Psychochild's Blog

A developer's musings on game development and writing.

10 December, 2007

Weekend Design Challenge: PvP design
Filed under: — Psychochild @ 12:54 AM

A recent post on Broken Toys talks about the problems the PvP game Fury is having. The short version is that they ran into a lot of the same problems that developers saw back in the days when M59 and UO launched.

So, let’s talk about PvP design and how to do it “right”.

Try to be constructive here. Let’s assume you’re a designer on a PvP project; if you’d just quit, go read something else on my blog. :)

The first major problem, as the CEO is quoted as saying, is that you need to do something to allow for a reasonable learning curve for new players. Getting in and getting ganked immediately doesn’t really make people happy. I also think it’s important that you give your players a “winning” experience to get them excited about the game. Most PvP combat tends to have absolute winners or losers, so having everyone, even inexperienced newbies, get a winning experience is hard. One thing mentioned in the comments of that thread is to even allow the losers to gain something from the encounter so that they feel some sort of advancement.

What are the other design features are important? And why? Bonus points for considering the smaller scale, since people with money are going to be hesitant about throwing large amounts of cash at a PvP-focused title.

« Previous Post:


  1. David Sirlin would argue that PvP with advancement is counter to the whole point of multiplayer games in that you should be delivering a level playing field to both opponents.

    But if you are going to do PvP with advancement, I’d recommending taking a leaf out of the Call of Cthulhu design and have a learn through failure approach. The loser gets to pick one ability that the winner has learned, which they don’t know. The winner improves the ability picked, and the loser learns it.

    That immediately makes the advancement through PvP technique have it’s own tactical component in terms of which opponents you select. By restricting the ability chosen to only something that the loser doesn’t know, you also encourage people to go out and find enemies that they haven’t fought before. They may even be encouraged to set up ‘dojo’ like systems, in order to train as many beginners in the ability they want to improve.

    Comment by Andrew Doull — 10 December, 2007 @ 2:01 AM

  2. Fury is failing because it is so limited in the things you can do. People will argue to the ends of the earth that the game has plenty of so-called “depth”, and it does – about as much as Magic: The Gathering cards. Like M:TG, Fury only has depth in one or two aspects and that’s were it stops. Once you’re sick of fighting and grinding to gain new abilities, there’s nothing else to do. People are quitting because there are no real quests to pass the time. There are no Mobs to practice on. Good weapons and armor cost an arm and a leg to buy and about half of that just to change their color. Top it all off with crappy loot that you can neither use nor sell and you have a game doomed to die.

    I really think there’s anything Auran can do to make Fury even moderately successful other than make it a completely free game. I mean, why pay $90 AUD for a game so restrictive when I can play one of the hundreds of free MMOs instead?

    Comment by Auguste Sentinel — 10 December, 2007 @ 3:48 AM

  3. Throughout this comment I’m going to talk about “numeric” rewards. It’s a convenient shorthand for “better gear, improved stats, experience points, and hot chicks”. I’m also going to frame this in terms of mistakes, rather than suggestions, because hindsight is easier than foresight (particularly at 3 AM).

    There are two “must not”s for making a successful PVP game:

    1. You must not reward PVP victory with increased power.

    This should be obvious, and I’m continually shocked when it isn’t. Fury (and Shadowbane before it, and countless other games before those) rewards players who succeed in PVP by making them numerically better at PVP. The people who fail at PVP and who need the most help, on the other hand, are for the most part left to rot. As a result, the people who are good at PVP wind up with both a skill advantage (which the player supplies) and a numeric advantage (which the engine supplies), which combine to eventually make the victors unbeatable.

    WoW gets a pass on this because you can become as numerically powerful as a PVP catasser by avoiding PVP entirely. This is one way to enable the game to numerically reward success: by providing alternate ways to get equivalent numeric rewards. Fury, on the other hand, has essentially no PVE game to support alternative rewards on.

    2. You must not throw new players in with the old.

    Players just entering the world for the first time, or re-entering the world with a new character, are at a numeric disadvantage. They almost invariably have crap gear, no powers or spells to call on, and essentially no reputation (and thus no friends to call on). More importantly, the truly new players entering the world are at a skill disadvantage. Even if they’ve played another similar game before, the presentation and exact execution of the game mechanics is new and unfamiliar.

    Creating an environment where established players can fight un-handicapped against new players is akin to shooting your own revenue stream in the foot. Nobody likes to be beaten by an apparently-overwhelming opponent regularly, but if new players are not permitted to hold back and learn with people at their own level then a signifigant number of established players will sit there and pick off new players for fun.

    It’s possible to go too far with this; new players who, for whatever reason, actively want to swim with the sharks should be allowed to do so if it’s remotely possible that they’ll win. If possible the game should also allow a newish player who has tried to go up against more-experienced or better-established opponents to retreat to safety. The decision to try harder opposition is much easier to make if you know you can back down again afterwards.

    Comment by Angstrom — 10 December, 2007 @ 4:06 AM

  4. Newbies are targets regardless if the game is billiards, poker or M59. Accepting that, you are presented with the basic Risk vs. Reward formula to deal with. If the reward is the same no matter the attacker and attacked you can’t escape the current problems.

    Slapping down a “Newbie-zone” is easier to program than a status database, so there is the question of how complicated the solution can be before the CEO balks.

    Empower the fallen with some righteousness-bonus against any they fall to in their newbness.

    If you can determine and define how much punishment newbies should encounter you could build the regulatory system into the game environment. If some PK is going hog wild on the newbs, knock him down with blue bolts from heaven.

    Interesting things can be done with faction and supply economics to make the system more robust.

    Comment by aozora — 10 December, 2007 @ 5:48 AM

  5. The biggest thing I would do is design the PvP game first, and then figure out how to add mobs to the system. Therefore the computer controlled enemies would probably be more like Quake deathmatch bots than WoW monsters. The idea is that PvE would be legitimate training for PvP. An expressly mandated policy for ongoing updates would be making computer controlled enemies adopt new tactics as the playerbase discovers them. In a lot of games, players who don’t focus on PvP are thoroughly unprepared for the differences between PvE and PvP combat.

    Comment by Vargen — 10 December, 2007 @ 7:02 AM

  6. My “quick fix” for Fury would be to:

    1) Reward everyone for participating even if they lose, but reward winners more.
    2) Never match up seasoned players against newbies.

    The way Fury PvP works (or worked in beta when I played it last) is basically instanced “battleground” type PvP, so it doesn’t seem terribly hard to separate players. If they used a similar matchmaking system to Halo 3 with the additional mechanic that higher rated PvPers got cooler stuff (and I think they do), that seems like it would be sufficient.

    Comment by Ryan Shwayder — 10 December, 2007 @ 7:19 AM

  7. I’m glad you asked this. Scott had a lot of good comments on his blog: I hope you professional types were taking notes there!

    Here are the elements I would design around for a rewarding pvp experience:

    1. For every attack, there is a counter. Many games tried this (daoc’s purge, for example, to counter mesmerize; or the anti-mezz some classes got later in the game). But the most basic is the concept of damage vs healing. One healer should always be able to outheal a damage dealer. If they can’t, why would you have healers on your team at all? Tuning those levels appropriately is hard because of “interrupts” like silence, sleep, etc. But make sure that the counters are meaningful, or everyone will just make teams of ten FotM DPS’rs.

    1a. A defensive class should be able to beat an offensive class, or everyone will make offensive classes, and your game will be boring.

    1b. No unbreakable/uncounterable attacks. I know they look fun. They’re not. Example: Wow’s paladin bubble (fixed: priests can mass dispel now); DAOC’s instant mezz (fixed: purge, unmezz). Notice how these games realized their mistakes and fixed them.

    2. Pacing: Combat should last a minimum of 30 seconds. It is not fun to die within three seconds of combat starting, and it is very hard to improve with combat being so quick. Many games make this mistake.

    3. Crowd control is limited to one enemy target. ie, no AE mezz. Use timers to prevent crowd control abuse, ie, warlock chain fear. Losing control of your character and watching him die is not fun.

    4. No instant kills. I know they look fun; they are not. Example: One-shot stealth kills in just about every pvp game made, ever.

    5. Promote the use of effective tactics over numbers. DAOC got this right. Superior mobility/rapid response forces were very powerful. Flanking maneuvers (where you could take out their ranged or casting classes first) were meaningful.

    6. Do not reward pvp success with game-imbalancing pvp gear. Critical! People will be motivated to win pvp battles if you give them a forum to brag in. Improving the victors just guarantees you’ll never have different victors.

    7. If you must use a rock-paper-scissors design, make sure your rocks, paper and scissors are spread out on multiple classes. Nobody likes having to wait around for two clerics, a sorcerer and a minstrel before they can start playing. Warhammer’s “multiple loadout” design suggests they’re paying attention here, but I don’t know enough about the game to tell for sure. Wow did a pretty good job of this in PvE, but pvp tuning is hard, and small differences between classes are more meaningful.

    8. Work hard to promote pvp conflict. Some skittish players are unlikely to try it otherwise.

    9. Make a “safe zone.” Few people like hardcore pvp games where they can’t leave their seat for five minutes to take out the trash.

    10. Promote experimentation: encourage people to try different builds and team combinations.

    11. Losing a few times should not mean the end of your career. Make sure you do not put players in a spot where they cannot win (gear loss, unbeatable teams).

    Comment by Axecleaver — 10 December, 2007 @ 8:11 AM

  8. I think it’s important to preserve some degree of hope that things might turn out differently next time.

    With that in mind, build arbitrary choices into the combat system that help define a player’s style, eg 3 spells that have the same effect but induce different vulnerabilities. By watching and being repeatedly beaten by the same player a pattern might start to emerge. Elite players beat up on hundreds of others, and as a result those hundreds of players are probably being beaten by the same few people. If the combat system is built so that it’s possible to develop a counter-strategy for one person, that puts the infamous at a disadvantage.

    Random ideas along the same lines as #1:

    - Ralphie’s rage: Chance on hit (or death) that a character will significantly increase in power, enabling a Scott Farkus-esque beating that satisfies not only the perpetual punching bag but also the horrified onlookers. The recipient will shake it off as a one in a thousand occurrence, and the issuer gets to vent built-up rage.

    - Role reversal: Chance on hit (or death) that players will switch character attributes (everything but looks/name/etc) for a period of time. 75% of everything gained (experience, loot, whatever) by one is given to the other. Punching bag essentially gains a slave (likely with a lot of time on his hands) for the day plus a loaner character that they likely would never be able to attain themselves. Don’t want to end up playing a level 4 character for the day? Don’t gank level 4s.

    Comment by mcj — 10 December, 2007 @ 8:53 AM

  9. You need a narrow power band so that the ‘elite’ aren’t orders of magnitude better than the ‘noobs’. You need meaningful PvP objectives so that people are fighting for more than just self aggrandisement. You need people who just lost to have fun as well. You need a meaningful death penalty but not one that allows for chain griefing.

    The current MMO orthodoxy doesn’t fit that as it revolves around making a character more powerful. What if you didn’t play a character but instead played a side? What if you had a bunch of guys, lets say in your Generic_fantasy_universe_01 that were a goblin tribe? You control one of these guys at a time. The ones that you aren’t playing have some kind of offline advancement system that allows them to improve based on the gains you make with the guy you are currently playing. If he dies, he dies. Yeah permadeath. It works if you aren’t putting all of your eggs in a single basket. There you have your meaningful death penalty which also allows self correcting systems like bounty hunting as a way to stop the top guys from getting too far ahead of the curve.

    The exit barrier is still plenty high enough as long as people have something to fight for over the long haul, an actual objective to aim for, rather than simple power-ups on the guys who win fights.

    Comment by IainC — 10 December, 2007 @ 9:32 AM

  10. PvP MMO Design

    [...] 10, 2007 at 7:49 pm (Musings) (design, industry) Brian ‘Psychochild’ Green has an interesting discussion going on at his place about PvP games and how to design them. I’ve put my tuppence into the [...]

    Pingback by Adventures in MMO Community Management — 10 December, 2007 @ 10:49 AM

  11. OK, we need a game that’s fun for newbies, so we could just remove any benefit from playing longer, even skill benefit. This isn’t going to be much of a game.

    So, we need to assume that players who have played longer are going to do better. Be this through mechanics (equipment, unlocks, character levels) or just through knowing what they’re doing better.

    Random example from the “real world”: Chess.

    How do they handle the masters “ganking” the newbies? Ratings, handicaps, if you have a Grand Master pwning a beginner they don’t get any improvement in their ranks.

    So, reward for killing newbs should tail off pretty dramatically – this encourages the masters to play other masters.

    We still have to solve the other half of the problem, making it fun enough for the newb to lose (since they will) to keep persevering and playing.

    A loss should never be a total one, give points for staying alive if need be :) (woot! I managed 20 seconds this time!). Points for putting up a good fight, points for nearly winning (but not quite).

    Newb group 1 goes into the arena vs uber group 2 – they last all of 10 seconds. But since the rank difference was so terrible uber group 2 get almost nothing (if not 0). However the newbs in group 1 get some compensatory points for lasting a whole 9 seconds longer than expected.

    Having goals which aren’t zero-sum can help – I kill you or you kill me generally devolves into “Uber group wins”. But if you have objectives to capture and hold, defensive fortifications to hide behind you can have a bit more fun with victory points: Sure, we lost to the uber group but damnit we took the town first!

    The top tier of your uber players will always want a challenge – lasting 30s isn’t exactly fun for them, so you need them to fight the other top tier groups. They probably will do this anyway, but you could throw in a ladder or a league or something, 0 points for newbie ganking (or even -ve points for newbie ganking ;)) and peer pressure should hopefully cover the rest.

    I dislike the “random chance of maybe winning” from both a ganker and a gankee’s perspective. It’s frustrating for the ganker when it happens and it gives no sense of achievement or progress to the gankee.

    Comment by Flim — 10 December, 2007 @ 1:25 PM

  12. @Ryan Shwayder (Post 6):
    Fury has a pretty well-written match making system that ideally will never match newbies with veterans. But when there are only 100 or so people online at once, its options are pretty limited, so it was designed to use whatever’s available rather than cancel the match. You could say Fury is digging its own grave, as the more players leave because they are getting sick of being pitted against veterans players, the worse the problem will get.

    Unfortunately, everybody still gets the same useless loot with winners only getting marginally better items than losers. Usually the items are too poor in quality to use and are worth almost nothing to sell.

    Comment by Auguste Sentinel — 10 December, 2007 @ 2:40 PM

  13. How do you do PvP design right? I assume we’re talking about MMORPG PvP here as there’s lots of great examples of PvP in other game types that don’t seem to garner the same level of scorn as MMO versions – FPS (Battlefield 2, Team Fortress 2, Halo etc), RTS (Starcraft – the Koreans have this as a national sport, C&C, etc), Turn based strategy (Civ, Laser Squad Nemisis, Empires etc), Board (Chess, Checkers, Go, Risk etc), Card (poker, bridge, magic the gathering etc), not to mention sports (football, bowling, athletics etc). In fact most games outside of the MMORPG realm seem to be doing OK in terms of PvP. Why is it that MMORPGs seem to have so much trouble getting it right?

    I wonder if its a function of the genre and the people who play it more than anything else. I think Angstrom hit the nail on the head when they mentioned not rewarding victors with increased power. The problem with this is that it runs counter to how your typical CRPG is constucted. The D&D level structure is just not suited to a PvP situation (I would argue that it severly limits your PvE as well, but I’m trying not to start too many arguments). I’m not sure if rewarding the losers more is appropriate though, most other game styles try to start with an approximation of a level playing field and once the game/match is won the game is reset back to initial states.

    For myself I like the approach that Guild Wars and to some extent Fury took in having a lot of abilities to choose from and allowing the player to construct their PvP avatar from amongst them. Where I think the problem comes is when different players don’t have the same base of abilities to choose from (I include items/equipment in the catch all ‘abilities’). The second problem is not having a good in game and safe method of testing out and trying the various combinations. Guild Wars has remedied this by having PvP tutorials in game, Fury’s effort however is a little lackluster. For it to be effective you need to have a well designed learning curve and a non-threatening but effective test arena – a good bot arena with great AI. I’d further design it so that the player can set all of the test arena (or as much as possible) parameters – choose the map, set bot abilities and skill levels as well as play styles (eg: aggressive assault, sniper, defender etc). I’d also want these test arenas to be available to groups of people so a player can gather a group of their friends or even a pick up group and play/test strategies before using them in the real world (this is why most sporting teams have practise… you need this to get good at the game and prepare strategies for play). This sort of thing has been done well in a few FPS games – timeplitters series on Xbox and playstation was great for controlling almost every aspect of the multiplayer game.

    I also like the Guild Wars ability to watch old PvP matches, the problem with it is that it moves a bit fast to get a real grip on what is going on unless your a veteran player anyway. So I’d look at having ‘commentators’ – great matches are played back with slow mo and ‘experts’ comments.

    In terms of facing off against opponents that are much ‘better’ than you are, I’d try to impliment an optional, by mutual agreement, handicapping system (if it works for golf why not MMOs). Another aspect of Guild Wars that I’ve liked is the special event PvP matches where you have your character set to a base state with the same abilities as everyone else – this levels the playing field (apart from ping and general player skill, not to mention familiarity with the match). Of course you’d still want the option of being able to test yourself against the ‘better’ player as a victory in this case is so much sweeter and a loss not so upsetting.

    I could go on, but this post is getting a bit long in the tooth. I guess my approach to the design would be to look to take an almost ‘sporting’ approach to it. Focus not just on the match, but also on learning the game (which tends to be an afterthought, but I’d like to try designing this as a central componant of the game – perhaps even extend it so that it becomes a PvE style aspect – allow players to have access to a fully functional practice environment is key I feel). Have rewards that don’t necessarily add to player power, perhaps they allow specialisation in role or are simple cosmetic improvements or bragging rights (look at how popular XBox achievements are).

    Comment by TickledBlue — 10 December, 2007 @ 3:23 PM

  14. Off the top of my head…

    I agree with Angstrom’s first point that PvP victory should not be a means (an optimal means, at least) of gaining individual power.

    But I think the opposite of his second point is a potential solution to the learning curve. Make PvP exclusively about RvR conflicts (which is to say that dueling is possible, but not rewarded) and reward veterans for looking after newbs. If the numeric power between them is not huge, then they can journey and fight together early on.

    RvR rewards even the weakest of players by including them in group victories. Individual victories can be possible, but focusing the game on group-vs-group encounters — with groups not limited in size — enables relatively ineffective players to meaningfully contribute and even turn the tide of battle. Star Wars: Battlefront and similar games have proven that many gamers (myself included) enjoy intense battles with compelling combat even when those gamers play only a limited role in the battles and suffer minor forms of defeat repeatedly.

    Then provide a variety of encounter scenarios so that different playstyles and character types experience experience their day in the sun.

    Comment by Aaron — 10 December, 2007 @ 3:28 PM

  15. I think a tiered system of unlocks, while simplistic, is also ingenious. Battlefield 2 does this very well. If you’re not familiar with the system, you gain ranks through accumulating experience. At each rank, you gain points you can use to unlock weapons. Essentially, you level your character to gain access to more versatile and different weapons. A player can also choose between classes each time they die, changing out weapons or switching between unlocks.

    Any unlock a player gains should not be more powerful than an existing weapon. It may be more useful in different scenarios than a starter weapon, but not necessarilly be more powerful.

    To make this work as a MMOG, you’d need conquerable outposts and cities. A player, upon creating a character, would then choose a faction to align with. Each faction would have headquarters that can be attacked, but can not be conquered. To conquer a city, your faction would need to gain control of say.. four of the five capture points within the city. If you’re within the area when your faction captures a city, that counts as a win. Players would build profiles online where their stats can be accessed and gain ranks and unlocks in-game.

    Throw away death penalties and time sinks. I know there are masochists out there, but ultimately, you want the game to be fun the minute the player logs on. I think it is important in any PvP game, to be able to log on and get into a battle immediately. Players could form fellowships, or squads, and squad leaders could create moveable spawn points. Battle commanders could be elected in each area, depending on who is at a command station. If someone is commanding and a player with higher rank comes along and wants to command, they could force the other player into battle.

    To summarize, I think a PvP MMOG should play much like a multi-player first person shooter. Emphasis on teamwork and tactics, without overly-complicated spell, item or skill systems. Don’t try to fit a FPS into the MMOG mold, but create a FPS in a persistant world.

    People pay thousands of dollars, every month, to maintain FPS servers. Create a game that is appealing to them, give them thousands of other players to play with in a persistant world and add content regularly. If you don’t screw it up, you’ll do very well.

    People like to compete. The learning curve is small. Some people will naturally be better than others. No one needs to be completely equal. There doesn’t need to be solo content, or NPC content, if there is always a battlefront.

    There doesn’t even need to be NPCs. But if they is, allow players to form squads and recruit AI soldiers to take into battle with them. Nothing more.

    Comment by Septa Scarabae — 10 December, 2007 @ 6:15 PM

  16. As a added note on fantasy games, Mount & Blade combat is very similar to what I would expect, with the exception of accuracy being skill based.

    Comment by Septa Scarabae — 10 December, 2007 @ 6:19 PM

  17. I feel that Guild Wars does PVP right in almost every way. But if I wanted to make a new system that didn’t rip from an existing videogame system it’d be as simple as these three words: What’s your handicap?

    The handicap system in Golf is ideal for videogames — most of the thrill that good PVP players get is from other players knowing how good they are. Let them display their handicap ranking as a title, and reduce the number of skills they can equip (or reduce the power of these skills) accordingly. Good players shouldn’t get stronger. They should get [u]weaker[/u]. As in, [i]I could beat you with one arm tied behind my back.[/i] I know I’d love it.

    Possibly the best part is that new and casual players get to try things out and stick in the noob battles if they’re not interested in fame, and they don’t have to put up with ‘competition rules’ and can just fool around.

    Comment by Mike Blackney — 10 December, 2007 @ 10:21 PM

  18. Same things I posted on Scott’s blog as comment.

    While there is no formula for creating the perfect PvP game (or it would’ve been made already) It is more important to adjust the PvP to match the design of the game and the vision of the creators. This way the immersion of PvP remains and can be tuned without ruining it.

    1) Do not let equipment take control over skill of player
    Equipping for PvP should not take up more of the week than actually playing the PvP
    This was a HUGE problem in the beginning of WoW.

    2) Don’t let instanced PvP take control over continuous PvP areas
    If the PvP is detached from the game world it will just be a minigame in a PvE game. Best example I can think of is WoW. PvP has absolutely no effect on the game world and world PvP is almost nonexistent since it doesn’t offer any rewards (lose-lose situation)

    3) Do not let player skill take over character skill
    What I mean here is that only give the player control to command the character they play not to control the character itself. This is a thin red line I know, but where I’d draw the line is having the character do the combat instead of letting the player do the combat. For example aiming shots/strikes.

    4) Last but not the least. make it possible for smaller groups (to an extent) to beat larger groups if they have more skill. In DAoC this is achieved by numerous types of crowd control. I know everyone complained/still complain about CC in DAoC, but it is none the less the soul of the PvP there since it gives this chance.

    Comment by Scorpio — 11 December, 2007 @ 5:40 AM

  19. Variant on suggestion by mcj:

    If attack first and kill a player with a lower level, you lose all abilities and stats above that level for 24 hours. If you kill a player with a higher level you gain those abilities for 24 hours. Losing doesn’t affect your status. Of course we’re now into the territory where it’s worth a low-level character provoking a high level character into attacking them just to grief them, but that hopefully would relatively rare and could be handled by normal abuse channels.

    Assuming there is some other reason for taking the combat risk – factional points, gold, kudos, ranking.. there should still be enough incentive for combat, with a real (but only short-term) risk.

    A skilled player could bring in a Level 1 char and be running at Level 80 pretty quickly – if they have the skill, but they have to keep killing (Players of a comparable level). Could be a interesting (or insane) mechanic in a vampiric setting. Not a great way of reducing obsessive play however. (Perhaps it should be 8 hours of ‘online’ time)

    Comment by Idlethought — 11 December, 2007 @ 6:36 AM

  20. Scott Jennings wrote (and then edited) a post that might be of interest to people discussing this issue:

    Comment by Psychochild — 11 December, 2007 @ 3:53 PM

  21. To be succinct: Move progression off the character.

    Comment by Grimwell — 11 December, 2007 @ 6:14 PM

  22. My thoughts on this (I am primarily a PvE person but I’ve enjoyed PvP somewhat).

    1. Give folks something meaningful to fight over. Many PvPers will get into ladder scoring and such, but if you’re going to set PvP in a VW, then you might as well give them some kind of additional hook – something that encourages them to keep fighting even when they’re not on the winning side. This is especially important if mass PvP is the goal – you need to have some kind of RvR-style objectives.

    2. Rock-Paper-Scissors balance. So many games get bogged down in “he said, she said” between classes or skill sets, and generally all this accomplishes is that it homogenizes people’s characters, or leads to flavor-of-the-month skillsets. It’s better off if you design from the beginning with a rock-paper-scissors balance – so you know up front that as designed, a rogue should be able to beat a priest, but a warrior can defeat the rogue, but a wizard can defeat the warrior, but a priest can defeat a wizard.

    3. Minimize the impact of gear, and avoid large incremental increases in character power. Sure you want to have some – everyone wants to see their characters advance – but ideally you want a system where the skillful newbie can bring down the lazy veteran in a fight.

    4. Include social hangouts. Just because most of your gameplay is built on gladiatorial combat doesn’t mean that gladiators don’t have lives outside of the arena. Give players places to meet, swap stories, and make friends, and reasons to do so (this is easy if you have a PvE game as well).

    5. Avoid complicated rulesets. If you need to throw a bunch of artificial restrictions into your PvP system in order to insure fairness, then you need to redesign the system. Complicated rulesets just encourage players to find loopholes and discourage players who want more “free-form” action.

    Those are my thoughts for now..

    Comment by Talaen — 12 December, 2007 @ 10:10 AM

  23. Wow, it seems I was right (Post 2). The other day Auran announced Fury will become completely free to play as of December 14th, the day the Age of the Chosen expansion will be added. Of course, subscriptions will still be available and free players will be unable to trade and earn less coins on top of the restrictions to players who bought the game but don’t subscribe.

    Comment by Auguste Sentinel — 13 December, 2007 @ 1:33 AM

  24. Idle thoughts (yeah, late to the game again):

    - Have an alignment/class system like Ogre Battle (really old SNES RTS).

    – Players can change class at any time (perhaps by spending money).
    – The advanced (hence more powerful) classes have alignment requirements: Good (e.g. pally), Neutral (e.g. druid), Evil (e.g. necro).

    – If you attack or kill a much lower-level player, your alignment goes down. Once your alignment drops below e.g. 70%, you become “neutral”. Once your alignment drops below 30%, you become “evil”.
    – You can raise your alignment through showing mercy (refusing to kill a defeated enemy) or helping the weak (rescuing a newbie from a ganker).
    – If you can no longer fulfill your class’s alignment requirements, your character reverts to a basic class with no alignment requirements.
    – One possibility: Players don’t die when their HP reaches 0. They enter a state where an opponent must use a special “execution move” to finish them. If they are not executed within a certain timespan (10 seconds??) then they have some kind of escape option (perhaps being able to teleport to a safe zone, or being invulnerable for a short while but unable to fight).
    – Another possibility: Players might have abilities which allow them to knock a weakened opponent out, rendering him invincible (to prevent them from helping an evil player to gank him) but unable to act, and effectively ending the fight.

    I’m sure there are ways to game the system, but it seems to me that this would introduce an interesting dynamic if managed right. Many players like to think they’re good people (see: Alliance zerg in WoW), so why not lean on them to “earn” their goodness? Picture the newbie dropped into a raging battlefield: within seconds, an evil necromancer swoops down on him, unleashing a bolt of dark energy. His HP drops to near-zero instantly, and he’s about to collapse. The necromancer raises his bone scythe to deliver the finishing blow, but before it lands, an armored paladin steps in between them and the scythe clashes harmlessly off his tower shield… Of course, the paladin is merely farming alignment, but our newbie doesn’t have to know this ;)

    Comment by n.n — 13 December, 2007 @ 7:46 PM

  25. n.n said “One possibility: Players don’t die when their HP reaches 0. They enter a state where an opponent must use a special “execution move” to finish them. If they are not executed within a certain timespan (10 seconds??) then they have some kind of escape option (perhaps being able to teleport to a safe zone, or being invulnerable for a short while but unable to fight).”

    I’ve seen something similar in a MUD, maybe Medieva? If your HP reached 0 you’d become unconcious and start losing blood. You’d only die when you got to -10 HP. If somebody could stop the heavy bleeding you’d be alright, otherwise you’d bleed out.

    Comment by Auguste Sentinel — 14 December, 2007 @ 12:30 AM

  26. PVP Done Right?

    [...] Psychochild started up a nice related entry to this with some great ideas posted by the community « Variance Shadow Mapping and Light Bleeding The Bandwidth Pit » [...]

    Pingback by Kreation’s Edge — 27 December, 2007 @ 12:02 PM

  27. I have a few of my own potential systems that might help (in addition to many of the good ideas above):

    Give newbies some “free” defensive/passive skills which make them very hard to kill by other players in a PvP situation to combat ganking. Let them choose whether or not to later “trade in” these skills for others they might want. Doing so is not mandatory, but may be desirable. Of course, doing so will “open them up” so to speak to PvP combat permanently once they trade them in, giving them the choice to stay PvE for their entire career or go PvP once they realize it isn’t.

    Provide a “revenge” system in which victims of griefing (high level PvPer attacking a lowbie, etc.) can get a free bounty against that player, for other players to accept for monetary gain – which is then split 50/50 between the player who kills the ganker and the low level player who was ganked.

    As much as I hate to punish players – punish gankers – NPC bounties, fines, higher merchant prices, “outlaw status” for x hours/days in which all NPCs will attack on site, etc. Many possiblities here

    Of course, some of the most critical points to good PvP are to have goals and objectives, not just mindless player on player killing. Give them a good reason to engage in PvP – and make it as enticing as possible to encourage non-PvPers to give it a shot, and to also help them realize that it’s not as bad as they once might have thought. This of course means we have to design systems that ensure it is in fact not all that bad. :)

    Comment by Matthew Doyle — 28 January, 2008 @ 8:58 PM

Leave a comment

I value your comment and think the discussions are the best part of this blog. However, there's this scourge called comment spam, so I choose to moderate comments rather than giving filthy spammers any advantage.

If this is your first comment, it will be held for moderation and therefore will not show up immediately. I will approve your comment when I can, usually within a day. Comments should eventually be approved if not spam. If your comment doesn't show up and it wasn't spam, send me an email as the spam catchers might have caught it by accident.

Line and paragraph breaks automatic, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Email Subscription

Get posts by email:

Recent Comments


Search the Blog


October 2020
« Aug    



Standard Disclaimer

I speak only for myself, not for any company.

My Book


Around the Internet

Game and Online Developers

Game News Sites

Game Ranters and Discussion

Help for Businesses

Other Fun Stuff

Quiet (aka Dead) Sites

Posts Copyright Brian Green, aka Psychochild. Comments belong to their authors.

Support me and my work on