Character games

5 views
Skip to first unread message

JAN THORSBY

unread,
Dec 7, 2003, 6:22:24 PM12/7/03
to
I was thinking it might be good if we got a new category of games, called
"character games". In character games you try to control the player
character in such a way as it would be beneficial for the player character
to behave.

For instance let's say you play a game where you start in a sinking ship. If
I where playing normal game I would begin to explore the ship, to see if I
could find stuff before they disappeared in the ocean. But if I where
playing a character game, I would run towards the lifeboat at once.

Now some of might claim that there are already lots of games that are
character games. Or that you always play a game as if it was a character
game. And you would probably be right. However I think it would be good that
the player was told at the beginning of the game that he was playing a
character game. The player could be told that he should try to react the way
player character would react. Or more practically (in most cases) the player
could be told that he should try to finish the game on the first attempt,
without getting killed. I think telling the player this could make a lot of
tings that don't normally work so well in games, work well. And some tings
that usually works well, would not work so well.

Time limits (particularly long ones) are often considered to be annoying,
because there is normally no challenge in keeping a time limit. If one fails
to keep it, one just restart and play through the game again, refraining
from doing all the useless things one did the first time. But time limits
might be interesting if one is determined to survive the first time. The
same might be true for limited amounts of items one can carry and limited
resources (like having only a few matches to explore a house) and randomised
combat and random dangers in general. Though I think there is still a fairly
large risk that many will find these features annoying.

Another example: say the player is in a house filled with monsters and are
trying to get out. He has a choice of two doors to open. Behind one of the
doors is a monster that will kill the player, or at least cause lots of
difficulties. Maybe there is some puzzle that when completed gives the
player a hint about what door is the right one. Maybe the player can listen
to the doors to hear where the monster is, or maybe he can smell the
monster. In a non-character game this would not work well at all. If the
player makes the wrong choice he could just reload and make the other
choice. There would probably be very little excitement involved. And there
would be no real reward for solving the puzzle. And if the player makes the
right choice, he would quite likely go back and try the other door anyway.
But in a character game it might work.

I think character games could work well in the action category. There are
very few action games in IF. (The only game that springs to mind that I
might consider to be action is Reverberations. Though there are some games
that have short action sequences.) Elaborate action sequences might not
work so well in a non-character game, because there is likely to be a need
for the player to do lots of things within small time limits. (If for
instance the player is being chased through city streets, and is supposed to
jump from car to car, and shoot at stuff etc.) Also if the player comes
across something which is likely to kill him, he might "save" before testing
if the ting indeed can kill him. This might get tedious if there is lots and
lots of things that can kill the player. But in a character game the player
knows he is supposed to try and win on the first attempt, and not get
killed, so he wont test if the tings can kill him. (He is still allowed to
save.)
If you want to make tings exiting, but don't want to actually kill off the
player, you could give the player the impression that things might kill him,
even if they can't, or that there is an important time limit, even if there
isn't. In a non-character game the player is likely to see through this much
more easily.

I think character games could work well in some types if horror. For
instance if the player where being chased by a zombie or a chainsaw wielding
psycho or whatever.

I think character games could work well in games with humorous plots. (There
are lots of games that have humour, but relatively few that have a humorous
plot. Often the humorous games are ordinary puzzle games, with lot of jokes
in the object descriptions, and in what you are told if you try to do stuff
you can't do etc. Not that there is anything wrong with that.)

I think character games probably should have easy puzzles, if any, so the
player has a change of winning on the first attempt. I think character
games, particularly those with time limits, should probably not focus much
on exploring (And maybe sometimes not even have the examine command or the
search command).

I also think that it would be good if the ones who writes games decides if
their game is a character game or not, and don't make a mix of a character
game and a non-character game. Say you where playing a game taking place on
a sinking ship, and you where supposed to explore the ship. Then it would
not be a character game. But say the same game also puts the player in a
burning house, where there is nothing to find, and you only have very few
turns to get out before the house collapse. Since the player knows you where
supposed to explore the ship he will likely try to explore the house as
well, and will therefore die many times, which is annoying. If might be
worse if the house came before the ship, because then the player might not
search the ship, and he will miss vital objects.

Also it would be very bad if the player was told he was playing a character
game, even though he was indeed supposed to search the sinking ship.


Piglet

unread,
Dec 8, 2003, 8:44:10 AM12/8/03
to
This is somewhat simmilar to an idea i've been thinking around recently,
sort of surviavalist if. The idea being that the game would be easy
(easy puzzles that sort of thing), if it wasn't for more continous
problems (such as food, drink, exposure, hypothermia etc).

It could all be quite interesting, I guess the most important thing is
to make sure the players know that they aren't trying to 'solve' the
game but simply 'survive' it. Some interesting problems crop up with
replayabilty though, is there a way that such a game could stay
interesing having completed it (or even failed it) before.

I guess there are also problems with the saving system of most adventure
creation systems and this type of game, how do you keep the immediacy if
a player knows they can simply restore the state to what it was 2
minutes ago.

Finally I'm not to sure about the name you have given, 'character games'
seems a bit of an arbitary choice in some ways, to me it makes me think
of a game that is based heavily on character interaction (like Galatea)
rather than a game in which the object is to complete it first try.

Piglet

Quintin Stone

unread,
Dec 8, 2003, 10:31:27 AM12/8/03
to
On Sun, 7 Dec 2003, JAN THORSBY wrote:

> I was thinking it might be good if we got a new category of games,
> called "character games". In character games you try to control the
> player character in such a way as it would be beneficial for the player
> character to behave.
>
> For instance let's say you play a game where you start in a sinking
> ship. If I where playing normal game I would begin to explore the ship,
> to see if I could find stuff before they disappeared in the ocean. But
> if I where playing a character game, I would run towards the lifeboat at
> once.
>
> Now some of might claim that there are already lots of games that are
> character games. Or that you always play a game as if it was a character
> game. And you would probably be right. However I think it would be good
> that the player was told at the beginning of the game that he was
> playing a character game. The player could be told that he should try to
> react the way player character would react. Or more practically (in most
> cases) the player could be told that he should try to finish the game on
> the first attempt, without getting killed. I think telling the player
> this could make a lot of tings that don't normally work so well in
> games, work well. And some tings that usually works well, would not work
> so well.

The idea that this classification should be created and applied to certain
games seems rather arbitrary to me. (The name also implies, as Piglet
points out, that game focuses on character interaction.) I'm tired, it's
early, so bear with me if I ramble.

In most good games, the goal should be to act in the best interest of the
PC. (Bedlam being a good example of where this isn't necessarily true.)
I take issue with the idea that the player should be given this
out-of-character information that the game follows a particular style
instead of prompting the player through in-game cues. That just smacks of
lazy design to me. A game can be given immediacy and survive time limits
if the writing is there to support them. That's what should put the
player in the desired mindset; not a line about this being a "character
game". And the idea that the player should try to finish this type of
game on the first attempt without getting killed? Sorry, but I rarely
play a game trying to off myself. And if I recognize that I'm likely to
snuff it in a game, I'm going to try and save regularly.

> Time limits (particularly long ones) are often considered to be
> annoying, because there is normally no challenge in keeping a time
> limit. If one fails to keep it, one just restart and play through the
> game again, refraining from doing all the useless things one did the
> first time. But time limits might be interesting if one is determined to
> survive the first time. The same might be true for limited amounts of
> items one can carry and limited resources (like having only a few
> matches to explore a house) and randomised combat and random dangers in
> general. Though I think there is still a fairly large risk that many
> will find these features annoying.

Yeah, I'd say so. Okay, I think I see where you're going with this. You
like the idea of these survivalist, time-limited, non-renewable-resource
type of games. There's nothing wrong with that. But at the same time,
you recognize that the "community" finds many aspects of them frustrating
(time limits, inventory limits, consumable resources, randomized events,
randomized combat). And so (this is how I see it) you want to categorize
and label games like this so that when people play them, they judge them
within this established framework, instead of hating them simply because
they have time limits, inventory limits, etc.

The problem is that most players are going to judge these games by their
own standards no matter what. Now if you want to start it out with a
little blurb specifying the type of game it is (like in an initial y/n
instruction page), that might work better than trying to get support for
establishing a "category" of game, when, as far as I know, no other kind
of formal "category" like this really exists within IF. (Though I could
be wrong.)

> But in a character game the player knows he is supposed to try and win
> on the first attempt, and not get killed, so he wont test if the tings
> can kill him. (He is still allowed to save.) If you want to make tings
> exiting, but don't want to actually kill off the player, you could give
> the player the impression that things might kill him, even if they
> can't, or that there is an important time limit, even if there isn't. In
> a non-character game the player is likely to see through this much more
> easily.

In the end, players will play however they want to play. Telling them
that they should try to survive their first time through, that seems
pretty meaningless to me. Really, is the player going to assume that they
can simply run through the game without taking chances or exploring items
that look like they might be interesting? Doubtful, if only because of
their experiences with other games. I'm sure you'll argue that these
character games are built to be different from those other games, but do
you have the player's trust that this is the case simply because you say
it is? I think people play the way they do because they are more or less
conditioned by the games they play.

Eh. I'm not in favor of establishing formal "categories" of games that
exist only to put players in a certain mindset. I say, if you want to do
that, you should do it in the game. I am in favor, on the other hand, of
people writing and playing the games that they enjoy. You don't need
anyone's approval to do that.

/====================================================================\
|| Quintin Stone O- > "You speak of necessary evil? One ||
|| Code Monkey < of those necessities is that if ||
|| Rebel Programmers Society > innocents must suffer, the guilty must ||
|| st...@rps.net < suffer more." -- Mackenzie Calhoun ||
|| http://www.rps.net/QS/ > "Once Burned" by Peter David ||
\====================================================================/

Kevin Venzke

unread,
Dec 8, 2003, 5:41:19 PM12/8/03
to
It might be interesting to have a game where you couldn't SAVE or

UNDO, in order to raise the stakes for the player. (Thinking of Diablo 2,

or those Scott Adams games where I seem unable to save.)

But what, of interest, could you really do in such a game? The game would

have to be designed in such a way that if I'm clever enough, I won't have

to die. That would be very hard to design, because the author has to give

the player enough hints to know what he should do, but sufficiently few

that there is still challenge.

> For instance let's say you play a game where you start in a sinking ship.
If
> I where playing normal game I would begin to explore the ship, to see if I
> could find stuff before they disappeared in the ocean. But if I where
> playing a character game, I would run towards the lifeboat at once.

It seems to me that if the player expects that he should be able to do

this, there will be a lot of puzzles that you won't be able to use. You

couldn't have crucial items that the player might not see fit to take, and

let the player leave them behind someplace they can't return to. Every

solution would have to be pretty obvious from a cursory reading of the

text, I think.

> I think character games could work well in the action category. There are
> very few action games in IF.

It's hard to design "action" puzzles without requiring the player to

read your mind as to which action is right and which are wrong. You

could bury hints in the text, but that would get pretty annoying.

Kevin Venzke


Piglet

unread,
Dec 8, 2003, 6:22:02 PM12/8/03
to
> It's hard to design "action" puzzles without requiring the player to
>
> read your mind as to which action is right and which are wrong. You
>
> could bury hints in the text, but that would get pretty annoying.

I guess one way you might manage this is to have many non-trivial
solutions to each puzzle, thus you dont have to mind read (if you think
of a potential solution there's a good chance it's implemented), and you
dont have to go hint hunting.

It was also reduce the problems associated with 'critical' items,
perhaps you didnt collect the rope, that might make it harder to climb
down the cliff safely, but crucially not impossible (maybe you can make
a parachute). A poor example I know, but I think it makes a point.

Piglet

Mike Roberts

unread,
Dec 8, 2003, 6:34:49 PM12/8/03
to
"JAN THORSBY" <jtho...@c2i.net> wrote:
> I was thinking it might be good if we got a new category of games,
> called "character games". In character games you try to control the
> player character in such a way as it would be beneficial for the
> player character to behave.
>
> For instance let's say you play a game where you start in a sinking
> ship. If I where playing normal game I would begin to explore the
> ship, to see if I could find stuff before they disappeared in the
> ocean. But if I where playing a character game, I would run
> towards the lifeboat at once.

To me, you're not describing a separate category of game here. I'd simply
call your first example old-fashioned, since the player character is just
collecting treasures without any motivation in the context of the story.
What you're describing is what Roger Ebert calls the "idiot plot": if for
one second any of the characters would stop acting like idiots, the story
would end. Yes, there's a lot of that sort of thing in IF, but it's
especially prevalent in older IF and it's seen as anachronistic these days.
It's already pretty well established in modern IF that it's a virtue for
player characters to have good reasons to pursue their quests.

In any case, I think your definition of "beneficial for the player
character" is too narrow. People don't always act in their own best
interests, and even when they do, personal safety isn't always their highest
priority. A PC in a modern, character-driven game could perfectly well risk
life and limb by going back into a sinking ship if she had a good enough
reason: saving a loved one trapped below decks, recovering the briefcase
nuke smuggled on board by the enemy spy the PC has been tracking, stealing a
valuable jewel left behind by a fleeing millionaire passenger.

If the PC's overriding motivation is personal safety, you might still be
able to create a decent story, but I'm not sure it would be as compelling as
one where the PC puts himself at risk for some loftier goal. Look at any
good man-against-nature story, or any disaster movie, and you'll find
something that's superficially about people trying to save themselves from
some calamity - but in every good example I can think of, the real story is
about the people who rush back into the burning building to save the others.
(Or whatever; the heroics don't always have to be motivated by good
intentions. An antagonistic character could be equally heroic about going
back into the burning building to make sure a foe *doesn't* make it out
alive.)

--Mike
mjr underscore at hotmail dot com

antispam...@bmrb.wisc-antispam.edu

unread,
Dec 8, 2003, 6:17:50 PM12/8/03
to
Kevin Venzke <ste...@yahooo.fr> wrote:
:It might be interesting to have a game where you couldn't SAVE or

:
:UNDO, in order to raise the stakes for the player. (Thinking of Diablo 2,
:
:or those Scott Adams games where I seem unable to save.)
:
:
:
:But what, of interest, could you really do in such a game? The game would
:
:have to be designed in such a way that if I'm clever enough, I won't have
:
:to die. That would be very hard to design, because the author has to give
:
:the player enough hints to know what he should do, but sufficiently few
:
:that there is still challenge.

It would also have to be designed in such a way that the
game doesn't take too long. After all, one of the main
reasons people save games isn't to cheat, but to go do
something else for a while and come back later. The game
would have to be short enough that you don't mind having to
do the whole thing in one sitting. And it would really, really
annoy me if I hadn't done anything wrong, but I had to quit
and lose my progress anyway because it was time to shut the
computer down and go to bed.

Quintin Stone

unread,
Dec 8, 2003, 6:53:56 PM12/8/03
to
On Mon, 8 Dec 2003, Kevin Venzke wrote:

> It might be interesting to have a game where you couldn't SAVE or UNDO,
> in order to raise the stakes for the player. (Thinking of Diablo 2, or
> those Scott Adams games where I seem unable to save.)

Interesting, maybe. Universally reviled, likely.

Michael Coyne

unread,
Dec 8, 2003, 7:26:36 PM12/8/03
to
Kevin Venzke wrote:

> It might be interesting to have a game where you couldn't SAVE or
>
> UNDO, in order to raise the stakes for the player. (Thinking of Diablo 2,
>
> or those Scott Adams games where I seem unable to save.)

You could call it "Annoyotron 2".


Michael

merri

unread,
Dec 9, 2003, 12:15:51 AM12/9/03
to
On Mon, 08 Dec 2003 22:41:19 GMT, Kevin Venzke wrote:

> It might be interesting to have a game where you couldn't SAVE or
> UNDO, in order to raise the stakes for the player. (Thinking of Diablo 2,
> or those Scott Adams games where I seem unable to save.)

It should be real-time, too.

Paul Prestidge

unread,
Dec 9, 2003, 2:27:18 AM12/9/03
to
> ... And it would really, really

> annoy me if I hadn't done anything wrong, but I had to quit
> and lose my progress anyway because it was time to shut the
> computer down and go to bed.

Perhaps one way to avoid this is adopt the a save system similar to the
'Iron Man' one from Alpha Centauri: you can save the game, but when you
do, it immediately quits with no chance for further action. The other
place you'll see this is permadeath roguelike games. This is sort of
similar to the save system in Diablo 2, except this way it *actually*
saves instead of dumping you back in town with new monsters on your
return ;-)

Paul

PS Yes, first time caller, long time listener. Hi everybody :-)

--
Having sex is like playing bridge...
If you don't have a good partner, you'd better have a good hand.

Anvilsmith

unread,
Dec 9, 2003, 8:11:03 AM12/9/03
to
There are two problems you'd be dealing with when making this kind of
game: the player's impression that he should "see it all", and his
disinterest in the character's well-being. I have already solved the
former in a rather efficient way - by giving the character moral
choices towards which the player himself would feel strongly. The
character in Daimons can follow one of his four guiding principles
whenever challenged to make a decision, and I made sure that at least
some of his decisions won't be palatable to the major audience. There
are even more powerful ways to keep players from wanting to see it
all... For instance, if you allow a character to commit child rape (a
bad example, since doing this would get everyone in the IF community
against you), a lot of players will never want to take part in it. The
same goes for a really well-done murder scene. If you can just type
"kill DuMonde" and be done with it, it'll be relatively easy. However,
if the game forces you to get the pillow and tell the your wife that
it'll be all right while you slowly kill her, it's an entirely
different thing. You can artificially increase this effect by adding
intermediare verbs, so that the command "stab Jason" would first be
greeted with the message "You feel you should grasp the knife
tighter... It might just slip."

>grasp knife
The squeezing sensation brings nausea along your arm, and from there,
into your mind... For a moment, your heart pulses with the force of
your clutched hand, pushing more blood into your broken teeth. It
tastes as you imagine...

>stab Jason
Still not close enough... Just take another step. One more step, and
he'll be in your reach.

>approach Jason
The warm knife shivers in your hand expectantly. He's there, an inch's
length away, and the books keep him turned from you. Your hand curls
lovingly around the weapon, further smearing it...

>stab Jason
The shiver turns into a shake, too fierce for your small hand to bear.
You foolishly release the knife above the floor, where it collapses
with a harsh clattering noise. Jason immediately turns to you, alarmed
as much by your appearance as the sound...

>stab Anvilsmith
Good luck with that. Oh, and there's a walkthrough available if you
need it.

'In character games you try to control the player


character in such a way as it would be beneficial for the player
character

to behave.'
How do you know what would be beneficial for the player character?
Sometimes, death is the best option he could have, but also the
quickest way to end the game. If I follow the general idea while
preserving my opinion, I'll end up with a very short and boring game.
I won't get into a discussion about how crucial it is to design a
dynamic story, and how one can accomplish that... You seem to be more
interested in games whose only purpose is survival, and the outwitting
of chainsaw-wielding zombies.

'The player could be told that he should try to react the way
player character would react.'
What you're really advertising is the kind of game that encourages
players to adhere to real-world logic. Nothing special with that,
though I would like it if designers made their games more realistic.

'Time limits (particularly long ones) are often considered to be
annoying,
because there is normally no challenge in keeping a time limit.'
Not really true. You assume that failing a time trial forces you to
restore the game, but there's nothing to keep you from just adding
another puzzle for the player to solve if he fails. A good example
would rely on Babel's overall time trial... While I'm not sure if the
lights do go out after a while, or if the whole thing's just a gimmick
meant to scare me into saving and reloading every time I type look, it
could turn into a "find an alternative power source" puzzle - a
welcome diversion from the "science is evil" plot.

'and randomised combat and random dangers in general.'
Players wouldn't hate these things if they led to interesting
situations. I guarantee you that if you make a wilderness survival
game with a partially random disease-catching system that stacks
percentile chances related to a lot of ambient factors, the player
will still do his best to keep his PC healthy, even though there's
always a chance that he'll get sick.

'In a non-character game the player is likely to see through this much
more easily.'
Very wrong. If the player expects the designer to give him reasonable
challenges, he'll suspect that the time trial is a fake, giving him a
lot more time than he deserves. Only the adventure gamer, used to
unforgiving time trials, will take something like this seriously, and
only if there are serious puzzles to complement it.

'I think character games probably should have easy puzzles, if any, so
the
player has a change of winning on the first attempt.'
Don't assume any of the puzzles have to be lethal, or lead towards
dead ends... If there's a chance the character will die doing
something and live happily ever after by avoiding it, the player might
want to choose the second option, even if this freezes the plot.

Jess Knoch

unread,
Dec 9, 2003, 9:30:46 AM12/9/03
to
Kevin Venzke wrote:
> It might be interesting to have a game where you couldn't SAVE or
> UNDO, in order to raise the stakes for the player.

> The game would


> have to be designed in such a way that if I'm clever enough, I
> won't have
> to die. That would be very hard to design, because the author has
> to give
> the player enough hints to know what he should do, but
> sufficiently few
> that there is still challenge.

It is very hard to design *any* game to give sufficient hints to the
player so that they know what to do, while sufficiently few so that
the game is still a challenge. That's almost the crux of puzzle
design summed up in a few short phrases. Why, oh why, would taking
out "save" and "undo" raise the stakes for the player?

Taking out SAVE and UNDO just means the player has to RESTART, or
more likely, just QUIT. It only adds annoyance without raising any
stakes. It's still difficult to design a game that is clued enough
but not too much, and it's still impressive if it's done well.

--
Jess K.


Piglet

unread,
Dec 9, 2003, 12:10:23 PM12/9/03
to
> Why, oh why, would taking out "save" and "undo" raise the stakes for the player?

Player's would know that each move is important, and that they couldn't
go back to a recently prior save, if that doesnt raise the stakes then
what does?

(not that I am neccesarily in favour of it)

Piglet

Andrew Plotkin

unread,
Dec 9, 2003, 12:42:06 PM12/9/03
to
Here, Piglet <pig...@ignika.com> wrote:
> > Why, oh why, would taking out "save" and "undo" raise the stakes
> > for the player?
>
> Player's would know that each move is important, and that they couldn't
> go back to a recently prior save, if that doesnt raise the stakes then
> what does?

There's raising the stakes, and there's raising the annoyance level.

The difference between these two things is very, very subjective. It's
a distinction of outcome, for a given player: either the player is
thinking "This is really important!" or the player is thinking "This
might be a waste of all the time I just spent."

The observable fact is that for the *general* case of IF, removing
save/undo causes the latter outcome, for nearly all players.
Contrariwise, leaving save/undo *in* the game observably fails to ruin
the player's sense of importance.

So while I'm not going to declare the practice verboten, it's not as
simple as hacking the parser. You have to set up an unusual form of
game in which the alteration works. (Simple example: a very very short
game. I think we've seen one-move games, or three- or four-move games,
where "undo" wasn't an issue.)

--Z

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
* Make your vote count. Get your vote counted.

antispam...@bmrb.wisc-antispam.edu

unread,
Dec 9, 2003, 1:35:51 PM12/9/03
to
Paul Prestidge <axo...@ihug.co.nz> wrote:
: > ... And it would really, really

:> annoy me if I hadn't done anything wrong, but I had to quit
:> and lose my progress anyway because it was time to shut the
:> computer down and go to bed.
:
:Perhaps one way to avoid this is adopt the a save system similar to the
:'Iron Man' one from Alpha Centauri: you can save the game, but when you
:do, it immediately quits with no chance for further action. The other
:place you'll see this is permadeath roguelike games. This is sort of
:similar to the save system in Diablo 2, except this way it *actually*
:saves instead of dumping you back in town with new monsters on your
:return ;-)

I suppose one way to allow for "save because I'm quitting for
a while" while disallowing "save because I'm going to try something
dangerous" is to: 1 - Associate one (and only one) savegame file
with each restart of the game, so once you begin a game, you can
only save to that same one file. To use a new file you'll have
to restart the whole game. 2 - upon death, delete that file.
How to do that with the current IF engines, however, I don't know.
(With TADS, for example, you run in a "sandbox" much like Java,
where you only have access to the hard drive if player explicitly
allows it in his settings. The default is to assume the player
doesn't trust the game program and it can't touch the disk drive in
any way other than the standard save/restore mechanism.)

antispam...@bmrb.wisc-antispam.edu

unread,
Dec 9, 2003, 1:40:05 PM12/9/03
to
Andrew Plotkin <erky...@eblong.com> wrote:

:Here, Piglet <pig...@ignika.com> wrote:
:> > Why, oh why, would taking out "save" and "undo" raise the stakes
:> > for the player?
:>
:> Player's would know that each move is important, and that they couldn't
:> go back to a recently prior save, if that doesnt raise the stakes then
:> what does?
:
:There's raising the stakes, and there's raising the annoyance level.
:
:The difference between these two things is very, very subjective. It's
:a distinction of outcome, for a given player: either the player is
:thinking "This is really important!" or the player is thinking "This
:might be a waste of all the time I just spent."

I've found that one of the common uses I have for savegames is to
try something explicitly stupid because I want to see what the
author's response is - even though I know it will get me killed.
Even if an action is stupid and deadly, I might want to see the
prose that goes with it because it's fun. If I was writing a game,
I wouldn't want to waste a lot of time writing fun prose to go
along with different death scenes and then never have players
actually see any of it.

Daryl McCullough

unread,
Dec 10, 2003, 6:01:42 PM12/10/03
to
Andrew Plotkin says...

>
>Here, Piglet <pig...@ignika.com> wrote:
>> > Why, oh why, would taking out "save" and "undo" raise the stakes
>> > for the player?
>>
>> Player's would know that each move is important, and that they couldn't
>> go back to a recently prior save, if that doesnt raise the stakes then
>> what does?
>
>There's raising the stakes, and there's raising the annoyance level.
>
>The difference between these two things is very, very subjective. It's
>a distinction of outcome, for a given player: either the player is
>thinking "This is really important!" or the player is thinking "This
>might be a waste of all the time I just spent."
>
>The observable fact is that for the *general* case of IF, removing
>save/undo causes the latter outcome, for nearly all players.
>Contrariwise, leaving save/undo *in* the game observably fails to ruin
>the player's sense of importance.
>
>So while I'm not going to declare the practice verboten, it's not as
>simple as hacking the parser. You have to set up an unusual form of
>game in which the alteration works. (Simple example: a very very short
>game. I think we've seen one-move games, or three- or four-move games,
>where "undo" wasn't an issue.)

You're right---the game has to be designed in a special way to
make up for the lack of undo/save/restore. But I think that it
might be possible to design a game so that player "mistakes" don't
ruin the enjoyment of the game. If you need a letter from the
President to win the game, then you're out of luck if you just used
it to light a fire. In that circumstance, you have to back up, or
start over, and if there is no convenient way to do that, you're
going to be mad.

On the other hand, to make a full-length enjoyable game in
which the player's mistakes don't ruin the game, you *either*
have to make the player's actions free of serious consequence
(if you accidentally burnt the letter, you can always
find a replacement) or else you have to give your game
a potentially enormous number of variant unfoldings.

The first option makes the player's options seem *less*
serious, rather than more.

I think that the typical way that undo/save/restore is
supposed to work is that the player is supposed to keep
his actions separate from his meta-actions in his mind.
Issuing undo/save/restore should no more spoil the flow
of the game than pausing a movie to go to the bathroom
spoils the flow of the movie. In your mind, you reconstruct
an "idealized" version that deletes the meta-actions.

I noticed that I have been completely inconsistent about
whether "you" refers to the player, or the game creator.
Oh well, you know what I mean.

--
Daryl McCullough
Ithaca, NY

Daryl McCullough

unread,
Dec 10, 2003, 6:26:38 PM12/10/03
to
Jess Knoch says...

>It is very hard to design *any* game to give sufficient hints to the
>player so that they know what to do, while sufficiently few so that
>the game is still a challenge. That's almost the crux of puzzle
>design summed up in a few short phrases. Why, oh why, would taking
>out "save" and "undo" raise the stakes for the player?

I think the idea is for the player to think "If I make the wrong
choice, I will have to live with the consequences" instead of
"If I make the wrong choice, I will have to undo, and then try
something different." If you can make the consequences for mistakes
*interesting* (as opposed to instant death, or losing the game)
then players won't be motivated to undo. I think that's easier
said than done, though.

This actually brings up an aesthetic complaint that I've had about
interactive fiction, which I voiced a few years ago in this newsgroup.
It seems difficult to impart a sense of tragedy to bad choices in IF.
In a novel, a character may make a mistake that has tragic
consequences: somebody dies, a relationship ends, etc. The character
has to live with the consequences of his mistake.

In IF, you can have tragedy by putting such a mistake into the
backstory, or by having an NPC make it. But it is really difficult
for the *player* character to make a tragic mistake, since he can
always undo/save/restore.

--
Daryl McCullough

Andrew Plotkin

unread,
Dec 10, 2003, 9:41:46 PM12/10/03
to
Here, Daryl McCullough <da...@atc-nycorp.com> wrote:
> This actually brings up an aesthetic complaint that I've had about
> interactive fiction, which I voiced a few years ago in this newsgroup.
> It seems difficult to impart a sense of tragedy to bad choices in IF.
> In a novel, a character may make a mistake that has tragic
> consequences: somebody dies, a relationship ends, etc. The character
> has to live with the consequences of his mistake.
>
> In IF, you can have tragedy by putting such a mistake into the
> backstory, or by having an NPC make it. But it is really difficult
> for the *player* character to make a tragic mistake, since he can
> always undo/save/restore.

It's most commonly done by tying the action to a short-term goal which
is clearly necessary. (Clear within the game structure, I mean.)

Not that different from horrific mistakes in horror games. Sure, the
player can avoid the zombie by not going into the graveyard. (Or by
"undo" after he enters the graveyard.) But standing around outside the
graveyard forever is not a satisfactory outcome either; it's just
boring. The story never ends if you do that.

ems...@mindspring.com

unread,
Dec 11, 2003, 4:11:57 AM12/11/03
to
da...@atc-nycorp.com (Daryl McCullough) wrote in message news:<br8a3...@drn.newsguy.com>...

> This actually brings up an aesthetic complaint that I've had about
> interactive fiction, which I voiced a few years ago in this newsgroup.
> It seems difficult to impart a sense of tragedy to bad choices in IF.
> In a novel, a character may make a mistake that has tragic
> consequences: somebody dies, a relationship ends, etc. The character
> has to live with the consequences of his mistake.
>
> In IF, you can have tragedy by putting such a mistake into the
> backstory, or by having an NPC make it. But it is really difficult
> for the *player* character to make a tragic mistake, since he can
> always undo/save/restore.

Well, as Zarf points out, only if you actually leave this freedom in
the hands of the player; it's possible to make the tragic mistake
necessary to advance the plot. But in that case -- and I've played a
couple of games that pull that trick -- I get irritated about having
been forced to do a stupid action in order to make the plot go forward
so that I could spend the rest of the storyline remedying it.

So the player character can easily make a tragic mistake, but it's
hard to get the *player* to make one, and harder still to make him
stick with that decision and not go back to change the outcome. I'm
not sure there's a good way around that, but (on the other hand) I
don't find the tragic mistake aesthetically very satisfying in many
cases. What I find more interesting than the tragic mistake is the
painful decision: things can go multiple ways, but there are costs to
all of them, and it's up to you to decide which you're going to take,
but you have a good story either way. That can be done in IF, though
I can only think of a few examples.

-- Emily

Mike Roberts

unread,
Dec 10, 2003, 8:48:53 PM12/10/03
to
"Daryl McCullough" <da...@atc-nycorp.com> wrote:
> In IF, you can have tragedy by putting such a mistake into the
> backstory, or by having an NPC make it. But it is really difficult
> for the *player* character to make a tragic mistake, since he can
> always undo/save/restore.

Not necessarily; you could always have the tragic event be an unavoidable
part of the story.

An interesting challenge would be to create a tragedy in the traditional
sense - not in the modern sense that something bad happens, but in the
traditional sense that something bad is *destined* to happen, that the
tragic outcome is inherent in a fundamental aspect of the protagonist's
character. The challenge would be to create the circumstances and the
character such that the player feels a freedom of action comparable to any
other IF, but still inevitably meets the tragic end. (You could argue that
Infidel fits the bill, but it seems to me to lack a crucial element of true
tragedy, which is the audience's ability to see the bad ending, and the
inevitability of the bad ending, from a mile away. Infidel's outcome is
pretty out of the blue, as I recall; more of a twist ending than a tragedy,
I think. Photopia might be a better example, but it doesn't have as much
freedom of action as some players might like, and its ending is more bad
luck than the inevitable result of a character flaw.)

--Mike
mjr underscore at hotmail dot ocm

Mike Roberts

unread,
Dec 10, 2003, 8:58:41 PM12/10/03
to
"Daryl McCullough" <da...@atc-nycorp.com> wrote:
> On the other hand, to make a full-length enjoyable game in
> which the player's mistakes don't ruin the game, you *either*
> have to make the player's actions free of serious consequence
> (if you accidentally burnt the letter, you can always
> find a replacement) or else you have to give your game
> a potentially enormous number of variant unfoldings.
>
> The first option makes the player's options seem *less*
> serious, rather than more.

I think that's true for action games, where the player's mistakes are more a
function of timing and physical agility, but my own experience as a player
is that "consequences" in IF don't do anything to make choices seem more
serious. Quite the opposite - once I start in with UNDO and RESTORE to roll
back PC death or lost objects or whatever, I'm just consciously and
deliberately taking apart the puzzle box to see how the plot branch tree is
constructed, and any narrative gravity is lost through the exposure of the
clockwork inside. If UNDO and RESTORE are disabled, then RESTART and an
input script accomplish the same thing, just more tediously.

> I think that the typical way that undo/save/restore is
> supposed to work is that the player is supposed to keep
> his actions separate from his meta-actions in his mind.
> Issuing undo/save/restore should no more spoil the flow
> of the game than pausing a movie to go to the bathroom
> spoils the flow of the movie.

Interesting analogy, but it doesn't quite hold for me. Pausing the movie
doesn't force me to start deconstructing it the way UNDO and RESTORE can
often do in IF. If I'm saving and restoring only so I can go to bed and
sometime later pick up where I left off, that's like pausing the movie; if
I'm repeatedly undoing to get past some timed puzzle or a conversation tree
with multiple dead-ends where the troll becomes irrevocably enraged, then
the patina of narrative over the mechanism gets worn pretty thin.

> In your mind, you reconstruct an "idealized" version
> that deletes the meta-actions.

Doesn't that argue against the notion that PC death (etc) invests decisions
with more gravity? If you edit out meta-actions and remember just the
winning storyline, then you never made anything but correct choices, so in a
sense the incorrect choices were never really possible. Or maybe the
argument is that it lends weight to the decisions by showing how dire the
wrong ones would have been: "I'm sure glad I chose the red door, because I
would have been eaten by a hungry grue if I'd picked the blue one!"

--Mike
mjr underscore at hotmail dot com

Peter Gambles

unread,
Dec 11, 2003, 6:59:52 AM12/11/03
to
>> In IF, you can have tragedy by putting such a mistake into the
>> backstory, or by having an NPC make it. But it is really difficult
>> for the *player* character to make a tragic mistake, since he can
>> always undo/save/restore.
>

>So the player character can easily make a tragic mistake, but it's


>hard to get the *player* to make one, and harder still to make him
>stick with that decision and not go back to change the outcome. I'm
>not sure there's a good way around that, but (on the other hand) I
>don't find the tragic mistake aesthetically very satisfying in many
>cases.

Is it technically possible, and / or aesthetically desirable, to modify the
UNDO command so that not everything is in fact undone? I don't mean anything
obvious, but, for example, leaving some state indicator such that certain
endings would not be accessible. Quick think: you take a decision which kills
Paul, an NPC. He would normally have appeared later on in the game to help you
get to a good ending. Even if you undo the killing, he will never reappear, so
you are heading towards less good endings. He's not dead (since you un-killed
him), just on a different branch which the player cannot now access. You
wouldn't even need to know this within the game, so the undo would appear
complete, though clearly the About or whatever should advise players of the
way the game worked. I can see problems with this with multiple undo's though.


PeterG
(author of Temple of Kaos)

aka Peter Gambles
Oxford
UK

e-mail peter....@admin.ox.ac.uk

Andrew Plotkin

unread,
Dec 11, 2003, 10:47:50 AM12/11/03
to
Here, Peter Gambles <peter....@admin.ox.ac.uk> wrote:
> In article <a69830de.03121...@posting.google.com>, ems...@mindspring.com (ems...@mindspring.com) wrote:
>
> >So the player character can easily make a tragic mistake, but it's
> >hard to get the *player* to make one, and harder still to make him
> >stick with that decision and not go back to change the outcome. I'm
> >not sure there's a good way around that, but (on the other hand) I
> >don't find the tragic mistake aesthetically very satisfying in many
> >cases.
>
> Is it technically possible, and / or aesthetically desirable, to
> modify the UNDO command so that not everything is in fact undone? I
> don't mean anything obvious, but, for example, leaving some state
> indicator such that certain endings would not be accessible.

It's not possible on the Z-machine. I stuck a Glulx feature in which
permits it (you can save data to a window of memory which
undo/restore/restart don't touch). I'm not sure about TADS.

> Quick think: you take a decision which kills Paul, an NPC. He would
> normally have appeared later on in the game to help you get to a
> good ending. Even if you undo the killing, he will never reappear,
> so you are heading towards less good endings. He's not dead (since
> you un-killed him), just on a different branch which the player
> cannot now access. You wouldn't even need to know this within the
> game, so the undo would appear complete, though clearly the About or
> whatever should advise players of the way the game worked. I can see
> problems with this with multiple undo's though.

Yes. You'll also never catch the case where the player saves, performs
the action, quits, begins a new session, and restores.

I think I would have mimesis problems. The world is now behaving
differently, without any in-world reason for it to be different. I can
see getting very frustrated -- you haven't removed the "undo" tool,
but you've made it work in a broken way. In some sense that's even
worse than removing it.

Daryl McCullough

unread,
Dec 11, 2003, 6:48:25 PM12/11/03
to
ems...@mindspring.com (Emily Short) says...

>So the player character can easily make a tragic mistake, but it's
>hard to get the *player* to make one, and harder still to make him
>stick with that decision and not go back to change the outcome. I'm
>not sure there's a good way around that, but (on the other hand) I
>don't find the tragic mistake aesthetically very satisfying in many
>cases. What I find more interesting than the tragic mistake is the
>painful decision: things can go multiple ways, but there are costs to
>all of them, and it's up to you to decide which you're going to take,
>but you have a good story either way. That can be done in IF, though
>I can only think of a few examples.

Yes, I agree. The painful decision is achievable in IF (even if it is
hard) even in the presence of undo/save/restore. The player is free
to back up and try something different, but that something different
has its own drawbacks.

But I think that tragic mistakes *can* be important elements of
some stories. King Midas wishing for the power of turning things
to gold, Sampson falling in love with Delilah, Othello believing
Iago's lies, Romeo misunderstanding Juliet's death-like sleep,
the consequences of the wishes in the "Monkey's Paw", etc. I
think that literature would be a little bit poorer if characters
never made mistakes.

ems...@mindspring.com

unread,
Dec 12, 2003, 3:31:01 AM12/12/03
to
da...@atc-nycorp.com (Daryl McCullough) wrote in message news:<bravo...@drn.newsguy.com>...

> But I think that tragic mistakes *can* be important elements of
> some stories. King Midas wishing for the power of turning things
> to gold, Sampson falling in love with Delilah, Othello believing
> Iago's lies, Romeo misunderstanding Juliet's death-like sleep,
> the consequences of the wishes in the "Monkey's Paw", etc. I
> think that literature would be a little bit poorer if characters
> never made mistakes.

Yes, I agree. On the other hand, if you're going to have a premise
that depends on the player character making a mistake, you can't offer
the player agency over that part of the story. The player can never
have the chance to negate the premise of the game.

You've got an interesting collection of examples, here. I think I'd
deal with them in different ways.

The Monkey's Paw is a no-win scenario. The point is that anything you
wish for could be turned against you; it would be easy enough to
render this as IF because there is no good answer. All the player's
choices would inevitably come out badly. The only effect he'd have on
the plot was in determining which disaster he wound up facing. King
Midas is a little less that way, and there's a little more emphasis on
the great stupidity of his specific request, but I think you could
still do something, if you like, where all wishes lead to disasters
and the only way to avoid that is not to wish for anything.

The Romeo and Juliet example, I think you could do with careful
misdirection. You wind up with something like this: everything in the
writing of the game directs the player toward a certain goal. He
pursues that goal, but discovers (only at the very end) that that was
in fact a bad idea. (See "Bliss", or, more subtly, "9:05".) It's a
bit of a gimmick, and I think, to be fair, that you'd probably want to
supply at least an implementation of what happens if the player *does*
replay and direct things the right way. But on the first playthrough,
all the way to the end, he'd probably be trying to carry out the goal
implicit in the text, because that's what players do.

Samson and Delilah... hrm. Is that really a *mistake* on Samson's
part? I mean, obviously, bad judgment call, but I'm not sure
"mistake" is the word I'd use. But here again, you give the player
every reason to pursue the goal of getting together with Delilah. He
does, obeying the pointing finger of Authorial Cues and the emotions
he is told he has, and then...

You are carrying:
no hair


Othello is perhaps the trickiest. Maybe the choice would be either to
take action (on Iago's advice) or to live with the uncertainty about
whether or not Desdemona is faithful or not. The need to come to some
sort of closure in the story might well drive the player to kill her
just to get it over with and resolved somehow.

So I think the function of the tragic mistake varies from story to
story, and how you handled it in IF would have to vary too. If the
story is really about how the protagonist copes after a tragic mistake
has already been made, then you start with that item in the past,
already done. If it's about blindness leading him to foolish
behavior, you write the game in such a way that he is misled (after
all, the player only sees what you tell him he sees). If it's about
fate being stacked against him, then you make it so that there are
options but they all lead to bad things no matter what.

-- Emily

Jess Knoch

unread,
Dec 12, 2003, 7:32:01 AM12/12/03
to
ems...@mindspring.com wrote:
> da...@atc-nycorp.com (Daryl McCullough) wrote in message
> news:<bravo...@drn.newsguy.com>...
>> But I think that tragic mistakes *can* be important elements of
>> some stories. ...Romeo misunderstanding Juliet's death-like
sleep, etc. I

>> think that literature would be a little bit poorer if characters
>> never made mistakes.
>
> The Romeo and Juliet example, I think you could do with careful
> misdirection. You wind up with something like this: everything
> in the
> writing of the game directs the player toward a certain goal. He
> pursues that goal, but discovers (only at the very end) that that
> was
> in fact a bad idea. (See "Bliss", or, more subtly, "9:05".)

The best example of this I have seen is "Failsafe". (In which, I
seem to recall, you were not allowed to save and restore.)

Still, with this or Romeo and Juliet, I would feel more tricked than
tragic, but that's the drawback of IF. Not that being cleverly
tricked into doing something is a bad thing, it's just not
necessarily very emotional. For decisions which truly affect a
player, they need to be aware of the consequences of their decision.
Heroism can be much more meaningful than an ironic misunderstanding.

> It's a
> bit of a gimmick, and I think, to be fair, that you'd probably
> want to
> supply at least an implementation of what happens if the player
> *does*
> replay and direct things the right way.

Yes -- and "9:05" is the only one of the three games mentioned that
does this satisfactorily, in my opinion.

--
Jess K.


Arthur Milliken

unread,
Dec 12, 2003, 2:56:40 PM12/12/03
to
Daryl McCullough wrote:

> You're right---the game has to be designed in a special way to
> make up for the lack of undo/save/restore. But I think that it
> might be possible to design a game so that player "mistakes" don't
> ruin the enjoyment of the game. If you need a letter from the
> President to win the game, then you're out of luck if you just used
> it to light a fire. In that circumstance, you have to back up, or
> start over, and if there is no convenient way to do that, you're
> going to be mad.

When we talk about "character games," what we're really talking about is
"immersion," or "suspension of disbelief." We want a game that draws
you in so deeply that you forget you're playing a game.

Removing undo/save/restore in order to increase "immersion" is like
shooting a fly with a machine gun, and actually has exactly the opposite
of the intended effect.

I.F. players *expect* undo/save/restore in their games, because these
features are designed to make the game more enjoyable and less tedious.
When these functions suddenly go missing, all it does is draw *more*
attention to the meta-game aspects of the piece, since players are now
faced with the horrible decision of starting over (and perhaps saving an
input script), or just abandoning the game out of sheer frustration.

If you want to discourage players from playing in a
"save/restore/save/restore" style, then you must design your game so
that a player never gets the sense that he's ruined the game for himself
just because he made one wrong step.

For example, a player can save the game before opening the basement
door. When the basement door is opened, a savage beast from the
netherworld crawls out.

Here are two scenarios:

1) Instant death. The beast rips the PC to shreds, ending the game.
Obviously, the player will want to UNDO or RESTORE, and not open the
door this time.

2) Instant *danger,* but not instant death. The PC can smell, hear, and
perhaps even see the beast, but he still has the option to slam the door
shut and run. What follows may be a very tense chase--frought with
danger, true, but ultimately survivable. I would guess that many
players wouldn't bother using RESTORE, because they would enjoy the
immersion and excitement that ensues from the chase.

When I play IF, I normally don't RESTORE unless I feel that I've somehow
made the game unwinnable or unplayable. I would guess that many players
do the same. If you design your game with a high level of "immersion,"
your player might forget that SAVE/UNDO/RESTORE even exist, and simply
try to win on the "first try." And isn't *that* really the goal?

-Arthur

Daniel Barkalow

unread,
Dec 12, 2003, 9:00:02 PM12/12/03
to
On 11 Dec 2003, Daryl McCullough wrote:

> But I think that tragic mistakes *can* be important elements of
> some stories. King Midas wishing for the power of turning things
> to gold, Sampson falling in love with Delilah, Othello believing
> Iago's lies, Romeo misunderstanding Juliet's death-like sleep,
> the consequences of the wishes in the "Monkey's Paw", etc. I
> think that literature would be a little bit poorer if characters
> never made mistakes.

A Change in the Weather is essentially based on the PC making a mistake,
and actually lets you not make the mistake. Of course, the game just ends
at that point, with an ending which is clearly satisfactory for the PC but
lacking in any sort of drama. The need for there to be a story drives the
player to make the critical mistake. Of course, I doubt that this would
work any time but the very beginning of the game, when the player is more
committed to playing a game than to reaching any particular goals.

-Iabervon
*This .sig unintentionally changed*

Reply all
Reply to author
Forward
0 new messages