Game Geographies and Dangerous Curves

56 views
Skip to first unread message

Emily Short

unread,
May 17, 2001, 2:33:15 AM5/17/01
to
Recently I started (again; I've started it before) "Dangerous Curves,"
which I have for some time considered sounded interesting just for the
thoroughness of implementation that everyone talks about. Unfortunately,
I found myself clueless about what to do, spent a long time wandering
around accomplishing absolutely nothing, and gave up in frustration. Some
of this has to do with, perhaps, a lack of plot direction, but some of it
undoubtedly has to do with my own fund of patience, so this isn't meant as
a concerted dig against the game, which may have many things going for it
I didn't get far enough to see.

What particularly irked me and contributed to my sense of futility,
however, was the sheer size of the game space and the way that movement
through the geography was handled (i.e., even when you use short cuts like
>GO TO PLACE, regardless of whether you are in the car or on foot, you
move one location per turn. This is good in the sense that it allows you
to see each location and in that you don't miss things by bipping past
them at high speed... except that, of course, in real life if you're in
your car hastening somewhere you might very well not get a good glimpse of
the neighborhood. But I digress.) In any case, I was annoyed by the fact
that I didn't have a very good sense of how the game space was put
together, never acquired that sort of internal map that allows you to play
rapidly, and generally felt lost most of the time. And I was annoyed that
I found myself annoyed, because I think, in general, that there's
something cool about having the patience to implement a big roomy city
with lots of space to move around in.

So here's the question. Supposing that one wants a game with a large
amount of game space available at one time -- no locked doors controlling
access to major areas, no impassible rivers or moats full of seething acid
-- how does one make it playable-in? I had a few thoughts, which were
more or less as follow:

1) Extra-game feelies: give the player a physical map. Wishbringer did
this, and it was cute. Seems potentially helpful. Maybe not the most
self-contained and elegant of solutions, however.

2) Always give the player a particular place to go. Would this work?
Suppose I have a city implemented with (glulx and spare time being on my
side) *many* hundreds of locations. Blocks and blocks. Just Looking
Around Thoroughly is not a play option, so instead, always have the player
have a concrete chore in a specific place. He can wander around if he
wants, but he really does have a place to go. In this regard, I think
Dangerous Curves might've worked if I had only had a little more of a
sense of what I was supposed to be doing at any given time.

3) Have >GO TO work faster (I have seen this done and approve), and/or
install some kind of transportation system.

or of course

4) Don't Do This in the First Place.

I guess the important point here is that early game design often seems to
have centered around using geography to shape the plot -- you could figure
out your progress in the game by how much of the territory you'd been
allowed to see so far -- whereas this method would require some external
mode of plot-enforcement if it is not just to become a giant city-sim.
[Which has already Been Done, or so I hear.]

Thoughts? Games that dealt with this well/badly? Small cheeses,
tastefully wrapped?

ES

--
Emily Short
http://emshort.home.mindspring.com/index.htm

Electronic

unread,
May 17, 2001, 9:47:55 AM5/17/01
to
On Thu, 17 May 2001 02:33:15 -0400, ems...@mindspring.com (Emily
Short) wrote:

>3) Have >GO TO work faster (I have seen this done and approve), and/or
>install some kind of transportation system.

I think this is one of the most valuable solutions. I opted for this
option for the game I'm currently trying to write. The player will
move from one distant area to another by way of taking a taxi, being
presented with a menu of possible destinations at the time he does so,
etc.

I'm not sure I like a gargantuan map where the majority of rooms are
there just for the sake of scenery. I feel it's a bit like stretching
a game's durability artificially. I mean, realistic simulation is
nice, except if it is to the detriment of gameplay. And wandering
around in a zillion places before finding something to do is
off-putting, as far as I'm concerned.

Just my two kopeks.

Lucian P. Smith

unread,
May 17, 2001, 10:30:28 AM5/17/01
to
Emily Short (ems...@mindspring.com) wrote in <emshort-1705...@user-2inik11.dialup.mindspring.com>:

: So here's the question. Supposing that one wants a game with a large


: amount of game space available at one time -- no locked doors controlling
: access to major areas, no impassible rivers or moats full of seething acid
: -- how does one make it playable-in?

: 1) Extra-game feelies: give the player a physical map.

: 2) Always give the player a particular place to go.

: 3) Have >GO TO work faster (I have seen this done and approve), and/or


: install some kind of transportation system.

: 4) Don't Do This in the First Place.

I just this week read the answer to this question, and now I can't for the
life of me remember where. Maybe over at gamasutra.com, though I can't
now find it.

At any rate, the answer given was: make each area thematically
consistent. Or, to put it another way, 'chunk' the sections of your game
into easily distinguishable sub-areas.

Perhaps your problem with 'Dangerous Curves' was that, not unlike Houston,
there was no zoning. So you couldn't say, "I need to go to the
library. Ah, that's probably back in the middle-class area of
town." Each building was dropped near the next one in a haphazard fashion
(with the exception of the west street, with the bar, strip club, and
seedy theater. That kind of think carried further might have helped. As
a side question: did you have problems visualizing where *those* places
were?)

This can be accomplished in a variety of ways. One method is to create
'constriction points' between your areas. So the only way to go from the
house to the garden is through the back door, and the only way to go to
the park down the street is to whistle for your dog, who leads you there
in a cut-scene. These can be natural constriction points, as above, or
artifical, like in 'Jigsaw', 'So Far', or (ahem) 'Edifice'.

Another method is to use your writing prowess to keep the sub-areas
distinct. 'So Far' is probably the best example of that--each of the
areas felt (for me) very different from each other. To this day, I can
still clearly picture the grassy plain, the abandonded castle and creeping
fungus, and the silent circus. The fact that each area had its own
*smell* was probably important to that, too.

Even 'Zork II' managed to convey a good sense of distinct areas, using
both techniques to some extent. The carousel room served as the
'constriction point' between the sub-areas, and even though it was
possible to find other ways between them, you still knew 'OK, the dragon
and bank area are to the NW, while the alice-in-wonderland section are to
the SE'. And each area had its own feel to it, too, even though things
weren't quite as thematically consistent as 'So Far'. I can probably
remember *those* sections, too. Let's see:

NE/E: Opening, topiary, gazebo
SE: 'Alice'
S: Menhir, diamond maze
SW: Wizard's Workshop
W: Volcano (well, and grue repellent)
NW: Dragon/Bank
N: Ledge, door

Yup. It's been years. Still got it. Zork I, on the other hand, I'd have
to go back to the map for the main area, though I do remember the areas
with the maze, portrait room, hades, coal mine, and the river. The middle
was a jumble, and I slavishly followed my map. Even the round room didn't
help, possibly because it wasn't really a constriction site.

Hmm, other examples. Hey, 'Small World' is yet another way to do
it. There, there's a *gradient* of change as you circle the globe. So
each area might not be that different from the areas immediately adjacent
to it, but would be most distinct from the area on the other side of the
circle. Contrast this with, say, Phlegm, where each room is utterly
distinct from every other room, and there's no logical connection between
them. (That's the point, of course, but it certainly doesn't convey a
sense of space.) Now I'm on a comp '96 kick. Reverberations! There's a
cityscape game. I don't remember what the geography here was really like,
but, then, I don't remember having problems with it either. I don't think
there were actually that many buildings you could explore, and instead,
each building was its own sub-area. It wasn't hard to distinguish in my
head the pizza place from city hall, oddly enough.

Any other examples? Good or bad? Did they use these techniques, or
others?

-Lucian

J.D. Berry

unread,
May 17, 2001, 10:44:31 AM5/17/01
to
>= Original Message From ems...@mindspring.com (Emily Short) =

>So here's the question. Supposing that one wants a game with a large
>amount of game space available at one time -- no locked doors controlling
>access to major areas, no impassible rivers or moats full of seething acid

>-- how does one make it playable-in? I had a few thoughts, which were
>more or less as follow:

Your points are accurate. I think together they form an overall picture.
(Which always seems to be 1)it depends and 2) it can work if done
correctly.)

>1) Extra-game feelies: give the player a physical map. Wishbringer did
>this, and it was cute. Seems potentially helpful. Maybe not the most
>self-contained and elegant of solutions, however.

Here's where commercial games have the advantage. I'll open the package and
find a colorful map, an instruction manual formatted in a
genre-compelementary
style, the actual CD and some survey-type cards that I instantly throw away.

The feelies are not only "right there" accessible, they are attractive.
And,
besides, I probably don't even have my computer running.

Compare this to downloading html files or printing out a map. Such items
are
not right there. I do have my computer running. Give me a game, now,
please.
Freeware IF has a few minutes to attract the player, and this must
accomplished through its initial gameplay alone.

>2) Always give the player a particular place to go. Would this work?
>Suppose I have a city implemented with (glulx and spare time being on my
>side) *many* hundreds of locations. Blocks and blocks. Just Looking
>Around Thoroughly is not a play option, so instead, always have the player
>have a concrete chore in a specific place. He can wander around if he
>wants, but he really does have a place to go. In this regard, I think
>Dangerous Curves might've worked if I had only had a little more of a
>sense of what I was supposed to be doing at any given time.

When I'm playing any IF, I need to know my purpose (Djinni pun not
intended).
I may not know what's going on OVERALL, but if I don't know what I need to
do,
in general, to advance the story I'm going to get frustrated. In IF, this
purpose is the second thing after the initial story hook that determines if
a
player will stay with the game.

"I can't figure out what I need to do!" is fundamentally different from "I
don't know HOW to do what I need to do!" The former makes the player feel
helpless and alienates him from the story. The latter empowers the player
with a feeling that he CAN do something, he'll just have to be creative.

This ties into trust. In DC's case, we trust Irene as an accomplished
writer,
but perhaps our personal immersion in this particular work of hers isn't
connecting--WE are the ones not getting it. And if, as we delve further, we
still aren't getting it, we lose hope.


>3) Have >GO TO work faster (I have seen this done and approve), and/or
>install some kind of transportation system.

This is dependent on how the environment has been already established in the
player's mind WITHOUT it. Ideally, the player has been walked through the
setting and the environment is felt, before the running.

>or of course


>
>4) Don't Do This in the First Place.
>

Blending with number 3, above, I think large scale can be done. But I think
it has to be done incrementally.

The Ultima series is a good example. Assuming, you've never played
before...
You get a map. Most likely you'll look at it--it's cloth, it's colorful,
it's
cool. You'll plant the overall continent and sub-islands in your head. You
won't associate many feelings to any of the places, but you'll know the
overall world.

The game starts you in a particular, "closed" location. It's not closed, of
course, but you're likely to explore just the local area. To get a feel,
but
also because you'll likely die if you just wander outside without equipment
and knowledge. Also, the NPCs you encounter often give you purpose--some
long-term, some short.

The world never feels intimidating because you learn it as you go. By the
time you learn the continent and realize it would suck to have to walk it
again, faster transportation becomes available: horses, ships, balloons,
moon
gates, etc...

IF is much smaller scale, but I think the Ultima principle applies.

>I guess the important point here is that early game design often seems to
>have centered around using geography to shape the plot -- you could figure
>out your progress in the game by how much of the territory you'd been
>allowed to see so far -- whereas this method would require some external
>mode of plot-enforcement if it is not just to become a giant city-sim.
>[Which has already Been Done, or so I hear.]
>

>Thoughts? Games that dealt with this well/badly? Small cheeses,
>tastefully wrapped?
>

See above. ;-)

Jim

J.D. Berry

unread,
May 17, 2001, 10:46:17 AM5/17/01
to

Lucian P. Smith

unread,
May 17, 2001, 10:58:14 AM5/17/01
to
Lucian P. Smith (lps...@rice.edu) wrote in <9e0na4$glg$1...@joe.rice.edu>:
: Emily Short (ems...@mindspring.com) wrote in <emshort-1705...@user-2inik11.dialup.mindspring.com>:

:
: : So here's the question. Supposing that one wants a game with a large
: : amount of game space available at one time -- no locked doors controlling
: : access to major areas, no impassible rivers or moats full of seething acid
: : -- how does one make it playable-in?

: I just this week read the answer to this question, and now I can't for the


: life of me remember where. Maybe over at gamasutra.com, though I can't
: now find it.
:
: At any rate, the answer given was: make each area thematically
: consistent. Or, to put it another way, 'chunk' the sections of your game
: into easily distinguishable sub-areas.

Hey, I found the article. And... it was you! What are you doing asking
questions which you've already answered? ;-)

Specifically:

"Another useful technique is to create obvious divisions or symmetries in
the game world, something to give the player a handle on distinct sets of
locations. Metamorphoses has a very symmetrical map, because it suited
the symbolism of the world to make it that way. One of my other works in
progress is less symmetrical, but it has sections, what I think of as
neighborhoods; each has distinct thematic features of its own."

(The article in its entirety, for those playing at home, can be found at
http://interactfiction.about.com/games/interactfiction/library/design/ucsetting1.htm

-Lucian

josh g

unread,
May 17, 2001, 11:03:12 AM5/17/01
to
Emily Short <ems...@mindspring.com> wrote:
> So here's the question. Supposing that one wants a game with a large
> amount of game space available at one time -- no locked doors controlling
> access to major areas, no impassible rivers or moats full of seething acid
> -- how does one make it playable-in? I had a few thoughts, which were
> more or less as follow:

> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.


I really like this option. It brings to mind my old favorite, A Mind
Forever Voyaging. Exploring the city in AMFV was well worth the effort,
but I would've given up on it early on without a map. Some people don't
mind mapping out a game world as they play - me, I like the lazy way out.

If your game is in a hard text-only format, this does put extra burden on
the player (who has to download and print the map out themselves.) But
if you're making an HTML TADS or glulx game, there's nothing stopping you
from having a graphical map in-game.


> 2) Always give the player a particular place to go. Would this work?


Seems I'm coming back to my favorite Infocom game as an example again.
Giving a player small objectives to motivate some exploration can be
very effective. This was pretty blatant in AMFV - you were tasked with
observing and recording a list of mundane events in the city. But it
had the desired effect - the player had to explore, had to observe, and
in the end would have come across all the richness of every nook and
cranny in the game world.


> 3) Have >GO TO work faster (I have seen this done and approve), and/or
> install some kind of transportation system.


I like and don't like this. Now my mind drifts to (of all things) a
favorite gameboy RPG (which shall remain nameless to protect whatever
iota of respect is offered me here). Having a quick transportation
system to avoid repetitive travel was a godsend. But you always had
to travel through things the slow, painful, interesting way first.

I think this is an important concept to note. If you give the player a
quick route from A to B, they're very likely to see that as the path
they're "supposed" to take. If you've created a fully described and
thought-out bit of game world in between A and B, don't sell yourself
short and have the majority of players skip it without knowing better.

Or if nothing else, give players a map so that they know there's
something interesting between A and B. The curious will explore and
enjoy, and the impatient will at least have a sense of context when
they take the quick route.


- josh g.
jo...@playground.net

Emily Short

unread,
May 17, 2001, 12:34:09 PM5/17/01
to
In article <9e0p7g$5...@dispatch.concentric.net>, josh g
<jo...@zork.plover.net> wrote:

> Emily Short <ems...@mindspring.com> wrote:
> > 3) Have >GO TO work faster (I have seen this done and approve), and/or
> > install some kind of transportation system.
>
> I like and don't like this. Now my mind drifts to (of all things) a
> favorite gameboy RPG (which shall remain nameless to protect whatever
> iota of respect is offered me here). Having a quick transportation
> system to avoid repetitive travel was a godsend. But you always had
> to travel through things the slow, painful, interesting way first.
>
> I think this is an important concept to note. If you give the player a
> quick route from A to B, they're very likely to see that as the path
> they're "supposed" to take.

Hmm, good point. I think one thing about Dangerous Curves was that I
rapidly realized (by walking around my own office building) that
traditional exploration techniques weren't going to work, so I started
using GO TO all the time, even for places I hadn't been before. (And
partly because I didn't know where the places in question were, whereas my
PC did know.)

Emily Short

unread,
May 17, 2001, 12:36:52 PM5/17/01
to
In article <9e0ou6$glg$2...@joe.rice.edu>, lps...@rice.edu (Lucian P. Smith)
wrote:

> Lucian P. Smith (lps...@rice.edu) wrote in <9e0na4$glg$1...@joe.rice.edu>:
> : Emily Short (ems...@mindspring.com) wrote in
<emshort-1705...@user-2inik11.dialup.mindspring.com>:
> :
> : : So here's the question. Supposing that one wants a game with a large
> : : amount of game space available at one time -- no locked doors controlling
> : : access to major areas, no impassible rivers or moats full of seething acid
> : : -- how does one make it playable-in?
>
> : I just this week read the answer to this question, and now I can't for the
> : life of me remember where. Maybe over at gamasutra.com, though I can't
> : now find it.
> :
> : At any rate, the answer given was: make each area thematically
> : consistent. Or, to put it another way, 'chunk' the sections of your game
> : into easily distinguishable sub-areas.
>
> Hey, I found the article. And... it was you! What are you doing asking
> questions which you've already answered? ;-)

Well uh...

But I'm thinking of a big setting this time. Really big. Not sure this
would be enough. Though I suppose chunking combined with a plot that
makes the player explore certain segments first might help, even if he
isn't actually prevented from going to the further locations.

David Samuel Myers

unread,
May 17, 2001, 2:09:56 PM5/17/01
to
Emily Short <ems...@mindspring.com> wrote:
> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.

I liked the way there was an in-game map in Christminster for the first
chunk of the game, followed by no map for the 75%-80% of the game after
that. This gives the player some comfort for a while.

> 3) Have >GO TO work faster (I have seen this done and approve), and/or
> install some kind of transportation system.

I like this.

> Thoughts? Games that dealt with this well/badly? Small cheeses,
> tastefully wrapped?

Mmm... port salut...

-david

Robb Sherwin

unread,
May 17, 2001, 2:26:56 PM5/17/01
to
On Thu, 17 May 2001 12:34:09 -0400, ems...@mindspring.com (Emily
Short) wrote:
>Hmm, good point. I think one thing about Dangerous Curves was that I
>rapidly realized (by walking around my own office building) that
>traditional exploration techniques weren't going to work, so I started
>using GO TO all the time, even for places I hadn't been before. (And
>partly because I didn't know where the places in question were, whereas my
>PC did know.)

I remember exploring Knight Orc in much the same way. That game came
with a paper map, so rather than try to explore the landscape in a
traditional way at first, I'd just identify things on it that looked
kind of cool and use the GO TO command. The downside to that is that I
never really got a good sense of the game's geography for the amount
of time I spent playing it. While I have the underground empire of
Zork I forever tattooed upon my brain, I just recall the geographical
setting for Knight Orc to be rough and foggy.

It's funny, I've been wrestling with the question of implementing GO
TO functionality in the thing I'm working on now. I can't envision the
task of programming that code to be anything other than a total
nightmare, though. For a rectangular game world (one where all rooms
from like X1, Y1 to X30, Y30 are implemented) I guess it would be
straightforward to do, but I'm not sure how I would tackle getting
from one place to another if there are oddly-angled streets, walls,
and such involved.

Robb


=-=-=-=-=-
Robb Sherwin, Fort Collins CO
Knight Orc Home Page: www.joltcountry.com
Reviews From Trotting Krips: ifiction.tsx.org

Guilherme De Sousa

unread,
May 17, 2001, 2:27:39 PM5/17/01
to
Emily Short <ems...@mindspring.com> wrote in message
news:emshort-1705...@user-2inik11.dialup.mindspring.com...

> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.

I`ve still got my battered Ultima maps somewhere. I always feel like I waste
a lot of time finding my way around games even when there aren`t a lot of
locations. I was thinking of including a map in my next game, possibly
within the game itself or through appropriate in-game items. I`ve also
played around with producing simple tile based maps in HTML TADS for an
automap. What do people think about automaps (if they could use one on their
interpretor)?

> 2) Always give the player a particular place to go. Would this work?
> Suppose I have a city implemented with (glulx and spare time being on my
> side) *many* hundreds of locations. Blocks and blocks. Just Looking
> Around Thoroughly is not a play option, so instead, always have the player
> have a concrete chore in a specific place

I think having a clear purpose or job to do is very important in large
games. I like the idea of a game with lots of extra locations, alternative
pathways and possibly mini adventures alongside the main story / pathway.
The player could explore as much of this optional material as they like.

> 3) Have >GO TO work faster (I have seen this done and approve), and/or
> install some kind of transportation system.

Considering that there may be dozens of locations en route I think rather
than seeing a list of locations as the player moves through them it might be
interesting to provide a description of the journey possibly through some
type of text generation. For example:

"The carriage pulled away from the house turning first north onto Main
Street and then east into a maze of unfamiliar side streets. It was early
evening when the carriage finally came to a stop outside my hotel..."

I`d still like to have the option of a location by location description
though and the ability to walk to the distant location on foot.

> Thoughts? Games that dealt with this well/badly? Small cheeses,
> tastefully wrapped?
>
> ES

I quite liked "Sherlock" where you used to catch Hansom cabs and trains
around various locations - after several turns of "You are travelling
through the streets of London..." you arrived at your destination. You only
learnt the names of new locations as the game progressed. You couldn`t walk
between distant locations but I think it made the game feel larger than it
really was.

Guilherme
ghde...@yahoo.co.uk


Iain Merrick

unread,
May 17, 2001, 2:46:19 PM5/17/01
to
Emily Short wrote:
[...]

> So here's the question. Supposing that one wants a game with a large
> amount of game space available at one time -- no locked doors controlling
> access to major areas, no impassible rivers or moats full of seething acid
> -- how does one make it playable-in? I had a few thoughts, which were
> more or less as follow:
>
> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.

As I mentioned in my review, this is something I'd like to see more of.

Not only would it make things easier for me, the player, but I think it's
also a good way of presenting lots of information that the player
character ought to know off-by-heart already.

Oh, and maps are just inherently cool. I suspect that most people who
happened to read Tolkien while young and impressionable would agree.

>2) Always give the player a particular place to go.

This just sounds a bit like 'design the game well', though perhaps I'm
misunderstanding.



>3) Have >GO TO work faster (I have seen this done and approve), and/or
>install some kind of transportation system.

I like this in principle, but I think it's likely to lead players to
ignore 99% of your carefully-crafted city. In which case...

>4) Don't Do This in the First Place.

...you might as well do this. The comp game _Madame L'Estrange_ had
instant GO TO instead of a complete simulation of the city, and I think it
worked very well (that aspect of the game, anyway).

But then I also liked _Dangerous Curves_, once I got into it. I think it
would work even better if GO TO was removed but you were given a clear
map.

--
Iain Merrick
i...@spod-central.org

josh g

unread,
May 17, 2001, 2:54:02 PM5/17/01
to
Robb Sherwin <bea...@zombieworld.com> wrote:
> It's funny, I've been wrestling with the question of implementing GO
> TO functionality in the thing I'm working on now. I can't envision the
> task of programming that code to be anything other than a total
> nightmare, though. For a rectangular game world (one where all rooms
> from like X1, Y1 to X30, Y30 are implemented) I guess it would be
> straightforward to do, but I'm not sure how I would tackle getting
> from one place to another if there are oddly-angled streets, walls,
> and such involved.


Off the top of my head, I'd say this isn't something you need to work
out blindly. It falls neatly under the topic of network routing
algorithms, which is a well-studied and popular field of computer
science / engineering. The concepts and algorithms will apply, no
matter if the nodes in your network are internet routers or IF rooms.

Of course this would be a much more useful reply if I had some decent
URLs to offer you as a starting point, but I don't know of any. And
the only tidbit I can remember off the top of my head from my general
CS algorithms course was Dijkstra's (sp?) algorithm for finding the
shortest path through a network. There's something to punch into
Google at least ... if no one's offered more useful info later today,
I'll see if I can dig up some old textbooks and do some web searching
myself.


- josh g.
jo...@playground.net

Magnus Olsson

unread,
May 17, 2001, 3:01:29 PM5/17/01
to
In article <9e16oa$4...@dispatch.concentric.net>,

josh g <jo...@zork.plover.net> wrote:
>Robb Sherwin <bea...@zombieworld.com> wrote:
>> It's funny, I've been wrestling with the question of implementing GO
>> TO functionality in the thing I'm working on now. I can't envision the
>> task of programming that code to be anything other than a total
>> nightmare, though. For a rectangular game world (one where all rooms
>> from like X1, Y1 to X30, Y30 are implemented) I guess it would be
>> straightforward to do, but I'm not sure how I would tackle getting
>> from one place to another if there are oddly-angled streets, walls,
>> and such involved.

>And


>the only tidbit I can remember off the top of my head from my general
>CS algorithms course was Dijkstra's (sp?) algorithm for finding the
>shortest path through a network.

There's at least one program (in TADS, IIRC) on the IF archive that
uses Dijkstra's algorithm to implement a "go to" command.

So if your game isn't *too* large (less than several thousand
locations, I'd guess), you can code the map in the traditional IF way,
as rooms connected by passages, and then let Dijkstra guide you.

--
Magnus Olsson (m...@df.lth.se, m...@pobox.com)
------ http://www.pobox.com/~mol ------

Andrew Plotkin

unread,
May 17, 2001, 4:02:12 PM5/17/01
to
Emily Short <ems...@mindspring.com> wrote:

> So here's the question. Supposing that one wants a game with a large
> amount of game space available at one time -- no locked doors controlling
> access to major areas, no impassible rivers or moats full of seething acid
> -- how does one make it playable-in?

My experience, from _Hunter, in Darkness_: some players *will* go
exploring if they are permitted to. Furthermore, if the exploring is
long and fruitless, they will hold it against you.

(I'm not criticizing such players, BTW. I'm documenting a demonstratable
trend of player-reaction, not particular people who played _HID_. *I*
demonstrate this reaction, sometimes, when playing other people's games.)

The only moral I can draw is, if you want to exert control by giving
the player particular destinations, make the quests pretty damn blunt.
If the player misses a clue, and doesn't know where to go -- even once
-- you've probably lost that player.

--Z

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
* George W. Bush was elected with just five votes.

OKB -- not okblacke

unread,
May 17, 2001, 4:39:30 PM5/17/01
to
>So here's the question. Supposing that one wants a game with a large
>amount of game space available at one time -- no locked doors controlling
>access to major areas, no impassible rivers or moats full of seething acid
>-- how does one make it playable-in? I had a few thoughts, which were
>more or less as follow:
>
>1) Extra-game feelies: give the player a physical map. Wishbringer did
>this, and it was cute. Seems potentially helpful. Maybe not the most
>self-contained and elegant of solutions, however.

I am in favor of this if the feelies really create a feeling. I've seen
games where the map was just a bland, dull sheet of paper with lines on it.
If, however, the feelies actually feel like they could have been objects in the
game world that were magically manifested in reality, that's cool.

>2) Always give the player a particular place to go. Would this work?
>Suppose I have a city implemented with (glulx and spare time being on my
>side) *many* hundreds of locations. Blocks and blocks. Just Looking
>Around Thoroughly is not a play option, so instead, always have the player
>have a concrete chore in a specific place. He can wander around if he
>wants, but he really does have a place to go. In this regard, I think
>Dangerous Curves might've worked if I had only had a little more of a
>sense of what I was supposed to be doing at any given time.

I would revise this to say "always give the player AT LEAST ONE particular
place to go". Personally, if I were writing a game of this size, I wouldn't
want to force the player to play it in a linear fashion. Likewise, if I were
playing a massive game, I would be annoyed to find that, at any one time, only
one location (or small clump of locations) was meaningful, and the others were
effectively scenery.

So maybe the player has one place which is obviously a good destination,
but there are other places which he can find out about through careful
observation, or doing certain things, or even ones that are meaningful even
though no indication of this is given in the game.

>3) Have >GO TO work faster (I have seen this done and approve), and/or
>install some kind of transportation system.

I like this.

I agree with josh that this can manipulate players into taking the quick
and easy GO TO instead of the more intricate path. A way around this, however,
is to make the GO TO not a meta verb but a part of the game world (like the
Hansom cabs someone mentioned, or the taxis, or bus(s)es, or whatever). Then
you can restrict the GO TO in a way that fits the game world -- for example,
not every street corner is going to have a bus stop. You could even
selectively restrict it depending on the current state of the game. For
example, if the player is at the Town Square and wants to take a taxi to the
Docks, but there's something important going on en route that he'll miss if he
takes the taxi, you can have the taxi drivers go on strike and force him to
walk.

>
>or of course
>
>4) Don't Do This in the First Place.

I'd say this is a good choice if your game is built around something other
than its size. If, however, a large world is an integral part of your vision
of the game, I'd say it's worth it to take one of the other options.

In general, I'd say that if you're going to go the simulationist route by
making a large, detailed game world, you should continue to simulate. Don't
just simulate the buildings and the streets -- simulate the buses and/or the
taxis and/or the futuristic moving sidewalks. I think a fully realized game of
this type would incoporate these concerns not only into its implementation but
into the game world itself.

Of course, this could be a life's work. But with Glulx and spare time,
almost anything is possible :-).

--OKB (Bren...@aol.com) -- no relation to okblacke

"Do not follow where the path may lead;
go, instead, where there is no path, and leave a trail."
--Author Unknown

Lucian P. Smith

unread,
May 17, 2001, 5:47:50 PM5/17/01
to
Emily Short (ems...@mindspring.com) wrote in <emshort-1705...@user-2inim64.dialup.mindspring.com>:
: > Lucian P. Smith (lps...@rice.edu) wrote in <9e0na4$glg$1...@joe.rice.edu>:

: > : At any rate, the answer given was: make each area thematically
: > : consistent. Or, to put it another way, 'chunk' the sections of your game
: > : into easily distinguishable sub-areas.

: But I'm thinking of a big setting this time. Really big. Not sure this


: would be enough. Though I suppose chunking combined with a plot that
: makes the player explore certain segments first might help, even if he
: isn't actually prevented from going to the further locations.

OK, I'm going to move the discussion to graphical games, but I think the
point still holds.

My wife and I are nearing the end of playing 'Planescape: Torment'. In
it, there are several large maps, and you can get from one to the next
by walking. After you have visited all such maps in one area, when you
leave the edge of one map, you then get a picture of the world map, and
can click on whichever area you want to go to.

Semantically, it was pretty easy for us to remember which region-map had
which building we wanted to go to in it. Again, each section (in
general) was pretty well thematically linked, or at least had a big
recognizable landmark, so you could say, 'Oh, that's by the Mortuary' or
whatever. Logistically, it was then easy to go there in a few clicks
(once you had been there before, of course).

Because it was an RPG, it had the additional benefit of making it
*dangerous* to explore too far afield. You wanted to make the most of
each map because you had cleared out the baddies in it, and you might
not have the resources to move on to the next.

IF doesn't have fighting, of course, but there's no reason you couldn't
put some sort of resource management system in to the game to make
different areas slowly accessible. Spiritwrak did this with subway
tokens, but was criticized because these were expendable. Tickets for the
red line, blue line, etc., maybe, and give the PC little to no cash. A
car with a limited amount of gas, and have the player not 'want' to stray
further than X blocks by foot (er, that could be annoying, if the place
you wanted to go to was a short distance out of your
'range'. Sub-sections, again, could help--walk anywhere in the 'upper
west side', but take the car to 'across the tracks'.)

At any rate, more musings...

-Lucian

Stephen Granade

unread,
May 17, 2001, 7:50:46 PM5/17/01
to
ems...@mindspring.com (Emily Short) writes:

This happened to me as well, and short-circuited the normal
world-building that goes on in my head. For smaller or more
self-contained games, I soon have a map in my head. For larger, more
sprawling game I draw a map, and in the process of drawing that map I
get an idea of the general layout of the world. In Dangerous Curves I
used >GO TO a lot, and then never built up that map. I let my player
do the navigating and never watched too closely what was where.

Clearly there are gradations: in some games, I never build up a good
mental image of the game world. However, the chunking that Lucian
described elsewhere in this thread helps. I may not know exactly where
everything lives, but I know in what section most locations live, and
how the sections relate to each other. That way, I can go to a
specific section of the game and then wander around as necessary until
I remember where I'm going.

Stephen

--
Stephen Granade | Interested in adventure games?
sgra...@phy.duke.edu | Visit About Interactive Fiction
Duke University, Physics Dept | http://interactfiction.about.com

Matthew Russotto

unread,
May 17, 2001, 8:02:01 PM5/17/01
to
In article <9e16oa$4...@dispatch.concentric.net>,

josh g <jo...@zork.plover.net> wrote:
>Robb Sherwin <bea...@zombieworld.com> wrote:
>> It's funny, I've been wrestling with the question of implementing GO
>> TO functionality in the thing I'm working on now. I can't envision the
>> task of programming that code to be anything other than a total
>> nightmare, though. For a rectangular game world (one where all rooms
>> from like X1, Y1 to X30, Y30 are implemented) I guess it would be
>> straightforward to do, but I'm not sure how I would tackle getting
>> from one place to another if there are oddly-angled streets, walls,
>> and such involved.
>
>
>Off the top of my head, I'd say this isn't something you need to work
>out blindly. It falls neatly under the topic of network routing
>algorithms, which is a well-studied and popular field of computer
>science / engineering. The concepts and algorithms will apply, no
>matter if the nodes in your network are internet routers or IF rooms.
>
>Of course this would be a much more useful reply if I had some decent
>URLs to offer you as a starting point, but I don't know of any. And
>the only tidbit I can remember off the top of my head from my general
>CS algorithms course was Dijkstra's (sp?) algorithm for finding the
>shortest path through a network.

Dijkstra's algorithm is quite simple, and I've been known to run it by
hand when playing certain IF games (like those from Phoenix). It's
basically a breadth-first search from your starting point, stopping
when you hit the destination.
--
Matthew T. Russotto russ...@pond.com
"Extremism in defense of liberty is no vice, and moderation in pursuit
of justice is no virtue."

Ricardo Dague

unread,
May 17, 2001, 10:14:13 PM5/17/01
to

Emily Short <ems...@mindspring.com> wrote in message
news:emshort-1705...@user-2inik11.dialup.mindspring.com...
> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.

How about a map object in the game? Looking up different locations would
describe paths to them, and maybe give other useful info. (Matter of fact,
if anyone wants to implement this in the next Art Show, I wouldn't mind.)

Such as:
]X MAP
It's the map you drew of this area. You are now in the Deep Ravine.

]CONSULT MAP ABOUT DAMP CAVE
You can go there by going down, up, east and east (through the Reservoir
South, Deep Canyon and Dam).

>
> 2) Always give the player a particular place to go. Would this work?
> Suppose I have a city implemented with (glulx and spare time being on my
> side) *many* hundreds of locations. Blocks and blocks. Just Looking
> Around Thoroughly is not a play option, so instead, always have the player
> have a concrete chore in a specific place. He can wander around if he
> wants, but he really does have a place to go. In this regard, I think
> Dangerous Curves might've worked if I had only had a little more of a
> sense of what I was supposed to be doing at any given time.

This is the best solution, in my opinion. I'd rather be given a certain
place to be and a certain thing to do. That would give me a reason to
examine a location and the objects in it and to try different verbs and
different actions that might lead to achieving something.

The biggest impediment, for me anyway, to playing a game is my attention
span. I'd rather be given a specific task and some obvious ways to achive
it, and when that's done I can see where the plot goes to next. If the game
just sits there waiting for me, I'll probably switch it off.

> Thoughts? Games that dealt with this well/badly? Small cheeses,
> tastefully wrapped?

Fooey. Gimme a big ole hunk of cheese any day. Lemme carve off a slab
myself.

--Ricardo

jjkc

unread,
May 18, 2001, 12:43:05 AM5/18/01
to
This makes me thing of Riven and Myst. Many people (myself included)
felt that they were large worlds which were fun to explore. But one
also knew at any given time what you were supposed to be working on.
But I can't place my finger on why this worked for me. Anyone else feel
the same way?

Hmm, one thing was the fact that once you had reached a desitination you
could get there faster. But that wasn't the only reason.

-Jim

Jason Melancon

unread,
May 18, 2001, 12:46:44 PM5/18/01
to
On 17 May 2001 21:47:50 GMT, lps...@rice.edu (Lucian P. Smith) wrote:

> My wife and I are nearing the end of playing 'Planescape: Torment'.

> [snip]


> Semantically, it was pretty easy for us to remember which region-map had
> which building we wanted to go to in it. Again, each section (in
> general) was pretty well thematically linked, or at least had a big
> recognizable landmark, so you could say, 'Oh, that's by the Mortuary' or
> whatever. Logistically, it was then easy to go there in a few clicks
> (once you had been there before, of course).

That's odd, I felt unending frustration trying to remember which
building was where in the spire city (the beginning of the game; I
forget it's name). I wasted a lot of time wandering around looking
for old quests to add a completed step to. And that was only four
screens or so, with user-defined, on-screen map markers. :-) I did
finish it, though, and loved it.

--
Jason Melancon

Fraser Wilson

unread,
May 18, 2001, 4:53:38 PM5/18/01
to
"Guilherme De Sousa" <ghde...@yahoo.co.uk> writes:

> I quite liked "Sherlock" where you used to catch Hansom cabs and trains
> around various locations - after several turns of "You are travelling
> through the streets of London..." you arrived at your destination. You only
> learnt the names of new locations as the game progressed. You couldn`t walk
> between distant locations but I think it made the game feel larger than it
> really was.

"The Dame Was Loaded" used a similar system. Getting into your car
showed you a map of the city with the locations you could drive to,
and as the game progressed more and more of it opened up. This gave
an impression of size, with no tedious walking around, and it's a
technique that would translate well to text IF, as Sherlock shows.
I'd lose the travelling turns though.

Fraser.

Urbatain

unread,
May 18, 2001, 8:12:48 PM5/18/01
to
Hi all!

First of all, i'm spanish, so sorry by my bad english.

ems...@mindspring.com (Emily Short) wrote in message news:<emshort-1705...@user-2inik11.dialup.mindspring.com>...

>
> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.
>

What ever you choose, this is a good thing to add a game. It can be
pretty, maybe you must add yours web's pictures in the ZIPs of the
games.

> 2) Always give the player a particular place to go. Would this work?
> Suppose I have a city implemented with (glulx and spare time being on my
> side) *many* hundreds of locations. Blocks and blocks. Just Looking
> Around Thoroughly is not a play option, so instead, always have the player
> have a concrete chore in a specific place. He can wander around if he
> wants, but he really does have a place to go. In this regard, I think
> Dangerous Curves might've worked if I had only had a little more of a
> sense of what I was supposed to be doing at any given time.
>

Well, you can do a thing like Snowball of level9: A lot of automatic
generated rooms but if you have a place to go, it won't be a boring
thing have this huge map.

> 3) Have >GO TO work faster (I have seen this done and approve), and/or
> install some kind of transportation system.
>

I don't like this... is the worst thing of the last games of L9. Think
about Knight Orc, it's right but the command RUN TO and GO TO, simply
break the literature, and when you are advanced in the game, It
transform in a simply game of "go to the place". This is the kind of
feeling that Knight Orc provides me.

> or of course
>
> 4) Don't Do This in the First Place.
>
> I guess the important point here is that early game design often seems to
> have centered around using geography to shape the plot -- you could figure
> out your progress in the game by how much of the territory you'd been
> allowed to see so far -- whereas this method would require some external
> mode of plot-enforcement if it is not just to become a giant city-sim.
> [Which has already Been Done, or so I hear.]
>

It's important that this city-sim, don't be a great void: a lot of
locs with nothing to do. So you can add sub-stories o sub-plots while
the player is walking from place to place.

It's important to be consistent in the distances, for example:

Your are moving in a house.

You are in the kitchen.

>go out

You are in the foyer

>go out

Your in front of your home. To the north in the distance you can see
the city park

>north

You are in the park


PUAJ!!!! great error!!!

Be careful to write literature when you change the relative distances.

>north

You walk then long street for a long time. It's a good afternoon to
walk in this town. At least you arraive to the park.

(It's only an idea. Don't take into account my bad english literature
;))

So, I recomend you:

1) make a pretty map in a picture.

2) make a big city-sim, with a lot of messages for entertaiment the
prayer if he wish to take a walk.

3) use the plot of the game to control that the prayer could be
walking wasting time all the game.

Byes.

Urbatain

Sam Thursfield

unread,
May 19, 2001, 9:40:04 AM5/19/01
to
I'm afraid I'm going to have to show how much I admire the on-screen
mapping in Zork Zero, Arthur and probably Journey and Shogun (I
haven't played the last two). It should be possible under GLULX.

golrien.cjb.net

John Colagioia

unread,
May 21, 2001, 9:09:33 AM5/21/01
to
Urbatain wrote:

> Hi all!
> First of all, i'm spanish, so sorry by my bad english.

Bah. One day, I'll tell you about the time I had to train two Venezuelan gentlemen on our software. Hint: I don't speak
Spanish, and they didn't speak English...

[...]

> Well, you can do a thing like Snowball of level9: A lot of automatic
> generated rooms but if you have a place to go, it won't be a boring
> thing have this huge map.

That's actually a fairly interesting idea--dynamically generate the city through an algorithm, rather than by hand-coding.
Then, the valiant explorer will always have someplace new to look at. The problem, though, is making this seem like it belongs
in the game, rather than room descriptions that look like a failed "Mad Libs" experiment:

[Adjective] [Noun] [Type of Shop]
You are standing outside of the [superlative] [same Type of Shop] on [algorithmically-generated address]. It's closed.

[...]

> It's important that this city-sim, don't be a great void: a lot of
> locs with nothing to do. So you can add sub-stories o sub-plots while
> the player is walking from place to place.

This would be the lovely thing about good, autonomous NPCs, I think. Scatter them throughout the city as "incidental"
characters, who are all pursuing their own goals--which you may help or hinder, if you should happen to find them and become
interested. Then, the city quite literally comes alive with activity.

[...]

> >north
> You are in the park
> PUAJ!!!! great error!!!
> Be careful to write literature when you change the relative distances.
> >north
> You walk then long street for a long time. It's a good afternoon to
> walk in this town. At least you arraive to the park.
> (It's only an idea. Don't take into account my bad english literature
> ;))

No, you're absolutely right. The effect created by tiny "incidental" comments--almost "micro cut-scenes," if you will--are
incredible. A security chime whenever you enter an expensive shop, a comment about the long distance you have just walked, and
random comments from passers-by all combine for a great atmosphere in a game.

[...]


PTN

unread,
May 23, 2001, 12:38:19 AM5/23/01
to
"Emily Short" <ems...@mindspring.com> wrote in message
news:emshort-1705...@user-2inik11.dialup.mindspring.com...

> So here's the question. Supposing that one wants a game with a large


> amount of game space available at one time -- no locked doors controlling
> access to major areas, no impassible rivers or moats full of seething acid
> -- how does one make it playable-in?

To take the question out of the theoretical, this is something I've had to
address in my game, 1893: A World's Fair Mystery. It's still being beta
tested, so I'm just now finding out how well I've resolved this particular
issue. Out of the four hundred or so locations in the game, only two dozen
are not available for the player to enter right away.

Partially, I wanted an open playing area on purpose. Too often, I play IF in
which you only have access to five or ten rooms at a time, then you pass the
next 'gate' or puzzle, then the next ten rooms are available, and so on. The
amount of choices you have as a player are fairly limited in this context,
and if you get stuck on a 'gate' puzzle, you're done.

With a more open architecture, getting stuck on one puzzle means that the
player simply needs to walk away from it for a while and look for another
one.

To address your ideas:


> 1) Extra-game feelies: give the player a physical map. Wishbringer did
> this, and it was cute. Seems potentially helpful. Maybe not the most
> self-contained and elegant of solutions, however.

I think the map is a good idea, and for this game, such a map is available
on its website, along with addtional background information. Note it isn't a
room by room map, but more of a general map so you know where things are,
more or less.

> 2) Always give the player a particular place to go.

To begin with, the player has a few simple tasks that they can perform
should they wish to, which takes them to a couple different locations. After
that, the player has nothing to go on except for a handful of clues. The
website 'feelies' also contain additional tips on how to explore the game
environment.

> 3) Have >GO TO work faster (I have seen this done and approve), and/or
> install some kind of transportation system.

In addition to being able to walk most everywhere, there is also a train, a
ferry boat, and three other less traditional means of more rapid
transportation from one location to another. But you have to find these and
figure out how they work, first.

> 4) Don't Do This in the First Place.

Well, I kept on thinking I wasn't going to. But the story just seemed to
call for it, and I couldn't think of any other way to do it justice. In many
ways, the geography is the story.

So, has this implementation been successful? Well, as the game is still in
early Beta, it's hard to say. Certainly an open geography makes a game
harder than it normally would be, and it doesn't, in and of itself, make a
game better. But I think it can, and in some cases, should, be done to
fulfill the potential of some story environments.

If you want to look at this case study some more, the website is up at

http://www.illuminatedlantern.com/1893/index.html

but it isn't public -- to get in, the ID and Password are both: 1893

(and if anyone would like to join the betatest, please let me know, as I
could always use some more testers.)

-- Peter


Kevin Bracey

unread,
May 29, 2001, 10:14:45 AM5/29/01
to
In message <a53b46df.01051...@posting.google.com>
sam.thu...@btinternet.com (Sam Thursfield) wrote:

It should be possible under V6 :)

--
Kevin Bracey, Principal Software Engineer
Pace Micro Technology plc Tel: +44 (0) 1223 518566
645 Newmarket Road Fax: +44 (0) 1223 518526
Cambridge, CB5 8PB, United Kingdom WWW: http://www.pace.co.uk/

Reply all
Reply to author
Forward
0 new messages