Psychochild's Blog

A developer's musings on game development and writing.

16 March, 2007

Weekend Design Challenge: Feedback loop
Filed under: — Psychochild @ 9:21 PM

We’ll go practical this week. Let’s say you’re working on a typical combat-focused online RPG. One feature requested on the forums is that players that take wounds should perform worse than a fully healed character. Your job is to design this system to work within a game.

A few thoughts below.

Don’t assume anything special about the game. Assume it’s a typical sword-and-sorcery type game. Use WoW as an example if absolutely necessary.

Think about these issues:

What are the potential problems with this design?

How do you work around these problems?

Bonus: What is the motivation for players for asking for this change?

I’ll post something a bit later.

« Previous Post:


  1. The classic problem with wound systems is the death spiral: the worse the character does in a fight, the less chance there is that he’ll be able to pull ahead and win. This tends to make it hard to judge one-on-one fights, as you not only have to have enough hit points to outlast the mob’s attacks versus your own, you’re getting progressively weaker throughout the fight. Players will tend to maximize their downtime, not progressing to subsequent fights until they’re completely healed so as to avoid starting off at a penalty.

    One way around the last problem is to make as many fights as possible face a time pressure of some kind or another; you can rest to 100% health, but if you do, it will have adverse effects on your next fight or overall mission. Another might be to put in a potential tradeoff: you lose overall effectiveness as you lose health, but at certain points you gain access to desperation moves that can only be used when at low health.

    Some players might ask for this change out of a sense of “preserving realism.” Others might hope that the change will apply to the mobs as well; if they almost always have access to a healer, they may hope to keep their own health totals high for major fights, or let the tank slowly lose damage as the health of the DPS stays at full. Long fights would get much easier for groups that can keep high health, since the standard mob is set up to use its health counter almost as a timer for the fight.

    Comment by Stephen — 17 March, 2007 @ 5:35 AM

  2. Wounds and disability

    [...] Saturday, March 17th, 2007 in Dream game, General Design Psychochild’s weekend challenge for this week is up, and hits on a topic that I’ve turned over in my mind for years. To quote… …Let’s say you’re working on a typical combat-focused online RPG. One feature requested on the forums is that players that take wounds should perform worse than a fully healed character. Your job is to design this system to work within a game… [...]

    Pingback by Voyages in Eternity — 17 March, 2007 @ 7:13 AM

  3. Problems:

    1) Destroys PVP balance, because a fight between two evenly matched characters comes down to who strikes first. This will cause rogues, ranged classes, and anyone else who is able to choose their fights to win against tanks and whatnot which cannot. Expect to see more ganking, more fighting engaged characters, and more zerging since anyone alone is meat.

    2) Increases downtime drastically. Players are psychologically very keen on avoiding risk, and if you’re damaged even 1% in this paradigm you’re actively debuffed. Even if the debuff is trivial players will sit and drink until its gone. Downtime is an absence of fun for players and one should only include it for a good reason.

    3) Complicates healer role/overemphasizes healer skill. In party/raid situations the entire team’s DPS is dependent on the healer not just keeping everyone alive but keeping their HP at optimum levels to avoid whatever the next damage/spellcasting/whatever penalty is. This is stressful for healers (“We wiped! Priests, WTF!”) and frustrating for everyone else, particularly classes which are designed to take damage (“Doods where the F are my heals! I can’t Execute if you don’t keep me above 30%!”) It gives healing classes responsibility for the DPS of the team, which is a major no-no — if the priests’ skill directly contributes more DPS than the mages’ then why ever play a mage? If the mechanism for performance degradation isn’t totally transparent expect an addon (legit or otherwise) which calculates it to spring up in a matter of days.

    Working around the problems:

    1) Slow down PVP. Target an average PVP encounter time which is closer to a minute than 5-15 seconds. This implies that the amount of damage which the first blow does is tiny relative to your max HP, decreasing the influence of who wins the first shots. Similarly, severely nerf any ability to kite (faster than average combat movement + ranged attacks, chill effects, etc). High DPS classes will complain since increasing everyone’s staying power helps tanks/healers the best. Consider making non-mutual escape from PVP fairly difficult so that people can still actually die (entering combat mode causes a massive slowdown in movement speed, for example, going to full sprint requires taking quad-damage as you expose yourself?).

    2) Radically decrease downtime. All characters from level 1 on should have the ability to fill themselves up to 100% effectiveness almost instantly in a non-combat situation. Consider what this does to group combats, PVP in dynamic environments, etc.

    3) Make the effects of damage simple to understand but offer clear exemptions. Tanks use a different scale than everyone else, say, if you lose 5/10/25% effectiveness at 75/50/25% HP then tanks lose 0/5/10% instead. Offer specialization via gear/talents/whatever to players who either hate this penalty or intend to take lots of damage (tanks/brawling classes/PVP-freaks-at-home-in-middle-of-battle/etc) which provides similar bonuses.

    Why they ask for it:

    “Realism”, principally, plus the perception that it makes the game tactically deeper. It comes down to players asking for things which are terrible ideas, like permadeath or corpse looting in item-centric games. Everyone will play so that they can AVOID the tactical issues introduced by this.

    (If you want the tactical issues to re-emerge, consider adding a class with short-term combat-ready buffs which weaken or break on damage. That way there will be no perception that you’re gimped without a botted priest keeping you topped off at 100%. Perception matters — look at how much better WoW’s Rested State works as a 100% Casual Bonus versus a 50% Powergamer Penalty despite the fact that the effects of those two framings are identical.)

    Comment by Patrick McKenzie — 17 March, 2007 @ 8:50 AM

  4. Step 1: Give everyone in your game the ability to heal themselves lest soloing will suck for a lot of people.
    Step 2: Make sure your game is intended for the more hardcore audience market.
    Step 3a: Make the impact rather minor and easy to understand when it happens.
    Step 3b: or… make the impact realistic based on where the player gets hurt and have it as part of the cool factor of your game.

    Comment by Ryan Shwayder — 17 March, 2007 @ 12:39 PM

  5. Two problems off the top of my head.

    1) Getting wounded isn’t fun, in RL or a game. I don’t think anybody ever said “I just lost the use of my arm again fighting an orc. Sweet!”

    2) Because players inevitably want to reset to full health before the next fight, the trouble you went to in making them look battle damaged only lasts a brief period between fights. That’s a lot of resources for something players won’t see long, not to mention that the sudden return to full health is counter to the realism of gradually getting wounded in combat.

    In effect, I think in an adventure-centric MMO it’s probably a system that isn’t going to provide enough fun to justify the overhead. It’s novel, but wouldn’t add much to an EQ/WoW style game.

    Comment by Moorgard — 17 March, 2007 @ 9:10 PM

  6. Or keep on going through the WoW design and replace mana with rage for all players. Then make rage gain proportional to damage taken and you might have a wierd version of flow control.

    Comment by Wolfe — 19 March, 2007 @ 12:59 AM

  7. From the description I’m going to infer that a complete game redesign around this concept is out of the question. To best address this I’m going to start with the bonus, because if we don’t identify that, it would somewhat null any implementation.

    I’m sure players would give any number of reasons, but here are my two underlying theories on what players are actually looking for:

    Increased tactical options: While the system does not guarantee this I see this as the prevailing valid argument for adding it.

    Increased narrative power: there is something to be said for making the players struggle seem more valiant even if it is only in a superficial way.

    I’m going to cheat and separate wounds from health, and not allow healing to affect wounds. Further outside of combat wounds are regenerated fairly quickly, and wounds will never kill you. As the feature is effectively tagged on, no abilities specifically wound the player, rather if the player takes damage equal to their base level health their “wound meter” would be filled. For narrative the best option is to give visual feedback of the player.

    Now the actual implementation of loss of effectiveness is where we can get creative, the obvious one is to just cut DPS based on your wound level, while that is certainly a sound way to deal with it, that seems a tad bit uncreative. I’d instead suggest that the wound meter have certain thresholds, when each of these thresholds is reached the last used ability becomes unusable until the wound meter is reduced below that threshold. The number of thresholds will never exceed the number of abilities of the character. Perhaps it would cause heads to explode with it’s quirkiness, but it does have the advantage of creating some interesting tactical possibilities.

    To keep things from becoming a turtling match, wounds are regenerated during combat albeit at a relatively slow pace, but a pace that requires players to remain engaged in combat to maintain a wound level on their enemy. I’m sure there are some emergent issues that would come up with balancing, and hopefully, the ability to edit the number and frequency of thresholds per class would handle most of it, but generally I think it would become fairly difficult to find a dominant strategy.

    Anyway, bit of insanity I suppose, probably put too much thought into it.

    Comment by CPinard — 19 March, 2007 @ 4:22 AM

  8. The problem, as always, is that when players ask for something they don’t bother to think of the “entire” system, just the part they want changed. Having player’s performance affected by wounds (fight slower, fight weaker, etc) must be balanced with something to offset the advantage and to keep it from becoming “he who lands the first blow wins”.

    One way to do this would be to also implement a type of “heroic opportunity” system, or even something as simple as “the more hit points you lose, the greater your chance to critically hit your opponent.” The justification for this is “you are, in fact, a hero, and in books and movies, wounded heroes win all the time”. The end result should be that hurting your opponent gives you a consistant advantage, but the risk is that the closer he is to death the more desperate he becomes giving him an unpredictable advantage.

    Yeah, when it doesn’t work, players will curse it… but when it does it will create stories they’ll tell their friends.

    Comment by Jason — 19 March, 2007 @ 5:00 AM

  9. Well my idea would a Bushido Blade kind of system but with shields
    and the ability to equip body-part specific armor that slows you down
    if you wear too much of it (like a nice big helmet if you want to avoid
    getting head-shotted all the time).

    Would probably be untenable due to lag, but I can dream…

    Comment by Daztur — 19 March, 2007 @ 8:57 AM

  10. You could complicate the idea to create a solution. Instead of a simple “combat effectiveness decreases as hit points decline” system, a system that includes the targeting of specific body parts and/or elements would allow injuries to take the form of disabling specific parts/elements. This doesn’t require FPS-style targeting, just the assignment of particular attacks to corresponding body parts. Then you correlate particular skills of the attacker with particular body parts. This way, the result of injuries is tactical limitation, rather than the player’s ability to deal damage, defend or otherwise respond adequately to the specific situation.

    You use skills that eventually disable my right arm. That means I can no longer use skills A, B, and C, which correspond with that arm. I can, however, continue to use skills D through L. I can still defeat you, but it requires a change in tactics. Perhaps I’ve lost my ability to strike crippling blows and damage-over-time attacks, so I must rely on debuffs and direct damage attacks.

    Of course, as is probably evident, the complications necessary for this sort of thing to work go on and on. And it would complicate the gamer’s side, making the MMO feel more like an RTS than [i]Everquest[/i] or [i]Warcraft[/i], thereby excluding low-input gamers from the target audience.

    I don’t think it’s so much a problem for gamers when the enemy is just below half health, but it stretches believability (necessary even in the wildest of fiction) when an enemy with only a sliver of health is still running around and dealing huge damage in the same ways as before.

    The key to believability here is more in animation and movement than in combat effectiveness. Players have trouble taking seriously a creature that moves quickly, attacks ferociously, and then just plops down dead and still. It’s the apparent lack of strain in the enemy’s dying moments that developers should address. Even if it’s dealing decent damage at the end, a dying creature must give an [i]appearance[/i] of being near death.

    Comment by Aaron — 19 March, 2007 @ 11:34 AM

  11. I was about to post this, but Wolfe and Jason beat me to it…

    1. Wounded characters get access to more powerful (one shot?) special attacks, even though their conventional attacks become weaker. The exact mechanics could be class-dependent – warriors get a fury bar or berserk mode, mages have some blood-related ability, etc. Tuned properly, this would negate the positive-feedback effects of striking first. Of course, the special attack has a chance to fail (the opponent could potentially do something to increase this chance – buffs which improve shield-blocking or dodging come to mind), and even if it connects, the opponent may have a chance to counter with their own special.

    2. Certain classes have a “shield” buff (I’m thinking paladins or some mages here) which protects an ally from a certain amount of damage. Dropping a shield on a low-health ally would allow them to use their special attacks with less risk (not zero, as the buff could be dispelled or overpowered). This would reduce the situation where everyone on the other team (assuming PvP party-based combat) dog piles on one guy to kill him before he can use the special attack.

    Comment by n.n. — 19 March, 2007 @ 6:47 PM

  12. One of the biggest problems in PvP combat is the instant death of the individuals that get the opposing team’s focused fire. Combat effectiveness reduction correlated with damage taken can retard the focus fire problem.

    Have the effectiveness fall sharply at first, making the first hits important, after that the effectiveness doesn’t drop much, entreating the attackers to switch to a fresh target hoping to throw them down that initially steep curve. A little like a hyperbola.

    Throw that in with some collision scheme and I’d pay to just see it.

    Players sit until full health before every battle I don’t see how a damage penalty would increase downtime.

    The who strikes first thing… If two evenly matched players are fighting it’s a coin flip anyway. The trick is how do you go about getting (or preventing) that first strike?? :)

    Comment by Ketzup — 19 March, 2007 @ 11:08 PM

  13. I’m thinking along the line of players being wounded in battle too, and avoiding simple respawning as much as possible. Heroes are tough, and the bodies ability to ignore wounds until after combat is pretty impressive.

    The lower the level, the briefer the effects. This means that higher level characters may have semi permanent wounds, and affect a limp etc, and may have to go on a quest to find a certain healer, gather certain components etc to repair their character. It should affect them, after all they just took on an Awesome Being, and lost. This should always be the preferred method rather than perma death, and should leave a nice scar, so they can tell about the time they almost beat a lvl 10000 archmage with their level 100 fighter…. “But I almost got him!”, or “You should have seen the other guy!” when asked why they are limping etc.

    It gives them incentive not to throw their character into harms way, but provides an out. Interesting scars help them build their characters image too, and opens the way to other quests should they want them removed.

    Comment by Mark — 20 March, 2007 @ 9:00 AM

  14. There are plenty of ways you could handicap a player based on how much damage they’ve taken, or other factors, like being poisoned and such. Lets just talk about damage taken exclusively. Damage taken could mirror damage dealt. If you’ve lost 50% of your health, you could deal 50% of regular damage. Another factor could be movement, the more damage you take, the slower your character move. Eventually perhaps the character could stumble every few steps. Another effect could be limitations of ability. For example, a Warrior couldn’t use bash past a certain health point. The advantages to this system are several. One, its more realistic. Two, it promotes a careful play style with effective tactics?

    Cons. As Stephen points out, a big disadvantge to implementing a damage system is that it possibly handicaps the losing player to the point where its impossible for them to win. Another flaw is (assuming we are working off a basic MMO design) that high-damage, faster characters will have an intrisinc advantage over other characters. Rogues would automatically be better characters.

    Comment by Finn Haverkamp — 20 March, 2007 @ 12:51 PM

  15. I would have thought the main reason for requesting this change is to reduce the variance (often referred to by players as “randomness”) in battles. If you’re badly hurt but your opponent is on just a sliver, you don’t want your opponent to fluke just three lucky hits against your three unlucky misses and beat you. And you don’t want to be lulled into thinking you can take that orange-con just because you got a few hits in at the start. Once you start losing, the writing should be on the wall, to give you the chance to escape. That lingering chance of success in full efficiency models has lead to plenty of my own deaths, I’ll freely admit :)

    Another interesting thought is what sometimes happens in PvP combat. When jumped by a superior force, sometimes the victims will simply sit on the floor rather than fight a losing battle. I think that is because then it’s not “their fault” that they died: since they didn’t compete, they didn’t “lose” properly. Similarly, if you lost efficiency with health, you could justly claim that you had already “lost” the fight well before the final blow. There’s also perhaps some comfort in losing ‘cleanly’ rather than getting into an embarrassing series of fumbles and misses at the end of an encounter.

    Comment by Mox — 22 March, 2007 @ 6:30 AM

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


June 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