[MUD-Dev] [DSN] Why Permadeath is Irrelevant

Paul Schwanz paul.schwanz at east.sun.com
Thu Aug 30 16:36:05 New Zealand Standard Time 2001

Yeah, I know this is one of those topics that just won't die
(permanently), but I thought I'd give it another go.  Really, this
isn't a troll, although I must admit to feeling a bit on the outside
of the current technical topics.  It isn't that the technical stuff
isn't interesting.  I just don't know enough about most of it to
even form an opinion.  Am I the only one on this list whose
programming expertise is summed up by the following?

  if (inWetPaperBag)
       then ????

Anyway...enough noise.

Why Permadeath is Irrelevant
(Or, at the least, not the Real Issue)

I believe that gaming is about choices and feedback.  In an arcade
style game, the choices may be based on which way you move a
joystick.  If you zig when you are supposed to zag, you get some
sort of negative feedback.  If you zag correctly, you get some sort
of positive feedback.  In a MUD or MMORPG, the interaction can be a
bit more sophisticated.  There can be a much greater range of
choices.  These might include where you choose to locate your
character, how you choose to invest time and money, what use you
make of information you receive, etc.  But in the end, I think that
those who play MUDs as "games" are simply making choices and
experiencing feedback.

Permadeath is one possible negative consequence.  There are of
course many others.  But, regardless of how it is precisely defined,
permadeath is perceived to be at or near one extreme of the range of
possible (in-game) negative consequences.  Consider the following
illustration (best viewed using a fixed-width font).

  Insignificant                    Momentous

If you were to chart possible negative consequences along this
scale, permadeath would be at or near 10.  Losing 1gp might be at or
near 0.  Perhaps losing 10% of your skill would rank around a 5.
This is really all about perceptions, and different gamers might not
see things precisely the same way, but in general, I think there
would be a great deal of consensus about where certain negative
consequences fell in the range.

But negative consequences (the risks) are only one part of the
picture.  Remember there are actually three parts to the gaming
experience.  The other two parts are the choices and the positive
consequences (the rewards).  A more complete illustration might look
like this.

         Insignificant                    Momentous
  Risk   0...1...2...3...4...5...6...7...8...9...10

  Choice 0...1...2...3...4...5...6...7...8...9...10

  Reward 0...1...2...3...4...5...6...7...8...9...10

The interesting thing is that the three scales don't really seem to
function independently.  In some ways the significance of the risk
seems to inform the significance of the choice and even the
significance of the reward.  Similarly, as has been expressed on
this list, gamers seem to believe that rewards should be appropriate
to risks.  Gamers expect mobs with more hit points to carry more
loot.  (If this were not so self evident, perhaps it would deserve
being written down as a law.)  The intuitive correlation between the
significance of choices and feedback help explain why I think that
permadeath is not really at issue.  If a game designer sets a limit
on risk (i.e. 10% skill loss is the greatest consequence your
character can suffer), all he is really doing is cutting the scale
in half.  He is effectively putting a wall up with a sign on it
telling every gamer, "In my game, you cannot experience any risk,
reward, or choice more momentous than this."

         Insignificant                    Momentous
  Risk   0...1...2...3...4...5|
  Choice 0...1...2...3...4...5|
  Reward 0...1...2...3...4...5|

I believe that this is a valid design choice (particularly in JB's
case, where he is purposely designing interaction that is of a more
casual nature), but I don't think it is what most designers really
want to do.  Personally, as a game player, I don't particularly like
having designers place limits on the significance of my choices.
I'd much rather be free to choose my own level of interaction and
significance.  I tend to like more momentous choices since they seem
to speak to my own ideas about heroism.  Yet it seems to me that
designers who often share this personal preference still resort to
cutting the significance scale in half and putting up a wall.

Why Permadeath is not Quite Irrelevant
(But Still isn't the Real Issue)

Why do game designers remove permadeath from their bag of possible
consequences?  I don't think this really stems from a desire to take
away game freedom or to limit a choice's significance.  Once again,
I don't think permadeath is the Real Issue.  The Real Issues have
more to do with Predictability and Control.

To illustrate, consider the final scene in "Indiana Jones: The Last
Crusade."  Presented with a number of goblets, of which, one is the
Holy Grail, our hero is instructed to choose wisely.  If he makes
the wrong choice, he will suffer a quick and horrible death.  If he
makes the right choice, he will save the life of his dying father.
Everyone knows that this is a momentous decision.  The hero knows
it.  The audience knows it.  This type of momentous decision makes
for a great movie.  I think it would also make for a good game
situation.  On our scales, it might look like the following, where
choice (A) will lead to either consequence (B) or reward (C).

         Insignificant                    Momentous
  Risk   0...1...2...3...4...5...6...7...8...9..(B)

  Choice 0...1...2...3...4...5...6...7...8...9..(A)

  Reward 0...1...2...3...4...5...6...7...8...9..(C)

In my mind, it is very important to gaming (perhaps not so much for
movies) that the gamer's perception about the significance of a
choice and its corresponding risks and rewards line up well on a
simple significance scale.  Indy knew the risks.  He knew the
rewards.  He understood the momentous nature of his choice.  This
gives a needed sense of predictability and control to the situation.
I believe that it is absolutely valid to have momentous risks like
permadeath in a game, if the gamer perceives that the choices and
rewards are predictably appropriate to the risks.  If Indy just
grabs a handy cup to take a drink of water (i.e. he thinks he is
making a mundane choice) and falls over dead, this isn't nearly as
compelling.  Similarly, if Bubba steps through his front door in
MUDdom and is instantly permakilled by an Ogre (NPC or otherwise),
this will seriously offend Bubba's sensibilities about how games are
supposed to work.  Intuitively, Bubba has a deeply held belief that
a choice that is perceived to be very low in significance should not
lead to such drastic consequences.  Violating this belief will
result in a lot of frustration.  Bubba's perception of the
occurrence (although I doubt he'd express it as such) might look
like the following, where choice (A) with a possible reward of (C)
led inexplicably to (B).

          Insignificant                    Momentous
  Risk    0...1...2...3...4...5...6...7...8...9..(B)

  Choice (A)..1...2...3...4...5...6...7...8...9...10
  Reward (C)..1...2...3...4...5...6...7...8...9...10

If you offend his sensibilities too often, Bubba will not continue
logging onto your MUD.  The quick fix is to chop the significance
scale in half.  This will effectively limit the disparity that is
possible between a choice and its consequence, but at the expense of
more momentous choices.  Furthermore, it doesn't really address the
true source of frustration since Bubba may still feel a lack of
control where a perceived mundane choice leads to a 10% skill loss.
In my mind, the better solution would be to find other methods for
limiting the disparity while not forfeiting momentous choices.  The
goal would be to have risks that predictably line up with
perceptions about corresponding choices and rewards.  However, such
an undertaking is nontrivial, especially in a multi-user world where
the actions of other gamers are so often unpredictable.  On the
other hand, I don't think it is impossible.  In any case, I prefer
to focus on this challenge instead of simply debating whether gamers
will or will not accept permadeath.

The second thing that I liked about the scene with the Holy Grail
(and that I think is conducive to good gaming) was the fact that
Indy needed to choose "wisely."  The choice was not entirely random.
Indy drew on his knowledge of archaeology and the story behind the
grail to choose the goblet that would most likely be used by a poor
carpenter's son.  This too gives a sense of control over the
outcome.  Whether the consequence of a choice is positive or
negative, as a gamer, I'd like to know that I received a reward
because I made a wise choice or that I experienced dire consequences
because I made a poor choice.  In my mind, momentous consequences
should not be based on the roll of a d20.  As choices move toward
the right on our significance scale, I think that the degree of
randomness in the outcome needs to move toward zero.

(Three Quick Recommendations)

1. Give a wider range of significance to actions, instead of a
narrower one.  It seems to me that too often, game activities seem
to fall into one of two categories: a) things that are boring or b)
things that will get you killed...oops, three categories...c) things
that are boring AND will get you killed.  Obviously, permadeath
would not work well in this type of game.  I'd like to see a lot
more interesting and challenging activities that resulted in all
kinds of consequences that had nothing to do with death, permanent
or otherwise.  The interesting thing is that when you beef up the
choices on the lower end of the significance scale, I think this
allows you to have more success with adding choices at the other end
as well.  The result is more freedom for the gamer.

2.  Use intuitive real-life concepts to design game systems that
help players manage their risks so that you can minimize the
disparity between the perceived significance of their choices and
the perceived severity of consequences.  These concepts might
include territory, vocation, justice, reputation, karma, access
control, and more formalized community building.  Help them have a
more predictable experience when it comes to the correlation between
choice, risk, and reward.

3.  As consequences become more significant they need to be tied
more strongly to the appropriate or inappropriate nature of a
corresponding choice.  It is extremely important that dire
consequences are seen to be avoidable.  Not being able to link a
negative consequence back to a poor choice on my part will always
leave me with a feeling of helplessness that can lead to
frustration, despair, and eventually departure.  It is never fun to
suffer the slings and arrows of outrageous fortune.  Give me more
control, however, and I will be happy to face whatever projectiles
comes my way.


MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list