Fixing the genre from the ground-up…
- May 27, 2010 11:49 AM
In my last blog, I talked about how the framework commonly used for MMOs is basically broken. While it lasted a good many years through several iconic games, players are now growing out of the novelty of the current order of gameplay and are looking for something new (or something different- which is new, considering that 80% of the current market is the same).
When development on Epic Frontiers started, Chris, Rick, and myself sat down and had several design talks that focused not on features that we wanted to have, but situations that we wanted to represent in the game. We figured that if we could choose the kinds of scenarios we wanted to present to the player and “solve” them in terms of game mechanics, then a system would come together that would be custom fit for what we wanted. As an added feature of whatever system we came up with, it had to be “unified”- that is, there would be no forcing of the player to grind on combat to level up his crafting character. The game had to be gameplay agnostic.
If you’ve followed my series of DigitalFlux Entertainment blogs (which are separate now, btw), you would have seen quite a bit revolving around interacting with NPCs. We’re old-school role-playing gamers, much like most of the guys making games, and what we wanted to represent aside from the combat scenarios was the role-playing aspect.
So far, role-playing is relegated to certain servers where people talk ye olde English and dwarves all sound like they’re from Scotland (I’ve heard Scottish people complain about this). That’s all well and good- players want to get into character, who am I to stop them? Except for one thing: The game will never recognize nor reward the player for that effort. The player will develop a reputation among their peers for exceptional role-playing skills, which is great- but wouldn’t it be greater if the game was able to use that?
So in brainstorming, we thought of several features that could get us from here to there:
- No Classes, No “XP”: From the start, we felt that the way that classes and experience levels were handled in current MMOs was being done incorrectly. The fact is, before the new player even starts playing a game, they set their character’s destiny in stone by choosing their class first. Instead of a character growing into the paladin role by playing a lawful-good character and making the hard choices needed to go down the long path to becoming a paladin, they start out as a paladin, and never have the opportunity to stray- or even be tempted. Not to mention the fact that should the player change their mind, they need to roll a new character up and start from scratch. What fixes that is a system where the player starts as a blank slate and picks Ability Skills instead of classes. And as those skills are used, they automatically increase in level. A player can start out wanting to be a paladin, but end up playing as a rogue or fighter, and pick up those skills as they go, slowly migrating from one type of character to another without losing anything, since all skills picked up would remain available to use. Because more than anything role-playing is about the journey and experiences of the character, not an arbitrary class choice.
- The Action System: The next thing we had to set down on paper was that the entire game would be based on an Object + Action = Result system. What we settled on was that the player would be the “container” for the actions rather than a weapon (though a weapon can have actions which only work with it). What does that mean? Usually, as your character levels up in a game, you need to constantly discard weapons because they don’t work as well anymore. In reality, if your character is a swordmaster, it shouldn’t matter if you pick up a rusty level 1 sword- you should still be able to hit (albeit with less damage) better than a lower-level swordsman. So we based the attack success of the sword on the skill level of the action the character is attempting. A level 10 sword attack would do 1D10 +/- the sword’s modifiers. So a level 50 sword master with a -2 sword (rusty? bah!) gets a 1D50-2 against, say, a level 20 swordsman with a +10 sword who gets 1D20+10. The max the level 20 player can roll is 30, while the level 50 player rolls up to 48. That’s how you know who the better swordsman is.
- Talking to NPCs: The biggest problem for MMOs to solve these days is being able to interact with NPCs in a meaningful manner. Some AAA MMOs are attempting this in different ways, but we thought we’d tackle the problem head-on and develop a conversation system that offered the player the ability to quickly get information from an NPC in a way that was intuitive without forcing the player to type out whole sentences. The Interrogative system, developed for Epic Frontiers, gave us all of that and more. The NPCs have a certain depth of knowledge in certain subjects of conversation, and that information can be accessed through use of menu items asking generalized questions such as: Who, What, When, How, Where, Elaborate, and so on. Additionally, the NPC can ask questions of the player, and also can have dialog that reflects their AI states and other information. A lightweight memory system allows the NPCs to vaguely remember what you talked about with them and develop familiarity as a result. But that’s not enough. What we are currently working on is extending the system in two ways. The first involves assigning missions through conversation instead of using the “Wall of Text” that has become so infamous in games, and the second involves giving NPCs the ability to talk about their AI states, mission status, goals, and even their views of the player.
- More Personality: Of course, the conversation system we worked on raised serious issues with AI. For role-playing to work, we needed a lightweight system that could represent a variety of personalities. After testing close to 15 systems, I settled on a set of personality “scales” which go from one extreme to another and which almost any (NPC) personality can be described. Personality traits such as Altruism, Openness, Discipline, Dominance, Sensitivity, and Trust are mixed with emotional scales such as Anger, Fear, and Disgust to create the dynamic NPCs that we need for such lofty goals as true role-playing systems. We’re also looking at ways to extend and work with this system with things such as memories, so that NPCs can “model” the player’s behavior towards them in the same way as they themselves are modeled, so they form opinions of players on an individual basis (in other words- if you piss them off, they’ll remember it).
- Content: So far, we’ve been doing a pretty good job of tackling technology for the framework of the game, and then when we arrived at the point of how to get more content into the game, it was time to press all the way forward and leverage “Dynamic Content”. It’s sometimes derided, but when done correctly, Dynamic Content can be very powerful. Some of the content in Epic Frontiers comes in the form of generating NPCs, complete with some back story, a set of relevant skills, AI, inventory, and even a name (the Charkritian name dictionary currently yields up to 24 million name combinations). Other forms of content come in the form of randomly generated dungeons that can be accessed in the countryside, within cities, or in other places. We also founded Bounty Agencies within the factions which farm out missions to players for a variety of purposes, most generated- some only partially. And that content generation is layered on top of the usual spawn points you’ll find for creatures and enemies in the game world, giving you a lot to do.
- Crafting: And last but not least, we decided to pay some attention to crafting, and give crafters- who are now on equal footing with fighters in advancement in the skill-based system- a deeper system to work with. Recipe-based crafting is the default mode for many players, and from weapons and gear to vehicles and clothing, the player has access to an array of skills and materials. But we went two steps further still: First by creating a special “Generic Recipe” system which allows crafters to experiment by substituting parts of an object. A generic car recipe may call for four tires to be added to the vehicle’s body, but the player can choose which kinds of tires to add, and the recipe will automatically adjust the stats of the resulting vehicle to reflect the non-standard change. The second step we took was to introduce a system of Material Properties: During the mining/smelting process, the player has the ability to combine rare materials in alloys which hold special properties that can be important to certain kinds of players. Lightweight materials, or materials with special magical or technological properties can be created from using these special alloys in regular or generic recipes.
In the coming weeks, we’ll be publishing more in-depth articles on these sections of the game, including screenshots and video demos of the features as they currently are. Stay tuned!
- Tags: