair enough. I should probably edit them for clarity - I think the point probably got lost in multiple edits.
My concern is the contention that seems to be rising amongst critics and designers that RPGs should be about player skills and abilities, not their character's. That somehow the reliance upon your character's (avatar's) "statistics" is old-fashioned and a legacy of pen-and-paper gaming that shouldn't be necessary in a world where a computer can simulate a somewhat realistic world.
My contention is if you go there, you lose what makes an RPG an RPG, and you get... well, Doom.
I think games like Oblivion straddle the gap as "action-RPGs," but I don't want to see all RPGs go that direction (and beyond).
To me, a big part of the challenge and fun of RPGs is building and growing your character, and using strategy and clever preparation to work within their limitations to defeat the odds. Maybe it's the semi-closet wargamer in me.
But I don't always want to play a game where I'm personally dodging the arrows and spells of enemies. I like the "thinking man's game" too much, and I find it every bit as compelling and immersive as the first / third person perspective actiony games where the stats and game rules are hidden from the player.
While it's somewhat related, this later article was a whole 'nother subject. More about what kind of factors you need to figure in when designing the rules of your game (specifically RPGs, though really any game).
Shamus was arguing that the one-size-fits-all mentality for a gaming system isn't appropriate (and I agree), so I wanted to suggest factors that need to be considered for creating (or choosing) a game system.
As to the use of the D20 system, it's the core ruleset of the Dungeons & Dragons 3.5 system. The 3-18 scale for attributes (not that it's really 3-18 anymore), Armor Class, Hit Points, Level-Based, Classes, Feats, Skill Points, Prestige Classes, use a D20 to determine success or failure, etc. I mean, even the name of the system is in reference to the core mechanic of rolling a 20-sided die to determine success or failure. A computer isn't limited to polyhedral dice for randomization, so does it make sense to incorporate this system lock, stock, and barrel?
Well, arguably yes, if it's supposed to be a licensed D&D game. :) But otherwise, probably not.
Shamus's articles explain more of what the problem of using the D20 system in computer games might be.
Maybe I ought to create a more practicle example article? I probably have been delving a little too long in the theoretical.
My concern is the contention that seems to be rising amongst critics and designers that RPGs should be about player skills and abilities, not their character's. That somehow the reliance upon your character's (avatar's) "statistics" is old-fashioned and a legacy of pen-and-paper gaming that shouldn't be necessary in a world where a computer can simulate a somewhat realistic world.
My contention is if you go there, you lose what makes an RPG an RPG, and you get... well, Doom.
I think games like Oblivion straddle the gap as "action-RPGs," but I don't want to see all RPGs go that direction (and beyond).
To me, a big part of the challenge and fun of RPGs is building and growing your character, and using strategy and clever preparation to work within their limitations to defeat the odds. Maybe it's the semi-closet wargamer in me.
But I don't always want to play a game where I'm personally dodging the arrows and spells of enemies. I like the "thinking man's game" too much, and I find it every bit as compelling and immersive as the first / third person perspective actiony games where the stats and game rules are hidden from the player.
While it's somewhat related, this later article was a whole 'nother subject. More about what kind of factors you need to figure in when designing the rules of your game (specifically RPGs, though really any game).
Shamus was arguing that the one-size-fits-all mentality for a gaming system isn't appropriate (and I agree), so I wanted to suggest factors that need to be considered for creating (or choosing) a game system.
As to the use of the D20 system, it's the core ruleset of the Dungeons & Dragons 3.5 system. The 3-18 scale for attributes (not that it's really 3-18 anymore), Armor Class, Hit Points, Level-Based, Classes, Feats, Skill Points, Prestige Classes, use a D20 to determine success or failure, etc. I mean, even the name of the system is in reference to the core mechanic of rolling a 20-sided die to determine success or failure. A computer isn't limited to polyhedral dice for randomization, so does it make sense to incorporate this system lock, stock, and barrel?
Well, arguably yes, if it's supposed to be a licensed D&D game. :) But otherwise, probably not.
Shamus's articles explain more of what the problem of using the D20 system in computer games might be.
Maybe I ought to create a more practicle example article? I probably have been delving a little too long in the theoretical.
I'm seeing a few different arguments here:
Quote:
"My concern is the contention that...RPGs should be about player skills and abilities, not their character's. That somehow the reliance upon your character's "statistics" is old-fashioned and a legacy...that shouldn't be necessary."
I think this is basically true. My understanding of RPGs is that they are predominantly about the creation and development of individual characters, and how the features of those characters grow and affect gameplay and outcomes. I'm not quite sure that I got this argument from Shamus's blog, although I don't know if that's what you were implying. My impression is that he agrees with you on this.
Quote:
"Shamus was arguing that the one-size-fits-all mentality for a gaming system isn't appropriate, so I wanted to suggest factors that need to be considered for creating (or choosing) a game system."
I see your point, for sure, and I think your arguments are on target. I'm still not quite sure what Shamus was arguing, however -- at least in his first blog. Looking over his blog again, I think he makes these key points:
- The focus of the d20 system is the die roll and the anticipation surrounding it.
- That same situation doesn't exist on computers.
- The d20 system allows for lots of randomness due to die rolls.
- Randomness doesn't seem as nice on computers since you don't see the dice.
- The d20 system helps reduce complexity for tabletop games.
- Computers don't need reduced complexity.
- Games often assume you understand the d20 mechanics.
- d20 mechanics are confusing and we should use something easier.
- Morrowwind's system is nice because it uses skills that increase as you play and which affect the outcomes of encounters.
My head is left kind of spinning after that spew of logic. I guess this is what I was getting at when I asked, "What really are we referring to when we use the term 'd20 system'?" It seems to me that Shamus doesn't really object to a game system that utilizes specific player attributes and skills to determine how the player advances and how he or she performs in combat. The Morrowwind system does that, it just uses different attrributes and skills than the traditional D&D system. It seems to me that he doesn't object to a system that applies skills and attributes at the level of individual rounds of combat; that's what the Morrowwind system does, to the best of my knowledge. I'm sure he doesn't object to the inclusion of randomness into the game's battle model, similar to how dice are used in tabletop games on individual strikes. So what is it he's objecting to? He's objecting to the d20 system, but what exactly is he referring to by this?
It sounds to me like he's really not objecting to the concepts behind the system, but to the typical implementation of this system. It sounds like he still wants a system that incorporates individual character attributes, character growth and development, and a battle system that allows attributes, skills, objects, and randomness to influence the outcome. This sounds to me like the d20 system, just maybe a different external implementation of this system.
Quote:
"My concern is the contention that...RPGs should be about player skills and abilities, not their character's. That somehow the reliance upon your character's "statistics" is old-fashioned and a legacy...that shouldn't be necessary."
I think this is basically true. My understanding of RPGs is that they are predominantly about the creation and development of individual characters, and how the features of those characters grow and affect gameplay and outcomes. I'm not quite sure that I got this argument from Shamus's blog, although I don't know if that's what you were implying. My impression is that he agrees with you on this.
Quote:
"Shamus was arguing that the one-size-fits-all mentality for a gaming system isn't appropriate, so I wanted to suggest factors that need to be considered for creating (or choosing) a game system."
I see your point, for sure, and I think your arguments are on target. I'm still not quite sure what Shamus was arguing, however -- at least in his first blog. Looking over his blog again, I think he makes these key points:
- The focus of the d20 system is the die roll and the anticipation surrounding it.
- That same situation doesn't exist on computers.
- The d20 system allows for lots of randomness due to die rolls.
- Randomness doesn't seem as nice on computers since you don't see the dice.
- The d20 system helps reduce complexity for tabletop games.
- Computers don't need reduced complexity.
- Games often assume you understand the d20 mechanics.
- d20 mechanics are confusing and we should use something easier.
- Morrowwind's system is nice because it uses skills that increase as you play and which affect the outcomes of encounters.
My head is left kind of spinning after that spew of logic. I guess this is what I was getting at when I asked, "What really are we referring to when we use the term 'd20 system'?" It seems to me that Shamus doesn't really object to a game system that utilizes specific player attributes and skills to determine how the player advances and how he or she performs in combat. The Morrowwind system does that, it just uses different attrributes and skills than the traditional D&D system. It seems to me that he doesn't object to a system that applies skills and attributes at the level of individual rounds of combat; that's what the Morrowwind system does, to the best of my knowledge. I'm sure he doesn't object to the inclusion of randomness into the game's battle model, similar to how dice are used in tabletop games on individual strikes. So what is it he's objecting to? He's objecting to the d20 system, but what exactly is he referring to by this?
It sounds to me like he's really not objecting to the concepts behind the system, but to the typical implementation of this system. It sounds like he still wants a system that incorporates individual character attributes, character growth and development, and a battle system that allows attributes, skills, objects, and randomness to influence the outcome. This sounds to me like the d20 system, just maybe a different external implementation of this system.
# posted by : November 08, 2006 1:46 PM
: November 08, 2006 1:46 PM
By my understanding, he's objecting to the specifics of the D20 system. Not the concept of quantizing attributes and skills and putting them into a computer game - not at all.
But the D20 system IS very specific mechanics, optimized for human-moderated play. For example, it dictates that a wizard of level X gets N spells of level Y per day, plus bonus spells derived from his Intelligence score.
If you are interested, the D20 game system can be found for free, here:
The D20 (Revised) System Reference Document
Even Dungeons and Dragons Online, the Massively Multiplayer RPG based on the license, had to take some liberties with the system to make it work. So they aren't really playing D20, but a variation thereof (and one of their biggest changes was, unsurprisingly, the spell system).
I think Shamus and I actually see fairly eye-to-eye on most elements of this. I take more issue with some of the comments I cited in my earlier article, where they seem to be objecting to the very concept of, say, quantizing attributes into some measurable value - regardless of system.
But the D20 system IS very specific mechanics, optimized for human-moderated play. For example, it dictates that a wizard of level X gets N spells of level Y per day, plus bonus spells derived from his Intelligence score.
If you are interested, the D20 game system can be found for free, here:
The D20 (Revised) System Reference Document
Even Dungeons and Dragons Online, the Massively Multiplayer RPG based on the license, had to take some liberties with the system to make it work. So they aren't really playing D20, but a variation thereof (and one of their biggest changes was, unsurprisingly, the spell system).
I think Shamus and I actually see fairly eye-to-eye on most elements of this. I take more issue with some of the comments I cited in my earlier article, where they seem to be objecting to the very concept of, say, quantizing attributes into some measurable value - regardless of system.
I see...I think that's what I meant when I said he objected to the implementation of the system. But there are plenty of sophisticated RPG games that don't employ the d20 system...if you don't like the d20 system, shouldn't you just not play games that advertise that they specifically employ the d20 system?
I think what it comes down to, more than anything, is preference rather than one overarching view of what RPG games should be. I think players generally appreciate complex systems that try to model all of the various details of "reality", so to speak. I think some people just prefer to push that complexity under the surface, where it doesn't interfere with game play, while others enjoy having all the bells and whistles out in the open.
And then, of course, is another camp that would prefer to leave the complexity behind and simplify things to streamline gameplay, but my guess is that these folks are more likely to play an FPS game than an RPG game.
Post a Comment
I think what it comes down to, more than anything, is preference rather than one overarching view of what RPG games should be. I think players generally appreciate complex systems that try to model all of the various details of "reality", so to speak. I think some people just prefer to push that complexity under the surface, where it doesn't interfere with game play, while others enjoy having all the bells and whistles out in the open.
And then, of course, is another camp that would prefer to leave the complexity behind and simplify things to streamline gameplay, but my guess is that these folks are more likely to play an FPS game than an RPG game.
# posted by : November 09, 2006 9:13 AM
: November 09, 2006 9:13 AMLinks to this post:
<< Home
