Psychochild's Blog

A developer's musings on game development and writing.

29 June, 2010

Is a death penalty appropriate?
Filed under: — Psychochild @ 2:48 PM

There’s been some discussion about death penalties lately, and I figured it was a good opportunity to discuss some issues with death penalties in MMOs, and penalties in games in general from a game design point of view. So, let’s take a look at this topic and see how many flamewars we can fan!

Identifying the problem

This round of discussion seemed to have been kicked off by the cheery Wolfshead in a recent post where he gets cranky about light death penalty, going so far as to point to light death penalties as one of the conveniences that replaced risk and therefore has emasculated MMOs.

Now, I will say that I think there is something to his core argument, since I’ve said that risk is a necessary component in our games. But, is a death penalty really an appropriate way to add risk?

Some people agree

Gordon of We Fly Spitfires says that he agrees that MMOs need harsh death penalties. He explains that playing EVE Online has given him something he hasn’t felt in a while: fear. And that fear of losing something of value has enhanced the game for him. He mentions that he remembers every death he’s suffered in EVE, mostly because it’s can be a fairly crushing loss if you risk too much.

So, for some people, risk does add more excitement to the game. For an experienced MMO player like Gordon, having some risk does enhance gameplay by making victory that much sweeter. Even when bad things happen and he’s penalized, a player like Gordon may viewsit as a story to share and a lesson to learn for next time.

A dissenter appears!

Larísa, the charming bartender at the Pink Pigtail Inn, says that death penalties really aren’t necessary. To quote the title of her post, “the Defias were scary enough” to her, no need to add more risk to enhance that.

The problem, as I point out in the comments, is that there are a few different concepts being confused here. To quote from a comment I left:

1. Risk is what Wolfshead’s article is arguably about. He’s arguing that most modern MMOs lack as much challenge because there’s less consequences for your actions. Dying is the canonical example: it used to be that dying put your whole character at risk. The original MUD had it so that if you died, the character was deleted. In EQ1 you could lose all your stuff. In most current MMOs (including WoW), you run back (or, better, wait for the healer to run back) and spend a bit more of your nigh endless gold supply that night.

2. Difficulty is what you seem to think Wolfshead was arguing for, making people work harder for their epics. Difficulty is different than risk. Difficulty must means it is harder to succeed, whereas risk means that failure carries a steeper cost.

3. Newness is what you’re talking about at the end of your post. Experiencing something for the first time is mind-blowing. The first time I was on a text game, playing in the United States and chatting with someone in England, it was amazing. Getting online and seeing other characters in graphics. A lot of people would like to recapture that feeling, but it’s almost impossible to do so.

Larísa’s fear seems to be that this is just about adding difficulty instead of risk. She mentions that she had moments of fear while playing the supposedly emasculated WoW: the Defias bandits in an early zone chased her down in the newbie area. This was added to her unfamiliarity with the game and made for a very trying first experience. Therefore, she reasons that some people might be merely remembering their initial experiences and wanting to recapture those fleeting memories when they were inexperienced.

It’s also important to know that Larísa is a less experienced MMO player. She just recently took a few tentative steps into LotRO as her second MMO. I suspect that risks like death penalties are definitely something that more experienced players want. Always make sure your design is appropriate to the audience.

What’s the design problem?

So, let’s dig a bit deeper and figure out what the design problem is surrounding a lack of death penalties. Elder Game has a great summary of death penalties in MMOs and how they affect players. The belief is that harsh death penalties will encourage grouping over soloing, provide more challenge, and also reduce the willingness of players to try something different.

But, let’s take a step back and look at the larger issues. Some players will chafe at the idea of being punished in a game; aren’t games supposed to be fun and carefree? Why does there have to be punishments in games? As I said before, though, risk can enhance the game because the troughs make the peaks seem that much higher. Winning a rare drop is much more exciting if you had to endure a lot of risk compared to simply putting in sufficient time.

The other important thing to note here is that punishment (and negative reinforcement) are great tools for helping people to learn. Punishment is a negative consequence for a undesired behavior, while negative reinforcement is avoiding a negative consequence through a desired activity (similar, but different). Only using positive reinforcement to help players learn tends to lead to diminishing returns, so these can be important tools.

Finally, I think that risk adds opportunities for heroism. Andrew over at Systemic Babble wrote about the story of a brave Canadian soldier who ended up throwing himself on a grenade to save his fellow soldiers. As I mentioned in a comment, we don’t give players opportunities to do this in our games. The equivalent in a game, say that a DPS character grabs aggro off of a healer in a raid, is not necessarily looked on as heroic. Plus, the DPS character will fall behind on the meters, meaning that he or she needs to lrn2play noob. But, seriously, I think giving players the opportunity to do heroic things would not only make the games more fun. It would also give people a real appreciation for what it takes to make such a sacrifice, more than the intellectual understanding we might have.

What’s the design solution?

So, what’s a good design solution to the problem. That’s a Terrible Idea has an interesting proposal: don’t have death. Instead of dying (or “being demoralized” in LotRO or whatever substitute for death you have), harmful blows reduce your combat effectiveness. This is potentially interesting, because there’s still a penalty there: eventually you’ll probably become completely ineffective at fighting to the point that you have to retreat and regroup. No silly “resurrections” at graveyards or whatever to “cheat death”. I think it adds a level of storytelling in that a “raid wipe” doesn’t have to be a complete failure, you could have the story about how most of the group gets knocked around, but then the group retreats, regroups quick, and springs on the encounter again before the enemies can fully recover. Dare we mention that this could also be an interesting opportunity for optional permadeath situations?

Let me point out a system I proposed in a comment to Larísa’s post I linked above:

Upon entering a raid you get X soul points (depending on the difficulty of the raid, let’s say up to 5). As you die, you lose a soul point. Once a boss is downed, any soul points are converted to favor points and a new batch of soul points are awarded.

What do favor points do? They increase your chance to find rare materials to make raid items (such items for flasks, etc.) Maybe also a small bonus to raid cash rewards and income from daily quests. The goal is that raiders would have to spend less time farming to get raid consumables the better they do.

Points might decay over time to encourage people to stay active. The exact bonus and maximum number of points would need to be balanced out. There would also need to be provision to prevent people from adding a new raider right as the boss is going to die in order to have the one “farming alt” that has the max bonus without the risk. Or, doing an “easy” raid to pad the point totals.

Of course, this still waters down the risk. But, again, it’s something where it adds a bit of risk to dying (requiring more farming).

How does this system address risk? The player has a potential reward that comes from exceptional performance in a raid (not dying). The player has a strong motivation to perform well, but the consequences for failing aren’t dire. The players are encouraged to learn their roles in a raid in order to help the group succeed. This doesn’t not necessarily add any more difficulty to the encounters. Yes, it can be disappointing to lose the opportunity to get a bonus, but failing to get the bonus leads to players experiencing the same thing as people who aren’t raiding.

I’ll admit, though, it’s not a great design. It’s something I threw together in a few minutes to provide a point of discussion. Perhaps you can do better. What’s a good way to increase risk without increasing difficulty? How can we manage player expectations to let them know that taking some risk might actually enhance the game?

« Previous Post:


  1. I think we should just follow the logical progression and reward players that die.

    Think about it. If you can only get experience from dying, you’ll stop grinding and seek actual challenges. No more hunting for healers, no more minmaxing.

    Comment by Zaratustra — 29 June, 2010 @ 3:13 PM

  2. While you didn’t seem to like your idea much, I think it has potential. For things like padding the raid, points could be based on the median rather than average, making the outlier have no effect, or less chance of an effect. Depending on how complex the devs want to make the math, they could even set it up to remove outliers, both high and low, to ensure that there’s no benefit to padding, nor any loss from bringing in a low person.

    @Zaratustra: Couldn’t players just die to trivial content?

    Comment by Klepsacovic — 29 June, 2010 @ 3:22 PM

  3. Hm… I may be charming :) but apparently I fail and expressing myself clearly.

    You see, I don’t really have any problems with a higher level of difficulty in the game! It’s just that I don’t need more punishment for failing to make it exciting. I may be a newbie, but I’m pretty hardcore in my mindset I’d dare say in the sense that I’ve got well developed winner/killer instincts or whatever you call it in English. If I fail, the failure in itself is a punishment. I’m totally motivated to avoid failure, to improve my gameplay, to go for a win, without adding more harsh death penalties. Such things would rather kill my enjoyment of the game than increase it – and the bad thing about it is that it might deter me from daring to take risks, to challenge myself with more difficult things, to aim higher and expand my boundaries as a player.

    A more harsh penalty for failing would probably result in me playing more like a coward, sticking to green safe quests rather than daring to try the red ones. I’m afraid that would result in a more boring game experience and in worst case – that beginners will be scared away. At least in my case I’m not scared of by added difficulty. Not at all. I’ll just climb the learning curve. I know there is barely anything I can’t learn to master if I really set my mind to it and invest a lot of time into it. But increased risk – that might result in just an endless grind to get back to where I was before I failed…. No. That sounds just too depressing.

    About your suggested system… Raiding is team work. Period. Sure, there are guilds that have punishments for mistakes by individuals. You know… minus x dkp. But it’s not something I’d advocate. The system you describe, if I’ve understood it clearly, could easily turn into a name-and-shame feature, depending on how it’s formed. Even if a death only affects the individual and not the entire raid, it can still lead to conflicts. “Why didn’t you heal me up, now I’ve lost my soul bonus thanks to you, sucker”….

    Unfortunately many players take every chance they can to behave like a jerk. I think you should be careful not to provide them with more tools to do that.

    I don’t dismiss it entirely, but I think it would need some serious consideration. Or maybe it’s just a mechanism that would work better for a smaller game, directed not to the broad masses, but to very hardcore and experienced players who love to get a good spanking every now and then.

    Comment by Larísa — 29 June, 2010 @ 3:30 PM

  4. As I’ve noted elsewhere, death penalties are simply a time sink in my eyes, and when time costs money, as in a sub game, increasing the penalty for failure is adding insult to injury.

    I like challenge, but only because I risk failure. Failure to succeed is penalty enough for me, and impetus to try again. Corpse runs or XP/gear loss are just something I have to *do over* in order to try the risky stuff again until I get it right. Too much of that, and I’m not interested in the challenging activity again… not because it’s inherently any less interesting *in itself*, but because getting back to try again is a waste of time. It’s not worth grinding through the prelims again and again; I’ve mastered that content, and I learn nothing from doing it over and over. It’s no fun, and it’s a waste of time and maybe money.

    I have much the same reaction to RPGs that have huge chains of boss fights with no save points or long treks between save points. It’s not the risk *or* the challenge that I dislike, it’s the grind of Do It Again, Stupid game design. (Shamus of Twenty Sided rightfully blasted that sort of game design a while back.)

    Comment by Tesh — 29 June, 2010 @ 3:53 PM

  5. Ah, and blast these double posts… sorry, I’ve been pretty disjointed today.

    I’ve also argued before that making death itself rewarding or interesting to *play* might be a good angle to explore. Perhaps a fallen character in a raid can make themselves useful in ghost form for a time with some sort of mechanic that affects the “living” world but that is exclusive to ghosts, or by providing important intel only accessible to ghosts who can pass through walls.

    I imagine raid encounters where one player’s death isn’t a domino that winds up wiping the raid group, but rather, just another tactical situation. The fallen player can still contribute, just differently. Perhaps they even get different rewards if the group wins, say, some sort of spectral loot or whatever.

    Perhaps there could even be raid encounters where dancing between life and death is critical to success. *That* might be a fun tangent to explore.

    In short, as I’ve argued before (as has Wiqd, notably), I’d rather death be a mechanic, not a penalty; something potentially fun in itself, that allows players to make choices other than “do it again or quit”. I’m decidedly not a fan of game design that punishes players, especially where the punishment is just a time sink.

    Comment by Tesh — 29 June, 2010 @ 3:59 PM

  6. Oh you express it much more eloquant Tesh! But I agree! About the ghost thing: we DO have something like that in WoW now, the raise-as-a-ghoul feature that came with the Deathknights. Sadly enough it isn’t used very much, probably because it isn’t evident to the players how they’re supposed to manage this sudden new class/vehicle. Finally the DK:s gave up on bothering about it. I’ve only been rezzed as a ghoul 7 times since Wrath came out and that’s NOT much considering how often I die! I tried to help out a bit againt the cluelessness recently, asking for blog posts on how-to-ghoul. Four different bloggers came up with brilliant guides on the subject.

    I’d love to see a development in this direction. I think the ideas you suggested here were very interesting Tesh.

    Comment by Larísa — 29 June, 2010 @ 4:14 PM

  7. Zaratustra wrote:
    I think we should just follow the logical progression and reward players that die.

    This was discussed for some text games. Of course, usually there was a limited number of times you could die. The point being that death would be very educational, and players could either take the quick road to power (die a few times to get a boost) or save those limited lives for when they really needed them.

    Larísa wrote:
    I may be a newbie, but I’m pretty hardcore in my mindset

    You’ll note I was careful not to use the term “hardcore”. Given that you care enough about online gaming to blog about it, I certainly would not accuse you of not being hardcore. :)

    But, you are less experienced. That probably makes you less jaded and easily bored as people like Gordon or myself. Nothing wrong with that! But, let’s talk in 5 more years about death penalties if you are still interested in MMOs. I would bet that your attitude might be a bit different.

    Such things would rather kill my enjoyment of the game than increase it – and the bad thing about it is that it might deter me from daring to take risks, to challenge myself with more difficult things, to aim higher and expand my boundaries as a player.

    I agree. As the Elder Game post mentioned, harsher death penalties do tend to hinder exploration and trying new things. Again, I think it’s a specific audience that is looking for excitement in the form of harsher penalties. You don’t seem to be part of that audience. Luckily, most of the current MMOs cater to you. :)

    Unfortunately many players take every chance they can to behave like a jerk. I think you should be careful not to provide them with more tools to do that.

    As Gearscore proves, people will take anything and be jerks about it. Should WoW remove gear from the game since some people are obnoxious about gearscores? At some point, you just have to make a good game and expect people won’t be total jerks about playing it.

    But, you are right in that raids would have to be changed to adapt to a change like this. Be careful in assuming that design proposals are “keep everything from WoW but change this one thing.”

    Tesh wrote:
    As I’ve noted elsewhere, death penalties are simply a time sink in my eyes…

    Gaming is a timesink for most people, if you get right down to it. Why do we fire up games? As has been pointed out, we could be out feeding the hungry or trying to find a cure for cancer instead of having my little video dude kill another little video dude. Anyway, this joke as been done already: ProgessQuest is the MMO with all the timesinks taken out. Although it’s still around after all these years….

    I suspect that you are not in the target audience that wants more risk in games. Again, you have a whole lot of MMOs to choose from (assuming you’re willing to pay a subscription fee…. ;) People who do want more risk don’t have that luxury.

    In short, as I’ve argued before (as has Wiqd, notably), I’d rather death be a mechanic, not a penalty….

    At that point, I’d rather not call it “death”, then. We already have narrative dissonance when an NPC worries about someone else “dying” in these worlds. Making “death” a mechanic would further strain this metaphor, I think, so time to think of a new one.

    Interesting discussions, though. Keep them up! :)

    Comment by Psychochild — 29 June, 2010 @ 5:38 PM

  8. Oh, sure, I’m all for making some games (or better, elements of games) with higher risk/reward for those who want it. I just won’t advocate making that the only option or trying to force it on everyone. Or if you *do* make a pure hardcore risk game, don’t be surprised if the mainstream doesn’t like it and shower you with money. ;)

    Was it you or Wiqd who suggested permadeath raids as *options*, and heroic quests as something a player could opt to do, but not the default setting…

    Comment by Tesh — 29 June, 2010 @ 6:10 PM

  9. @Zaratustra:
    There is actually a title in Guild Wars that can only be achieved by “Death Leveling”, in which the player intentionally dies in order to give mobs exp so the mobs can gain levels and continue to award the player exp beyond the point at which they would normally grey out. Progress? :)

    The problem with the payout for not dying is that players very quickly begin to look at such “extra” bonuses as part of their pay when evaluating what they are and aren’t willing to do. This very quickly translates into risk aversion in the form of unwillingness to group with players who would put “your” rewards at risk.

    (For example, the two raid emblems from WoW’s random daily dungeon can’t even be lost by a failed pug, and the lockout time for refusing to run a dungeon is longer than the dungeon run typically takes, but players still bail out or vote-kick on a whim.)

    Grouping with other players presents the risk that the player will be killed through no fault of their own, but rather the fault of the group. I don’t think that this is a good place to put negative reinforcement (such as the death penalty) unless you’re making an educational game for young children about why they should not talk to strangers.

    Comment by Green Armadillo — 29 June, 2010 @ 6:51 PM

  10. Up until Module 5, DDO had XP penalty when you died.

    That had a few negative consequences:
    1. Discouraging people from challenging themselves. People often avoided running anything difficult until they reached cap because it other could otherwise cost them a lot.
    2. It was frustrating (loss aversion).
    3. Resulted in an inexistent death penalty at cap.
    4. Caused a lot of drama in groups if the healer failed at his task or if someone just played horribly in a way that caused everyone (or a lot of people) to die of a very painful death.

    The first three problems were due to the type of death penalty that was used but the last one is common to any type of penalty that could be labeled “harsh.” Think about it: if the penalty was in any way painful (ie more than a temporary debuff), how enjoyable would it be to be an healer?

    In the case of DDo, it turned out that the answer was “not very.” It was not uncommon for the rest of the group to hate the cleric because “he sucked.”

    Comment by Borror0 — 29 June, 2010 @ 9:21 PM

  11. Let me explain why I like permadeath.

    Firstly I’m a voracious reader of heroic fiction. David Gemmell, Robert Howard etc. It’s always implied that death is possible even if the hero always manages to evade it. I don’t see Conan as a hero in a non-permadeath world, I see Conan as a hero who manages to evade death by luck judgment and courage.

    Secondly I cut my gaming teeth on games like Chess and Monopoly. If you lose it’s complete, you don’t semi-lose and get another go.

    I cut my RPG teeth on 70s and 80s table top. Not only was permadeath possible but often dying was woven in as part of the entertainment. In Call of Cthulhu you spent a considerable time scared. In Paranoia you were actively hunted by a malevolent GM.

    With computer games I come from the wargamer side rather than the FPS side. In wargames units die (and I see my MMO characters as units in my faction’s army).

    One of my finest computer gaming experiences was Diablo 2, which I played hardcore (permadeath mode) for about 2 years. It was enjoyable because of the associations I formed with other players to cooperate against the harshness. People would write obituaries on our guild forums and generally felt very close.

    So for me I don’t see harsh death penalties as punishment. I see them as not fudging the dice rolls in favour of the player.

    Comment by Stabs — 29 June, 2010 @ 10:21 PM

  12. The problem with death penalties, is that they only add more grind. The only consequence is that you have to re-do stuff you’ve already done to get back to the same point. I’ve yet to see an approach which is different (the one you propose is interesting, using positive feedback instead of negative, in an improvement on what what WoW has done with “immortal” achievements). Still, it has again a grindy look to it (die this time, restart dungeon and do the same until you don’t die and get the shiny item). Old muds could get away with things like losing your stuff, because acquiring it was infinitely easier than the long progression which is happening today.

    BTW I agree with the “failing is already a penalty”, and I find permadeath to be a complete joke. Die and all you have to do is re-level you character. Wonderful….. If you’re serious about permadeath do it the hardcore way: if you die you delete your accound and the game once and for all. :)

    Comment by Ishark — 30 June, 2010 @ 12:43 AM

  13. Tesh wrote:
    Or if you *do* make a pure hardcore risk game, don’t be surprised if the mainstream doesn’t like it and shower you with money. ;)

    You mean like EVE Online as Gordon describes it in his post? ;)

    And since when have I been concerned about mainstream acceptance. Have you seen the game I worked on for over a decade? :P

    Green Armadillo wrote:
    The problem with the payout for not dying is that players very quickly begin to look at such “extra” bonuses as part of their pay when evaluating what they are and aren’t willing to do.

    Well, yes. By definition, risk is about potentially losing something. You can present it as losing something the player already has or losing a potential bonus they might earn. Yes, some people are going to see past the clever wording, but WoW has shown that people will put up with a missed advantage more than having something taken away. Go look up discussions about the rested xp bonus in Beta to see that in action.

    Borror0 wrote:
    Think about it: if the penalty was in any way painful (ie more than a temporary debuff), how enjoyable would it be to be an healer?

    The penalty is already pretty severe, though. -10% xp from someone dying. More if they have to re-enter the instance. If someone wants to, they can still blame the healer for a loss of xp. In fact, one recent group I had outside of my normal guild resulted in the leader dropping because someone zerged ahead. People seem to blame the zergs more than the healer these days; not sure that an xp penalty is responsible for that.

    I think there’s definitely a shift in perception, as I mentioned above. The current system takes away a bit of a reward once instead of potentially multiple times.

    Stabs wrote:
    It was enjoyable because of the associations I formed with other players to cooperate against the harshness.

    I think this is something that a lot of the people who only really know WoW don’t understand: adversity can forge strong bonds between people. Not to say that harsh penalties and tremendous risk eliminates jerks, but it’s a lot easier to get by in a harsh situation if you have other people. Being a jerk is a great way to become isolated and ostracized. Decent people will tend to stop and take more note to help others out, expecting reciprocation in the future.

    Lots to think about, I guess. Nobody said game design was easy. :)

    Comment by Psychochild — 30 June, 2010 @ 1:51 AM

  14. Here’s what I reckon may be a reasonable death penalty solution.

    #1 no real penalty for death, it is just the standard inconvenience a mild slowdown IE tone this part right down.

    #2 a “death penalty” cost is applied upon entering a zone/dungeon whatever, a cost that can be quite harsh. This also helps with creating consumables in the economy.

    Now, you must play to recoup this investment you have made, the inconvenience of dying is just a mild inconvenience.

    You start off having already lost and you keep playing to turn that loss into a win. Or you give up, admit this area is too hard and stop trying. Even recouping a small amount of your initial loss will still seem like victory.

    It’s risk/reward but still seems close to the standard model. Shifting the fail penalty to the beginning is just a mental trick to fool the loss adverse humans.

    I’ve seen this partially applied but not as strongly as I’d like to try :)

    Happiness for hardcore and carebears?

    Who knows, I figure it may be worth a try though.

    Comment by Kriss — 30 June, 2010 @ 4:53 AM

  15. Death penalties aren’t bad.

    Time penalties are bad.

    The trouble is that death penalties are usually implemented as time penalties. Even when it is “loss of equipment” it often actually means “You must now grind for 5 hours to replace the equipment”

    The problem, IMHO, is that MMORPGs are built around time sinks. Thus, “having” to regain XP is seen as a time sink. Having to re-equip is seen as a timesink. Which is bizarre – gaining XP should be done by playing the game, and playing the game should be fun, so thus not a time sink! The trouble is if your gameplay is treadmill interspersed by fun raids, anything that sends people back to the treadmill is seen as punishment.

    In Ultima Online, a game whose effective death penalty was “lose everything your carrying”, a friend used to quip: “If you don’t die once a day, you aren’t trying hard enough!” Of course, this was spoken at a certain level of advancement where losing your inventory was almost a feature as you’d clear out all the crap you had accumulated and could just grab a new suit of armour from your bank.

    I disagree with Psychochild that negative reinforcement is needed. Positive reinforcement doesn’t have diminishing returns, or casinos would be out of business by now. The trick is to turn the external positive reinforcement into internal positive reinforcement. Before you get to diminishing returns, you need to turn the means into the end. So, rather than having a bigger and bigger sword for the player to strive for, you want the player to refocus their goal away from the prize and onto the journey. Thus, by the time they hit your content cap, they don’t care about the items anymore. They care about the game.

    I like Kriss’ idea: it really addresses the true problem, that we are asking the wrong question about death penalties.

    Another solution is to ask seriously why we are so scared about death in a video game. Why not die and roll another character? The usual Role Playing argument hardly holds for most players, they see it as a mere puppet anyways. Playing Brask-352 wouldn’t affect their experience negatively. Losing hard won items and levels? Aren’t we looking at that the wrong way? Isn’t it instead getting the chance to win new items and levels? Except… In WoW, you are looking at 2-3 months of play to get back to Level 80 so you can play with your friends again. Which brings us back to the treadmill. So long as that hangs over peoples head as a threat, it is hard to apply any meaningful setback on players, as they will always react by grinding, and hence boring themselves to death.

    Comment by Brask Mumei — 30 June, 2010 @ 7:19 AM

  16. “And since when have I been concerned about mainstream acceptance. Have you seen the game I worked on for over a decade? :P”

    *chuckle* I know, I know. That was more of a generic “you” there, with Darkfall and Mortal Online as poster children. Great little games for some, but hardly commercial juggernauts. Not that such is a bad thing, of course, so long as expectations are tempered. EVE is a bit of an aberration, as you lose *stuff*, but generally you can be up and *playing* in very short order again, close to what you were doing before. Permadeath on a level capped WoW character would be comparatively devastating, as getting back to where you were would be an onerous task. Yes, playing the game itself should be fun, and gear deemphasized, but for some, only the endgame is challenging or fun. The leveling grind to qualify for the fun part is just a time sink treadmill for them. (It’s a structural thing; if your game is more like EVE, or even Darkfall, and a fresh character can be up and running, doing fun and useful stuff in very short order, even something as harsh as permadeath isn’t a big deal, since the coin of the realm isn’t progress or time, it’s play and player skill.)

    …maybe permadeath for level capped characters in WoW wouldn’t be so bad if the player could instantly reroll a minimally equipped level capped new character and be back helping their guild in short order?

    A further thought on time sinks and wasting time, though: yes, you could argue that gaming on the whole is a waste of time, but that’s missing my point. When I have to do the same thing over and over, grinding through those treadmills to qualify for the fun challenging bits, *that* is a waste of time. It’s like taking a math class, and if you “die” on a test and only get 80%, you have to go home, walk back to class, and redo all of the chapter’s homework again before you retake the test. Never mind that the relevant part you really need to learn is that one section in the middle of the chapter, and that actual teaching from the teacher isn’t forthcoming, you have to waste time and redo everything you’re already proven mastery over, as punishment for “dying”, and you don’t even get help on what killed you.

    There are a couple of problems that highlights, then: One, sometimes, dying in a game is something you have no idea how to counter or avoid. If the game doesn’t teach you what you did wrong, you’ll be spending a lot of trial and error time dealing with bad designer feedback. Two, when punishment is more a time sink (XP loss, corpse runs, that sort of thing) rather than a teaching tool, it’s not serving a good purpose.

    Comment by Tesh — 30 June, 2010 @ 8:47 AM

  17. Ah, one of the evergreen topics. :)

    I think it’s worth noting (as Tesh and Stabs alluded to) that the question of character death only matters if we’re talking about games where you play a character. In other words, RPGs.

    A roleplaying game is substantively different from a multiplayer shooter like Team Fortress 2 or online Quake/Halo/Call of Duty/etc. In a multiplayer shooter your character is just a kind of vehicle — there’s no character development; “you” are not distinguishable from any other character (in the same role) or any previous or future incarnation. When your character dies, nothing personal *to that character* is lost; you just respawn and jump back into the action.

    In an RPG, though, there is something to lose. Whether we’re talking very personal story-oriented RPGs or highly mechanistic and numbers-driven RPGs (like today’s MMORPGs), the usual design is that characters accrue unique attributes over time. You may look at those attributes as mere numbers (“my STR is 15 and yours is 8; I win”) or as elements of characterizing a person (“I got this scar putting down a guard in the Quaagar uprising”), but because they’re things you had to earn for your character they become things you don’t want to lose. That’s where the motivational element of risk comes from.

    So to make the risk of death generally acceptable in a roleplaying game, you have to either improve the perceived benefit for dying or reduce the perceived cost.

    Improving the perceived benefit for dying is, as noted above, relatively easy to implement as a mechanic but hard to design in a way that makes logical or story-based sense in the gameworld. In a large commercial MMORPG, rewarding players for dying (either numerically or narratively) is pretty much guaranteed to encourage lemming-like behavior that will put everyone else right out of the magic circle. Unless that kind of behavior is something you actually *want* in your game (as in Paranoia), it’s probably best not to give special in-game incentives for getting your character killed.

    In the other direction, reducing the perceived cost of character death in an RPG means eliminating the perceived value of character attributes. That’s why I’ve said before that the only viable way to have meaningful death in an RPG is to eliminate character advancement. Design the game to work like Traveller where everybody starts playing with a fully-developed character (so that the death of a character doesn’t mean losing weeks or months of time spent increasing that character’s attributes), and implement a “bequeathal” system that transfers most (but not all) of the possessions of a dead character to the new character. This allows characters to be unique — they can have distinctive qualities that allow their players to enjoy personalized stories through them — while minimizing the numeric pain of losing that character and allowing the loss to have story-based meaning.

    There are probably other ways to make character death acceptable without making it desirable. This is just the approach I’ve been considering lately; I look forward to reading other ideas here.

    Comment by Bart Stewart — 30 June, 2010 @ 9:18 AM

  18. One of the issues that has always irritated me about this topic is the artificial conflation of “death” and “loss” in these games. In my pen-and-paper RPG games, the party loses fights all the time without “dying”. Having myself commented and lurked and spouted off on this topic a dozen times before, I sometimes wonder if we even realize there can be a difference.

    And isn’t the real reason that this topic comes up time and again (I liked the “evergreen topic” side comment) basically due to this conflation/dichotomy? “Meaningful”/”heroic” death in these games simply can’t exist if the only way to lose a fight is to “die”… especially when the only activity worth taking part in is combat to begin with. “Death” gets dumbed down/nerfed in part _because_ it’s so common under this definition… am I right?

    I am a fan of perma-death. I also agree with Tess that (in most existing games, at least) “Failure to succeed is penalty enough for me, and impetus to try again.” Avoiding the conflation of “death” and “lost a fight” is why I do not think these two statements are not necessarily diametrically opposed.

    So, the $64,000 question… Does that make any sense to anyone but me?

    Comment by DamianoV — 30 June, 2010 @ 9:56 AM

  19. First, a correction… the 2nd-to-last statement in the preceding post has a double negative, I apologize. It should read:

    Avoiding the conflation of “death” and “lost a fight” is why I do not think these two statements are necessarily diametrically opposed.

    A previous round of runs at this topic:… there are a few links that aren’t quite dead yet in there…

    I haven’t seen much discussion of the concept from “That’s a Terrible Idea”, which is one I happen to play with a lot in pen-and-paper, being a RoleMaster fan: the concepts of injury and loss of effectiveness, not to mention (gasp) “unconsciousness”, as opposed to outright “dying”. (See City of Heroes and inspiration-passing for one way to make “unconsciousness” less onerous.)

    I’ve always wondered if that would/could work in an online game… you have to deal with the issues of avoiding “first hit wins”, as well as “perma-gimping” via injury. What other considerations are there that I am missing?

    Comment by DamianoV — 30 June, 2010 @ 10:18 AM

  20. I just stumbled across an interesting FFXIII review that makes an argument against “setback punishment”. It’s nicely relevant to what I’m arguing as the distinction between failure and punishment.

    Comment by Tesh — 30 June, 2010 @ 2:50 PM

  21. Toldain weighs in on this topic (Blogger doens’t seem to do trackbacks properly):

    Another interesting perspective there.

    Comment by Psychochild — 1 July, 2010 @ 12:44 AM

  22. Kriss wrote:
    a “death penalty” cost is applied upon entering a zone/dungeon whatever

    An interesting proposal. I wonder if this wouldn’t amplify a lot of the problems people have worried about here: a decrease in risk-taking to make sure you get your investment back.

    Brask Mumei wrote:
    Death penalties aren’t bad.

    Time penalties are bad.

    Yes, but time is the ulimate coin of the realm in MMOs. Risk requires the loss of something, so anything you lose in a game will require time if you want to get it back. But, I think this is the key here: if you want to get it back. There’s a default assumption in these games that we must pursue perfection in a character. It’s got to be max level, it’s got to have the best gear we can acquire, it must be top of the line. When your gameplay doesn’t enforce that type of character, such as Ultima Online, you have a lot more freedom.

    Positive reinforcement doesn’t have diminishing returns, or casinos would be out of business by now.

    Casinos work primarily off of random reward schedules, not positive reinforcement. But, if you don’t believe me about the diminishing sensitivity to gains, go look at a real psychologist’s perspective.

    Bart Stewart wrote:
    Ah, one of the evergreen topics. :)

    Yeah, but it’s gotten some good discussion going again. :)

    and implement a “bequeathal” system that transfers most (but not all) of the possessions of a dead character to the new character.

    I’ve heard about systems like that… ;)

    DamianoV wrote:
    “Death” gets dumbed down/nerfed in part _because_ it’s so common under this definition… am I right?

    One of the issues that has always irritated me about this topic is the artificial conflation of “death” and “loss” in these games.

    I think this strikes at the heart of the issue. The only real mode for failure in combat-centric games is “death”, especially in scripted encounters you find in WoW raids. That’s one reason why I thought the “you just get weaker” idea from That’s a Terrible Idea was so interesting, because you could add more failure modes. There’s a lot of room for other interesting conditions. Yes, sometimes my character would die in a pen and paper game if something went horribly wrong; that’s the risk I took. But, most of the times I was able to turn around and join the party again in short order and have fun again.

    Some further thoughts.

    Comment by Psychochild — 1 July, 2010 @ 1:28 AM

  23. Thoroughly enjoyed your post, Brian! A lot I want to chat about here but I’ll try to be succinct :)

    Difficultly vs risk is a very interesting thing however I believe that whilst death penalties increase risk, they don’t necessary need to affect the day-to-day difficulty of a game. For instance, soloing and standard groups could be designed in such a way that most encounters will be overcome by an average group with only a low chance of failure. The point of risk though becomes a factor when players decide to partake is more unusual activities, such as certain quests or group or raid encounters. The chance of failing these would be higher and the death penalty severe to sting the players that these activities become very risky. And then of course this risk would allow developers to create nice rewards. The whole thing, if carefully balanced, could create a nice mix of risk and reward, the result being that players can truly overcome difficult odds, bound together and feel like true champions after they’ve done it. Plus, the beauty is that it wouldn’t affect the easy and casual day-to-day activities of normal soloing or groups. Personally, I think the mistake that some developers make in very “hardcore” games is that they try to create risk throughout all areas and levels of game play and enforce huge penalties on every player at every stage… which is just darn frustrating. The idea, for me, should be that players make concious choices to undertake more risky activities because they want the special rewards.

    On another note, we also need to look at why death penalties exist in the first place. Aside from creating a feeling of risk, they’re there to help create immersion by reflecting the normal cycle of real life. Just as MMOs have food and drink and night and day cycles, death is there to help our characters feel more human and for us to bound with them more as a result. However, every mechanic needs to have meaning and can’t just exist for the sake of it otherwise it defeats the point of having them in the first place! For instance, if death becomes something that people don’t even think twice about, why bother having it?

    Comment by We Fly Spitfires — 1 July, 2010 @ 3:11 AM

  24. I think punishment/death/risk and loosing are all a little too tightly bound on average. It’s easy to loose the point of a discussion about one of them because they all connect so easily, but much like you don’t want to confuse the positives and negatives in a bundle of wires, it can be problematic to NOT distinguish between them. Death is really just one way of expressing the point at which somebody out and out looses, Evizaer’s suggestion to remove death is to remove loosing in favor of punishment. The players decide for themselves what a loss is, the game only has levels of punishment to meat out for poor performance. The system you proposed also seems to be solely punishment to me, once in a raid you WILL move at the pace of your team and you WILL take on the challenges they tell you to. Since not running the raid means getting nothing, there isn’t really any risk since they aren’t putting anything on the table, instead they are just loosing rewards as punishment for poor performance.

    Risk is a huge part of the “loosing your stuff” approach to loss, and it’s probably one of the best risk heavy death penalties we have. You have to decide what to wager on your chances of success, and how much you invest in making that success more likely. Therefore when you loose, you’ve lost what was risked, are only punished in the time it takes you to get back on your feet, and have clearly lost at some point. However, when you start to take loss out of the picture, things tend to flow more towards punishment than risk since there isn’t a clear point at which the wager ends and your losses are measured. In this case a way to deal with your proposed system, as an example, might be to alot the players those same points, but to instead have the total that they can “wager” on the current boss. Let’s say they have 10 points, if they invest 3 in the first boss, then they can die 3 times before they are reduced to getting the bosses consolation prize, on the other hand if they keep all three, then they get 3 pieces of/3rd leveled gear from that boss. Now while they are still punished for dieing, they are also risking the chance of better future gear on a bet that this boss will be easier or that they know this fight better.

    If good games are a series of interesting choices, risk is an interesting choice. Loss and punishment are measurements, results, outcomes of those choices but not choices in and of themselves. Difficulty is a subjective measure of how “fair” the punishments in a game are, or how “fair” loosing is. Adding more risk doesn’t necessarily change the amount of loss or punishment, but it can effect how people view them, causing greater risk to be associated with greater difficulty even though the changes aren’t objectively substantial.

    As an aside, I would like to say that players who desire more risk are probably not nostalgic for their first experience in games again. Sure everything seems difficult early on, but if it weren’t for the immense usage of shared tropes it would be that way with any game (in fact, if you wanted to be completely crushed by a game’s learning curve there are always the football manager series). Personally, I think the desire for greater risk is a more refined taste, one that comes with having memorized the similarities so well that only the differences are of any worth to you. Also as a refined taste, it’s personal and generally associated with the head fakes it brings with it, closer, more mature communities in this case, making it’s pull to those who have made those associations stronger than the objective measure of it’s parts.

    Comment by Sara Pickell — 1 July, 2010 @ 4:53 AM

  25. Psychochild misread what I wrote on removing death. I don’t want to punish players at all. I just want to let them do what they want without potentially losing time from dying. I suggested no punishment whatsoever except for the inherent loss of time that results from fighting a battle that isn’t as efficient in terms of rewards for time as some other battle.

    Here is the original text of my suggestion:

    “Make the game so that the player character can’t die. The player can fight indefinitely against any enemy and eventually probably win, but he can’t be killed and forced to respawn. Give the player an ability that allows them to teleport out of battles (or bad places that would usually cause death) at will. Let the player disengage an enemy’s aggro, but then bump the enemy’s health and expendable resources back to where they were before the battle. Build the game around rewarding players for efficiently dispatching with enemies. It’s already this way in effect, why not make it the central issue?”

    Comment by evizaer — 1 July, 2010 @ 7:45 AM

  26. We Fly Spitfires wrote:
    Personally, I think the mistake that some developers make in very “hardcore” games is that they try to create risk throughout all areas and levels of game play and enforce huge penalties on every player at every stage… which is just darn frustrating.

    I think this is fair. It might be what a lot of people who are wary of adding more risk see as a problem, too, where they envision that there will be more risk everywhere in the game.

    In general, I’m a proponent of letting people people make their own choices. Just need to make sure that people don’t choose the overly easy option then get bored because they refuse to take risks. ;)

    Sara Pickell wrote:
    …they can “wager” on the current boss.

    My concern is that this might confuse some people, especially if they have yet another thing to do while setting up for a fight. Perhaps having a “default” value might work. It is an interesting concept, and does introduce some risk of the player’s choosing.

    I think the desire for greater risk is a more refined taste, one that comes with having memorized the similarities so well that only the differences are of any worth to you.

    I think this is fair. As I’ve commented several times now, this is definitely something that seems to appeal more to the experienced players.

    evizaer wrote:
    Psychochild misread what I wrote on removing death. I don’t want to punish players at all.

    I haven’t mentioned anything about punishing. I assumed that even without death there would still be a potential failure condition in that a player might be ineffective in combat; you mentioned about being able to teleport out of battles at will. If there’s no risk of not finishing the battle, what’s the point in being able to teleport out?

    I’d also worry about what I mentioned above: people choosing the easy method then complaining that the game is too boring. “Okay, everyone, turn your autoattack on and wedge a paperclip in the hotkey that does your best damage attack. Come back in an hour to check to see if we won!” I don’t see that as providing more interesting gameplay for players. Perhaps there’s more to your proposal, then?

    Comment by Psychochild — 1 July, 2010 @ 2:02 PM

  27. “Just need to make sure that people don’t choose the overly easy option then get bored because they refuse to take risks.”

    Isn’t that their choice, though? ;)

    I loved the pacing and balance of Chrono Cross, where the “star level” meant you were never overleveled or underleveled for a challenge, and you knew going in that if you botched it, it was your fault for not understanding the system well enough. Oh, and you could run from almost any fight, even most bosses, to regroup and try again. The risk/reward balance and challenge were a bit more even than spiky, but it allowed for much finer tuning and a better sense that failure was the player’s fault for not understanding what they were given, rather than not grinding enough.

    Comment by Tesh — 1 July, 2010 @ 2:41 PM

  28. @Tesh I agree that players should be allowed to take the easy options if they want. For me it’s all about balancing the risk vs reward. If someone wants to take the easy route then that’s all well and dandy but if someone wants to take the risker road then the rewards will be better to compensate.

    Comment by We Fly Spitfires — 1 July, 2010 @ 2:46 PM

  29. N’Gai Croal talks about failure in (single-player) games, which touches on some of the issues here. Failure Is An Option talks about “challenge, difficulty and progression”, particularly in the story-focused game Heavy Rain.

    Tesh wrote:
    Isn’t that their choice, though? ;)

    Yes and no. We’ve called this behavior “bottom-feeding” in the past and we know that if you allow players to bottom-feed, some will see it as the most profitable option and pursue it. They’ll get bored and then blame the game designers for making a “boring game”, ignoring their own choice that lead them to that problem. Being a good designer requires you to guide the player’s experience, even to areas he or she might not normally have chosen. The trick is to make it interesting but not impossible. Allowing the player to ramp up the challenge (possibly for greater reward as Gordon points out) is the real choice the player should make, not the ability to bore themselves to death by bottom-feeding.

    We Fly Spitfires wrote:
    …if someone wants to take the risker road then the rewards will be better to compensate.

    The problem here is if the perception changes so that the better rewards become the only desirable ones. Then people will complain about how they are “forced” to do the harder route to get the “real” rewards.

    Of course, at some point as a designer you have to accept that someone will always complain about something and forge ahead anyway. :)

    Comment by Psychochild — 1 July, 2010 @ 4:16 PM

  30. Just to add food for thought on the issue of non-commerciality or what players will accept.

    In 2000-3 Diablo 2 had far more online players than any MMO. The hardcore community was just under half of the size of the non-permadeath community.

    In other words more people played a permadeath online RPG character development-based game, with associated deaths to lag and ganking, than played any contemporary MMO.

    It IS commercially viable.

    The hurdle is how you provide customer support for a game of such emotional highs and lows, not whether people would come.

    If WoW opened a permadeath server tomorrow it would be so full they’d have to open a second one a week after. Their CS team would need therapy however.

    Comment by Stabs — 1 July, 2010 @ 6:15 PM

  31. Speaking generally about failure in games, what about the point that some people tolerate failure-states better than others?

    In particular, consider rules-based play versus simulation play. Most games today (and definitely most MMORPGs) simply don’t let players fail. This is because the developers assume or know that most of their audience will be folks who consider a game broken — not even a “game,” in fact — if there’s no way to win through sheer bullheaded determination and repetition. I suspect this is one reason why MMORPG developers don’t even try to eliminate bottomfeeding mechanics; the existence of such mechanics insures that there’s always a way for every player to win, even if it takes longer than a more carefully considered or riskier process.

    Simulations, by contrast, can very definitely be failed. In any run of a simulation, it is always possible to get the world-state into a wedged condition or otherwise reach a point where the “win” state is no longer attainable. The landing can be botched; the girl can slap you and never speak to you again; and so on. So in a highly simulationist game, simple repetition won’t work. The only way to win (after restarting the sim) is to recognize the patterns that lead to the various end-states and choose the actions that lead to the “success” state. Gamers who like simulations tolerate and even welcome failure because it provides additional data through which a path to success can be identified by effective observation and creative thinking.

    I’ve come to characterize these playstyles as “Persistence” versus “Perception.” The point here being that Persistence-oriented gamers are intolerant of failure and will generally have only bad things to say about simulationist games in which failure is possible and perceptiveness is required for success. Likewise, a gamer who wants to be rewarded for his Perceptiveness is generally going to disdain highly rules-based games in which it’s always possible to win through mindlessly repetitive action.

    So if Persistence-oriented gamers don’t tolerate failure-states well, and Perception-oriented gamers do, what does that suggest about how to implement a character death mechanic that a majority of a game’s players can tolerate?

    Comment by Bart Stewart — 1 July, 2010 @ 9:03 PM

  32. We had/have permadeath in MUD1 and MUD2.

    Something that seems to be missing from this debate is how often you die. In today’s MMOs, characters “die” fairly regularly – barely a playing session goes by without its happening at least once. Sometimes it happens so often that you need a secondary mechanic to get the message through to players that they’re not going to win (eg. WoW‘s cumulative armour/weapon damage penalty on death). In MUD2, high-level characters will go months without dying. Sure, they lose fights, but they don’t die in them – they flee. Dying, when it happens, is usually because the player got over-confident (figured they could take some opponent on while too heavily injured), too greedy (stayed too long because the potential rewards overrode their sense of risk), or the result of a well-planned and systematic lynching having made one enemy too many.

    When people today talk about death penalties, they assume that the frequency of dying must be the same as it is in their current MMO of choice. In such circumstances, sure, permadeath is ridiculous. Think of it this way: what you currently have as “dying” is like a “flee from fight when hit points get this low” threshold. The default value of this threshold is to flee when the enemy you are fighting could kill you in a single blow. If you never change the threshold, you will never die. However, in some cases you may wish to reduce it so you flee when, say, you have 90% of the “safe” number of hit points left. You might do this to save a colleague, or to get some epic loot, or because you know something the game doesn’t (there’s a second healer about to arrive).

    That’s not quite how MUD2 and its ilk do it, but it gives you a sense of the difference between what you see as a “death” event in WoW and what would be a “death” event in a permadeath MMO.


    Comment by Richard Bartle — 2 July, 2010 @ 1:10 AM

  33. @Brian “The problem here is if the perception changes so that the better rewards become the only desirable ones. Then people will complain about how they are “forced” to do the harder route to get the “real” rewards.

    Of course, at some point as a designer you have to accept that someone will always complain about something and forge ahead anyway. :)”

    Very true :) I’d hope that people would be a bit more logical but maybe that’s just wishful thinking ;) It’s kinda funny because when I play MMOs I accept that if I want to raid, I need to invest a lot more time and effort. I also understand though that these rewards are going to be better than if I just soloed and did quests all of the time.

    If we put time sinks aside (which is something I’m glad to see disappearing in MMOs), I’d be more than happy to accept a higher risk and penalities in order to achieve better rewards. Of course it would have to be voluntary, just like partaking in a raid is, but if I chose to do it then I would know that failure would mean very hard consequences. I wonder if the majority of other peoples would also accept that?

    Comment by We Fly Spitfires — 2 July, 2010 @ 2:10 AM

  34. Failure Should Really F***ing Hurt (But It Shouldn’t Be The Death Of You)

    [...] by all of the talk about death penalities in MMOs and the risk and difficulty it brings, I wanted to share an idea about about a type of failure that [...]

    Pingback by We Fly Spitfires – MMORPG Blog — 2 July, 2010 @ 2:05 PM

  35. The Best Of The Rest: Why’d You Have To Go And Declare Independence Edition

    [...] Brian ponders death. [...]

    Pingback by We Fly Spitfires – MMORPG Blog — 4 July, 2010 @ 3:32 PM

  36. Full Spectrum Challenge

    [...] solved. There tends to be one right answer that maximizes the potential reward (or minimizes the potential punishment) from the decision. Players almost inevitably “min/max” their game choices, then, [...]

    Pingback by Tish Tosh Tesh — 2 August, 2010 @ 5:32 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


January 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