Monday, March 21, 2011

A Proposal

Since my first post this year, I have been writing a lot about Character Power Progression (CPP), Holy Trinities, Skill Systems, Class Systems and I'd like to come to a preliminary conclusion; a Proposal.

Unfortunately, features in MMORPGs can never be analyzed without context. That's why I will first describe the game the proposal is meant for. I try to keep it short.
  • Little teleport
    • No character teleport
    • No item teleport
    • No currency teleport
    • But limited information teleport
  • No instances
  • No predetermined factions
  • Character housing
  • Group housing (read castles)
  • Control zones (borders)
  • Sieges that take weeks
  • Crafting
  • Item decay
  • 99% player-run economy built on buy and sell orders
  • Trade
  • Cultivation of wild lands
  • No Holy Trinity, but GW2-like PVE
  • No boss monster encounters that need to be trained or can be trained
  • Collision control instead of threat
  • Full PvP, but strong guards where reasonable. Some areas are very secure
  • All player vs player combat takes at least 2 minutes
  • Items are necessary to stand a chance in combat, but the difference between good and phenomenal items is not that big
  • Full Looting
  • Credible weight limitations on inventory
  • Bounty system
  • It requires roughly 4 new players in basic equipment who know what they are doing to beat the best-equipped/skilled player on the realm
  • Very diverse PvE content. Some areas in the world inhabit very powerful mobs, some inhabit really weak mobs. Mobs mostly look according to their power
  • Design focus on creation, but destruction by players/mobs is a possibility
  • Design focus on group content. Very strong incentives for groups to recruit and contact new players
  • Possible goals
    • Defending together against monsters
    • Defending together against other players
    • Attacking monsters to gain items/gold
    • Attacking other players to gain fame and influence
    • Creating a skilled character
    • Becoming rich by trading
    • Creating an army of players, conquering the world
    • Being a shop keeper
    • Offering transportation services
    • Playing bandit
    • Bounty hunting
  • You could also say: 70% Eve Online on the ground.

What is missing, obviously, is the CPP. And that is what the Proposal is about. Adding to last post, I suggest this system:
  • A skill system (OMG!)
  • Every skill has associated actions. Exercising these actions improve the skill. Some more, some less
  • Every action has associated skills. Exercising the action improves those skills. Some more, some less
  • Skills only apply if the correct equipment is used. The equipment you wear defines your role
  • Efficiency at increasing a skill is influenced by other skills you have
    • Some skills cannot be trained unless other skills have been trained enough (skill trees)
    • Some skills make learning another skill faster
  • After some amount of skill gain (equivalent of about 2 hours) your character becomes tired and starts to gain skills ever more slowly until rested (logged off) for at least 8 hours
  • To become 80% effective at a role requires about 50 hours /played if you focus on it. To become perfect requires 1000 hours of focused play
  • Gaining some points in some groups of skills reduces other skills in some other groups

Example:
Defeating somebody with a sword (action) makes you better at 'melee combat' and also better at 'combat with swords'. Defeating somebody with a mace also increases 'melee combat'. But it doesn't make you better at 'combat with swords'. Thus, fighting any melee combat, you become better at general melee combat, but to become a master swordsman you need to be more than just good at 'melee combat'.
If you become good enough at skills you gain new 'moves'. A move is something like a 'whirlwind attack' or a 'fireball'.

The higher your swords skill, the harder it is to become better at melee combat by fighting with swords. Therefore, you have an incentive to also train maces and daggers a bit.

Obviously, all pure swords-related skills are useless if you don't use a sword. Some skills require a sword and a specific kind of armor to take effect. Some skills require a very specific equipment (assassin skills). Some skills require a very expensive equipment (remember item decay).

Starting to use magic draws on your physical capabilities and makes you a worse warrior. And vice versa.

What do you think?
One thing: Before you tell me this system is incredibly hard to balance: Since players are locked in their equipment that allows only a small subset of usable skills, this needs not be true.

Now, please don't tell me you like it. I know you do. Please tell why this might be a terrible mistake. Critical minds welcome. I need criticism! But please concentrate on the proposal, not on the game I want to embed it in. That's just necessary background information.

24 comments:

  1. That a great bullet point list, obviously you been thinking about it quite a while.

    Now the slight problem with exclusively bullet point approach is that imho coherent design can not be described just a collection of bullet points (at least not a flat one) . Especially for MMO -because most of the game design itself is tightly tied to the technical design of server/client

    Another point I would like to mention is that since it is supposed to be a game it has to be designed to be fun. Fun within context of your target demographics of course, but still fun nevertheless.

    There are many games which had many of the bullet points you described, yet I would say most of those resulted in a bad game. They detracted from experience, not added to it . Not because the feature is just horrible , but because it was not designed from the fun perspective, or not really designed at all: (lets put full lot in ! -why? how does it add to player experience? how does it affect new player experience? how does it impact new player vs newbie player? is it fun only for old players? Do only veteran players have advantage?) .

    Take every point and ask yourself "was it implemented elsewhere" (answer is yes for almost all of them). Was it fun? (answer is no). Why the design failed exactly ? How to avoid it?


    #Little teleport

    - How do you ensure travel does not become tedious?

    #Sieges that take weeks
    How do you ensure they are actually fun to play? 6h+ sieges are anything but fun.How do you imagine players keep focus on something for WEEKS? How about 3am raids?

    #Collision control instead of threat
    Do you have approach to solve technical problem of collision detection in MMO setting? What makes you think it will work?

    #Full Looting

    How do you ensure it will not turn off new players? bad experience
    for new players = dead game

    #Very diverse PvE content.
    How do you do that? Are you aware the content creation cost ,especially modeling and animation are very high?

    #Design focus on creation, but destruction by players/mobs is a possibility

    Hmm how exactly it is a possibility? Imagine you are a player who spent months on building beautiful elaborately detailed castle. Do you think anyone would come to terms with its destruction in game? Keep in mind there will be groups of players hell bent on destroying it

    # Cultivation of wild lands
    #Control zones (borders)
    #Full PvP, but strong guards where reasonable. Some areas are very secure

    This needs to be fleshed out a lot more. Why would player cultivate lands when they risk em? What prevents wolves making wolves guilds and preying on sheep till it is extinct (and thats the main dynamic which happens in every open world pvp game and it will kill game really fast)

    Now I didnt go into CPP part of the game, because while it is a very important part of the player experience imho it is a relatively thin layer on top of everything else.

    The character template balance thing imho mainly revolves around deciding

    1) how much you want mask dps vs HP scheme (which at the end all those sytems are).Do you want roles split (which allows greater diversity, but kills small scale balance) or you prefer true balance (100 hp and 10 dmg abilities is true one for example)

    2) How balanced you want it to be on 1vs1 ,2vs2 ,5vs5 scale?

    3) do you really care for
    it to be "balanced"? - not every combo has to be viable ,you just have to make easy for people to recognize mistake with their choices and easily revert them ( not involving stuff like re-rolling new toon)

    ReplyDelete
  2. Thanks for this really long comment, Max. Blogspots brilliant spam filter sorted it out right away ;)

    I should have known that it is impossible to resist the urge to comment the background instead of the topic. mmh...

    As you know I consider the task of the MMORPG designer not to not implement desired features that aren't fun, but to implement them in a way that they are fun. I do have detailed ideas on how to answer almost every issue you brought forward, but that is not a thing to discuss on this a blog at this time.

    I'm not currently being paid to design a MMORPG, so I really just wanted you to have this as a rough background information
    It is important, because the kind of CPP I described wouldn't work with WoW, for example.

    On your questions of balance. I think I remember you favouring a true 1:1 balance in games. This kind of MMORPG couldn't deliver that, of course. The game would not only be about combat. Combat would be some 50% of it. It is designed as a group game, so group vs group has to belanced. The individual new player would be 'wooed' by existing groups to join them, because in a game without instances and a small power intervall (4x), every player counts.

    Since you can change your role anytime, and new roles can be added anytime, balance problems can be dealt with quite fast and are not as serious as in other games.

    Since CPP allows you to reach 80% of power of a role within just 50 hours /played (about 1/10 of classic WoWs time_to_maxlevel), the individual role doesn't matter as much.

    Since all reasonable roles have a power difference of no more than factor 2x, it is easier to prevent roles that are some 10x as effective as others.

    Since we have full loot and, I didn't write that, a rather serious death penalthy (compared to modern games), groups wouldn't attack other groups with the same number of people.

    Those who enter a battle and don't outnumber the enemey are stupid. This game is about strategy and control. It puts you into the position of a foot soldier. And it still aspires to be fun to play - even solo, but always in a group context.

    Sure, that's a hell nof a lot to promise and actually it's no promise, because nobody pays for this game right now ;)

    But within that framework, what do you think about the CPP ?

    ReplyDelete
  3. Nils,

    I'm going to leave the CPP for now and comment on the other ideas. The thing that stands out the most for me is the no predetermined factions. I left AoC because of this, simply because players could attack and kill me at any time, for no reason, and with no consequences for themselves. This will run into a shopkeeper losing his shop, or a farmer his farm, or a house or castle.

    There needs to be some sort of system in place that prescribes consequences for these actions. EVE of course has such a system in place, but the only thing you can really lose in EVE is your ship. In your world, which I have to say will really encourage the builders and creaters amongst us, there will be a lot more to lose.

    The key point will be how to stop it all degenerating into 90% of the populations being made up of random armed thugs hell bent on destroying and pillaging everything that they can see.

    ReplyDelete
  4. Thenoisyrogue, you are just as terrible as Max. ;)
    Those ideas are really just that: Stupid ideas that 50% of MMORPG players come up with 10% of the time.

    Of course, we could delve into how castle sieges could be made fun and all the rest. I am sure you have quite some ideas, so have I. None of them will be tested and I really don't want to talk about it in this blog post. I'd love to make an own post for all the 'features'. The problem is that you mostly cannot discuss these them while ignoreing the rest - nor can you discuss them while also discussing all the rest at the same time.

    So please (please !:) if you have an opinion on the CPP, tell me.

    ReplyDelete
  5. Hrm... I don't like full looting, and I love teleporting (at least hearthstone tech), but other than that, I like what I see here. The narrow power band caught my eye. I'd love that, as well as the longer PvP. I hate ganking; I think it's bad design and bad for the community.

    ReplyDelete
  6. "What prevents wolves making wolves guilds and preying on sheep till it is extinct"

    Item decay, plus the high cost of attacking vs defending, plus high cost of carting off your plunder (no teleport, remember), and so on.

    Wolves will need weapons to attack. Their only source of weapons is by raiding the sheep. Who have weapons. And walls. And paid militia. And technology. And science (or magic).

    Settled agrarian societies are more efficient and prosperous than nomadic hunter/gatherers, who are in turn more efficient and prosperous than raiding bandits (who have to spend a lot of energy & time simply evading the pursuit of justice).

    The wolves could well establish their own little colony of crafters, mining up iron ore and smelting that, paying a bunch of snooty pointy eared geeks to sit about and ponder the mysteries of the universe (and thus eventually discover Fireball rank 1) ... at which point how are they different from the colony of sheep over the hill?

    ReplyDelete
  7. I think the CPP proposal looks good. You might want to think about a couple of issues common to skill-based systems:

    (1) What exactly will players be doing to increase their skills? Use-based systems often encourage grindy behavior like casting the same spell over and over, hitting practice dummies for hours on end, etc. You want to make sure you're encouraging players to do fun things.

    (2) How will you prevent players from getting every single skill (in which case everyone will be more or less the same)? You say that "Gaining some points in some groups of skills reduces other skills in some other groups." Is that intended to prevent players from becoming masters in every skill? How exactly will it work?

    I know you're not looking for comments on the embedded game, but I have to say it looks like exactly the kind of game I'd like to play! :D I hope you'll expand on some of those ideas in future posts.

    ReplyDelete
  8. I wasn't sure what information teleporting meant, is that just global chat and tells?

    On the CPP, are you wanting to make make new characters competitive as quickly as possible? I enjoy playing guild wars from time to time, and with their emphasis on players skill making new characters useful more than leveling as a restriction, I have a fully capable character who only has about 20 hours played. This is probably too fast for new players, but If you've already played the game, it lets you make a new character that's useful quite quickly. Of course, they also give you the option of just starting a pvp character at top level. So, given the comparison, would that be too fast, and unrealistic?

    Also in a connected question, how important to your character progression is player skill? Would you try to group people by skill at all? Allow more skilled players to advance more quickly? Or do you intend to deliberately mix players of different skill levels? It's my impression that this is at least part of what FPS's do with kill score advancement. Which could be completely wrong, but that's what it looks like to me.

    ReplyDelete
  9. I think I've mentioned Darkfall before, but this does sound like you're describing that game's skill system almost to a tee. Not to mention most of the bullet points are in Darkfall too.

    ReplyDelete
  10. "Some skills make learning another skill faster"
    Please no. This pushes players into taking these skills as soon as they can, reducing choice. For the player they do it so they can get to a higher power level faster (over the long term), but in the short term it's pretty bad. Didn't EVE semi-recently remove these sorts of skills because they were doing little more than creating a huge time sink before players could learn 'real' skills?

    ReplyDelete
  11. @tolthir re [grinding skills] - well, Nils did mention there would be diminishing returns on learning without rest. This could also be expressed/framed as a daily allowance of clear mindedness (i.e. a state of mind receptive to learning), in the same way "rested XP" is a re-framing of penalized grinding.

    "How will you prevent players from getting every single skill?" Another way to Nils (skill decay) is to lock the advanced moves of skill trees behind content, time, and distance barriers.

    ReplyDelete
  12. Nils did mention there would be diminishing returns on learning without rest.

    I think diminshing returns are a great idea, but I was more concerned about the grindiness of the underlying activity. For example, if the most efficient way to level the fire magic skill is to shoot fireballs into the air over and over, it doesn't matter whether you're limited to doing it for two hours a day; it's still very grindy. So you'd need to make sure that the optimal way to advance skills is by doing something interesting.

    I think what you'd need to do is make sure that skill points are awarded only for accomplishing something useful with a skill (like killing a monster, completing a quest, etc.) rather than simply for using the skill. The interesting question is how to determine what actions should award skill points, and also how to allocate the points if the player earns them using more than one skill.

    ReplyDelete
  13. Thanks for commenting on topic, Tolthir. I'll try and answer your points.


    (1) What exactly will players be doing to increase their skills? Use-based systems often encourage grindy behavior like casting the same spell over and over, hitting practice dummies for hours on end, etc. You want to make sure you're encouraging players to do fun things.

    Gameplay dominates simulation here. As you already said, you need to get rewarded for doing fun (and if possible immersive) things. You don't get rewarded for simply swinging your sword at something, but for actually defeating opponents or achieving something using a sword.

    ---
    (2) How will you prevent players from getting every single skill (in which case everyone will be more or less the same)? You say that "Gaining some points in some groups of skills reduces other skills in some other groups." Is that intended to prevent players from becoming masters in every skill? How exactly will it work?

    Three points:
    1) As you already noticed, gaining skills in some skill groups, lowers skills in other skill groups. Magic/Melee combat would be an example.

    2) You gain 80% effectiveness within about 50 hours /played. But to become 100% effective (and gain all the fame) you need to spend about 1000 hours /played. This should encourage many playrs not to try to become a master of all trades - even though that might be more effective overall.

    3) Even if long-term players are very effective at the same roles, they will always be locked into one of their roles, as determined by their used equipment. Therefore, they will hardly be similar in a specific situation - even though their characters might be able very similar.

    Generally, this is a slightly modified Eve Online approach, so it's not grey theory, but already tested.


    ---
    I know you're not looking for comments on the embedded game, but I have to say it looks like exactly the kind of game I'd like to play! :D I hope you'll expand on some of those ideas in future posts.

    Thanks. The ideas aren't really that innovative or even valueable. The exact implementation, that is actually fun to play, is what's innovative and maybe valueable ;)

    ReplyDelete
  14. Steven:

    On the CPP, are you wanting to make make new characters competitive as quickly as possible?

    New characters will be viable as soon as the tutorial is finished. They will become competetive within about 50 hours. However, keep in mind that duels take at least 2 minutes, more likely 5 minutes. This is something where the simulation aspect is completely ignored. Also retreat, fleeing, is often a possibility. The design intend is that pvp is possible, but not always encouraged. Playing bandit is going to be high risk, high reward. Playing trader in reasonably secure areas is going to be low risk, low reward.

    ---

    Also in a connected question, how important to your character progression is player skill? Would you try to group people by skill at all? Allow more skilled players to advance more quickly?


    Player skill is inevitable going to play a role, but is not a focus of design. The 'easy to learn, hard to master' does not apply to playing your char, but to playing within your group.

    If you want to be the hollywood hero, go play WoW. If you want to have the theoretical option to become the mastermind who defeats his enemies without even taking part in the final battle, come with us ;)

    ---
    Or do you intend to deliberately mix players of different skill levels? It's my impression that this is at least part of what FPS's do with kill score advancement. Which could be completely wrong, but that's what it looks like to me.

    The design intend is to make the hardcore work for the casuals by organizing and possibly grinding for them! Mixing players of various skill levels is actively encouraged. One major focus are tools that allow hardcore players and estabilished guilds to attract new players to their group. The rest of the game will make this very desireable.

    A very well equipped and extremly organised strike team of 10 players might win against a random collection of 15 reasonably skilled players. But they will lose to 20 of them! To be that well equiped, the strike team will also need to have a headquarter. Simple bandits in the wood do not have the economic capabilities to wage a long term war on anybody. Almost realistic weigth limits, among other things, play a role here.

    This is not at all a typical PvP game. Don't mistake 'full loot' for mercyless PvP. Being the evil guy is going to be risky and players will find that, just like in real life, playing along the rules is attractive, and playing against the rules can gain you fame - for a little while.

    ReplyDelete
  15. Jesse, Darkfall messed it up by encouraging destruction over creation. A fatal mistake. They also had a naive skill system that was almost designed to be abused by botting.

    And Darkfall is almost a FPS and focuses on player skill.

    I see that the bullet points are superficially similar to Darkfall, but I didn't even play Darkfall for more than 14 days. Me, personally, I enjoy PvP for tension and immersion, not for competition.

    ReplyDelete
  16. Klepsacovic,

    Please no. This pushes players into taking these skills as soon as they can, reducing choice. [..] Didn't EVE semi-recently remove these sorts of skills because they were doing little more than creating a huge time sink before players could learn 'real' skills?

    You misunderstood me. As described in the example, you have an easier time raising your general 'melee combat' skill by fighting with various weapons instead of onyl one type. Also some skills will make learnign other skills slower.

    I am not talking about pure learning skills. They are a gameplay mistake that don't even suuport the simulation aspect.

    ReplyDelete
  17. Garumoo, not much to add. ;)

    ReplyDelete
  18. "I think diminishing returns [on learning] are a great idea, but I was more concerned about the grindiness of the underlying activity."

    A skill-points CPP game doesn't need a system of accumulating generic XP, but that doesn't mean you can't do the same kind of testing. That is, if the situation would (in an XP based game) result in an award of XP then permit the awarding of skill points. Shooting fireballs at target dummies or grey-con trash mobs wouldn't award XP, and thus also shouldn't award skill points.

    Cast those same fireballs at some red-con mob well above your level, and if you survive to the end of the fight you get whatever skill points you potentially earned during the encounter. If you die then you don't, so no suicide skilling thank you very much. (Ooh bonus: a life bonus aka death penalty which skillful play, vs graveyard zerging)

    ReplyDelete
  19. A really quick note:

    "After some amount of skill gain (equivalent of about 2 hours) your character becomes tired and starts to gain skills ever more slowly until rested (logged off) for at least 8 hours"

    Most games avoid doing this sort of thing now, because penalizing players for playing more is pretty much opposite from what a game should be doing. A simple fix is to just set it up so that being logged off for at least 8 hours leads to gaining skills much faster, as in WoW. Then just have the base speed of skill gain be fairly slow.

    ReplyDelete

  20. Wolves will need weapons to attack. Their only source of weapons is by raiding the sheep. Who have weapons. And walls. And paid militia. And technology. And science (or magic).


    wolves can have alts providing them weapons. But most of the time raiding is usually enough

    If you want go with historical examples raiders (such as vikings or mongols) constantly dominated for much of early history. No walls could stop them. Only when sheep got organized and well run large military it started working somewhat

    Never underestimate what a dedicated group of hardcore players can do . Balancing such thing in MMO setting is not easy

    ReplyDelete
  21. Verilazic, a valid concern. What you need to consider is that the game is in no way meant to be only a character-power-achieving game, like todays MMORPGs usually are.

    Just because you gain skill points more slowly doesn't mean that you cannot make gold anymore or cannot help your guild. Or cannot simply have fun.

    You still have a point, though.

    ReplyDelete
  22. Never underestimate what a dedicated group of hardcore players can do . Balancing such thing in MMO setting is not easy

    Max, I think developers just don't try enough.
    I might write something about sieges done right in the future. Essential part of it is that a siege is not a two hours experience, but an effort over many days or weeks. You need to win the economic game to win the siege. Being good at PvP helps, but is not sufficent.

    If you want to make a foreign-founded group of 6 elite PvP players, you are welcome to do so. But there ways to make sure that this number of players has no chance against a force that is vastly superior in numbers.

    What the 6 elite PvP players would never be able to do is capture a reasonably defended castle, because the actual PvP is only part of what is required to do this. They might want to sell their help to other groups, of course. In the right situation they could be highly effective.

    The superiority of numbers vs. skill is, btw. the central idea behind balance. Sure, you might find the perfect way to create a strong character. But you will still lose against a force that is significantly larger; the rules of the game enforce it. That doesn't mean that you are not rewarded for being better than the rest. It just scales down the entire benefit of min/maxing the individual character.

    In non-instanced conflict, numbers dominate everything else. You win wars and battles due to preparation, not due to execution. Being good at execution is just a bonus.

    ReplyDelete
  23. "If you want go with historical examples raiders (such as vikings or mongols) constantly dominated for much of early history."

    be careful here - the "Hollywood" version of history was that Vikings were inveterate raiders and seafaring brigands, but the reality was they were more involved in trade and commerce.

    "wolves can have alts providing them weapons."

    those alts would thus either be sheep themselves (and thus open to the same depredations), or fifth column elements within sheep communities. For the latter, how would you propose such a sheep would manage to get the goods to the raiders, given that there are no teleport travel, no goods teleport, etc?

    ReplyDelete
  24. What is required to get equipment? Do you need certain stats for equipment, certain level, both? You might have said this and I missed it.

    I don't think that the "equipment" method is the right direction, having players be able to switch roles around cheapens the sense of place their character has. If the game has enough depth in its systems a player should be able to find satisfaction in whatever direction they take.

    For horizontal progression: have different aspects of the three main roles and have different ways of approaching those aspects. Some games already doing this to some extent.

    An example: there could be multiple types of "healer" or "tank" or "combat/dps" that balance between different aspects and play-styles of their specific roles.

    On top of this you could have a "stat-based" progression, a "skill-based" progression, or something in-between for each of these.

    ReplyDelete