IF Map Design Question

13 views
Skip to first unread message

Charles

unread,
Sep 1, 2005, 11:52:35 AM9/1/05
to
When designing map locations, do y'all prefer keeping the maps as
logical grids, or is that unimportant?

In some of the early Infocom games, I've notice that the maps are not
made into a perfectly logical grid, which leads one to sometimes say
south, then north will not necessarily bring them back to the point
where they were.

This always irks me as it makes making a map more difficult and feels
awkward to me. So, I wanted to see what y'all thought about this
design issue. Thanks for the feedback.

Bob

unread,
Sep 1, 2005, 12:06:17 PM9/1/05
to
I like logical grids better. But it's ok for me, if there are e.g. four
rooms surrounding a house and some more in the house. I tend to make
more than one map in these cases.

Bob

Damian Dollahite

unread,
Sep 1, 2005, 7:16:46 PM9/1/05
to

You're asking two different questions here. To your original question,
whether rooms need to be laid out on a regular grid like a chessboard,
the answer is absolutely not; in fact, do everything you can to
discourage that impression. The 8-way compass navigation will naturally
tend to make players feel like they're moving on a chessboard, as the
author your job is to break down that impression as much as possible and
make the world feel varied and organic, and thus, realistic.

Your second and third paragraphs, however, are talking about straight
vs. crooked room connections, which has nothing to do with whether the
rooms are a laid out in a chessboard pattern or not. The answer to that
question is, avoid crooked room connections at all costs. They irk just
about everyone as much as they do you.

--
Ryukage

Rexx Magnus

unread,
Sep 2, 2005, 4:45:08 AM9/2/05
to
On Thu, 01 Sep 2005 15:52:35 GMT, Charles scrawled:

If I have a location that does not immediately connect in the reverse
direction, I'll often explain why as the player leaves the first location.

--
http://www.rexx.co.uk

To email me, visit the site.

Richard Bos

unread,
Sep 4, 2005, 1:14:05 PM9/4/05
to
Damian Dollahite <ryu...@aol.com> wrote:

> Charles wrote:
> > When designing map locations, do y'all prefer keeping the maps as
> > logical grids, or is that unimportant?
> >
> > In some of the early Infocom games, I've notice that the maps are not
> > made into a perfectly logical grid, which leads one to sometimes say
> > south, then north will not necessarily bring them back to the point
> > where they were.
> >
> > This always irks me as it makes making a map more difficult and feels
> > awkward to me. So, I wanted to see what y'all thought about this
> > design issue. Thanks for the feedback.
>
> You're asking two different questions here. To your original question,
> whether rooms need to be laid out on a regular grid like a chessboard,
> the answer is absolutely not; in fact, do everything you can to
> discourage that impression. The 8-way compass navigation will naturally
> tend to make players feel like they're moving on a chessboard, as the
> author your job is to break down that impression as much as possible and
> make the world feel varied and organic, and thus, realistic.

Agreed; except where the chessboard impression is appropriate. If you're
treading the street grid of Manhattan, for example, or if you're in a
modern, faceless, all-square office building.

> Your second and third paragraphs, however, are talking about straight
> vs. crooked room connections, which has nothing to do with whether the
> rooms are a laid out in a chessboard pattern or not. The answer to that
> question is, avoid crooked room connections at all costs. They irk just
> about everyone as much as they do you.

Beg to differ, rather strongly. Making the map feel natural is much more
important than symmetrical connections. Avoiding asymmetrical
connections, especially at all cost, can mean your map feels contrived
or unnatural.

For example, if you're on a side lawn to the east of a building, I'd
expect to have to go north to reach the back yard. To go back, the most
natural direction would be east. This is asymmetrical, but to "repair"
this, you'd have to apply tricks merely for the sake of symmetry, none
of which feel right.
You certainly can't make going south from the back yard go to the east
lawn. Not only would that mean having no way inside the house from the
back, but what would you do about the west lawn?
You can't reasonably make northwest and southeast the directions to go,
because, from a realistic point of view, they'd take you straight
through the corner of the house. I'd possibly allow nw and se as
additional directions, for the sake of the player who insists on
straight connections, but in would depend on whether these are also
needed to enter, say, a kitchen door. In any case, they would be
additional; they'd never be able to replace the more natural north and
east.
You could, perhaps, add another connecting location, the "northeast
corner of the garden"; but to add another, likely empty, location,
merely for the sake of straight connections, would feel very artificial.

In short, do _try_ to have most connections symmetrical; but if the
natural layout of your map asks for it, do not be afraid to have some
crooked ones. The real world isn't perfectly straight, either.

Richard

James Mitchelhill

unread,
Sep 4, 2005, 1:45:44 PM9/4/05
to
On Sun, 04 Sep 2005 17:14:05 +0000, Richard Bos wrote:

<snip>


> Beg to differ, rather strongly. Making the map feel natural is much more
> important than symmetrical connections. Avoiding asymmetrical
> connections, especially at all cost, can mean your map feels contrived
> or unnatural.

To me, asymmetrical connections feel much more contrived than a grid-like
map ever could. There doesn't seem much point maintaining a false
verisimilitude. The world itself is not neatly broken down into rooms and
I never think "How do I get to there? Ah, E.SE.N.NW.N.UP.E!" No amount of
design is ever going to make me think that the map isn't an abstraction,
but I'm happy with it being an abstraction, because most of the time I can
ignore it.

Asymmetrical connections are jarring. They mean I have to think about the
map which shatters any immersion I was feeling.

<snip>


> In short, do _try_ to have most connections symmetrical; but if the
> natural layout of your map asks for it, do not be afraid to have some
> crooked ones. The real world isn't perfectly straight, either.

But I don't have to think about getting around in the real world and I
don't think in compass directions. And that's the problem - compass
directions (the abstraction of personal mapping used in IF) imply symmetry.

--
James Mitchelhill
ja...@disorderfeed.net
http://disorderfeed.net

Andrew Plotkin

unread,
Sep 5, 2005, 1:07:51 AM9/5/05
to
Here, James Mitchelhill <ja...@disorderfeed.net> wrote:
> On Sun, 04 Sep 2005 17:14:05 +0000, Richard Bos wrote:
>
> <snip>
> > Beg to differ, rather strongly. Making the map feel natural is much more
> > important than symmetrical connections. Avoiding asymmetrical
> > connections, especially at all cost, can mean your map feels contrived
> > or unnatural.
>
> To me, asymmetrical connections feel much more contrived than a grid-like
> map ever could.

How do you visualize a round room, then? I mean a game area which is
divided up into "North of the White House", "East of the White House",
"South...", and "West..."

The convention from the beginning has been that east from #1 takes you
to #2, and north from #2 takes you to #1. And so on. Yes, diagonal
directions can *also* work, but they're not always implemented -- and
they can't be used at all when you divide a game area into *eight*
segments.

--Z

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
I'm still thinking about what to put in this space.

J. Robinson Wheeler

unread,
Sep 5, 2005, 5:02:10 AM9/5/05
to
James Mitchelhill wrote:

> Richard Bos wrote:
> > Making the map feel natural is much more important than
> > symmetrical connections.
>
> Asymmetrical connections are jarring. They mean I have to think about
> the map which shatters any immersion I was feeling.


What I tend to do when writing a game is walk around the map a lot,
back and forth. Especially in an area where the map connections
are not exactly gridlike, when I catch myself trying to connect
from one location to the next with a direction that seemed intuitive,
but which results in a "You can't go that way," I change the coding
(and perhaps the description) to go with the flow of what felt right.


--
J. Robinson Wheeler Games: http://raddial.com/if/
JRW Digital Media Movie: http://thekroneexperiment.com/

James Mitchelhill

unread,
Sep 5, 2005, 1:44:07 PM9/5/05
to
On Mon, 05 Sep 2005 05:07:51 +0000, Andrew Plotkin wrote:

> Here, James Mitchelhill <ja...@disorderfeed.net> wrote:
>>
>> To me, asymmetrical connections feel much more contrived than a grid-like
>> map ever could.
>
> How do you visualize a round room, then? I mean a game area which is
> divided up into "North of the White House", "East of the White House",
> "South...", and "West..."

I tend not to visualise, actually. I have fairly poor spatial reasoning.
This may be why it's particularly jarring to me.

> The convention from the beginning has been that east from #1 takes you
> to #2, and north from #2 takes you to #1. And so on. Yes, diagonal
> directions can *also* work, but they're not always implemented -- and
> they can't be used at all when you divide a game area into *eight*
> segments.

Diagonal directions would be my preferred implementation.

I'm not entirely sure what you mean by dividing a game area into eight
segments. Without resorting to crude ascii maps, I'm thinking of a similar
set up to the house example, but with two rooms to a side. But I'm not
entirely sure why diagonals wouldn't work with this, so I suspect that
what I'm thinking isn't what you mean. (I can also think of eight rooms
arranged three to a side, like a 3x3 square with no centre square, but
this is a simpler set-up, so I suspect that's wrong, also.

there...@yahoo.com

unread,
Sep 5, 2005, 7:58:38 PM9/5/05
to
> If I have a location that does not immediately connect in the reverse direction, I'll often explain why as the player leaves the first location.

THERE you go. I don't care how tricky the map gets, so long as the
game gives me a modicum of warning that I'm getting twisted around.

In real life . . . When do we ever find ourselves in a room, having no
idea which way we came in?


~JD

Andrew Plotkin

unread,
Sep 5, 2005, 8:21:04 PM9/5/05
to
Here, James Mitchelhill <ja...@disorderfeed.net> wrote:
> On Mon, 05 Sep 2005 05:07:51 +0000, Andrew Plotkin wrote:
>
> > Here, James Mitchelhill <ja...@disorderfeed.net> wrote:
> >>
> >> To me, asymmetrical connections feel much more contrived than a grid-like
> >> map ever could.
> >
> > How do you visualize a round room, then? I mean a game area which is
> > divided up into "North of the White House", "East of the White House",
> > "South...", and "West..."
>
> I tend not to visualise, actually. I have fairly poor spatial reasoning.
> This may be why it's particularly jarring to me.
>
> > The convention from the beginning has been that east from #1 takes you
> > to #2, and north from #2 takes you to #1. And so on. Yes, diagonal
> > directions can *also* work, but they're not always implemented -- and
> > they can't be used at all when you divide a game area into *eight*
> > segments.
>
> Diagonal directions would be my preferred implementation.

If you're explicitly in a circular hallway, then diagonal directions
are a contrivance: the path *really does* lead straight east and west
from "North of the White House". (And starts to curve immediately.)

> I'm not entirely sure what you mean by dividing a game area into eight
> segments.

"North of the White House", "Northeast of the White House"...

If you can arrange it as a square, the connections can be straight.
But if you're explicitly modelling a circular hallway, the result is
weird. Unlikely to satisfy everyone, most likely.

Here, there...@yahoo.com wrote:
> > If I have a location that does not immediately connect in the

> > reverse direction...


>
> THERE you go. I don't care how tricky the map gets, so long as the
> game gives me a modicum of warning that I'm getting twisted around.

Yes, I certainly want that warning.

Mantar, Feyelno nek dusa

unread,
Sep 5, 2005, 9:16:43 PM9/5/05
to
On Mon, 05 Sep 2005 16:58:38 -0700, therealjdc wrote:

> In real life . . . When do we ever find ourselves in a room, having no
> idea which way we came in?

Try spelunking. ;-)


James Mitchelhill

unread,
Sep 5, 2005, 9:19:09 PM9/5/05
to
[The level of quotes have got a little confusing, so... context: I have
been arguing that asymmetrical connections on a map are confusing. Andrew
has presented a couple of examples where, he argues, asymmetrical
connections make more sense.]

On Tue, 06 Sep 2005 00:21:04 +0000, Andrew Plotkin wrote:
> If you're explicitly in a circular hallway, then diagonal directions are
> a contrivance: the path *really does* lead straight east and west from
> "North of the White House". (And starts to curve immediately.)

OK... I guess the way I process directional descriptions in IF is a little
different to some (most?) people. I tend to think of the direction given
as a composite of the direction of the path to the next location. So, to
me, "a path leads east" does not necessarily mean that I begin walking in
that direction, but that the location in that direction is east of where I
am now.

I think this is close to how we navigate in real life, too. When someone
gives directions saying: "Get on the M27 heading South" (the M27 being a
motorway - replace this with the major road of your choice), I don't
necessarily expect the motorway to head directly south. As in the circular
path example, it may well head east at first, then curve to the south. But
even if, by the compass, it's heading east, I'd still be southbound,
because that's where I'll end up.

The diagonal directions *are* a contrivance, but they're one that I'm
happy to indulge in. Besides, the whole directional system in IF is (in
general) a contrivance. Where we differ, I think, is the the level of
abstraction we feel it should provide. I prefer the directional system to
represent the position of locations at the global level, while you (and
others) prefer it to represent the direction of exits at room level.

Going back to your example, would this feel 'wrong' to you if it was
implemented using diagonal directions, and would this break immersion?

Andrew Plotkin

unread,
Sep 5, 2005, 11:06:57 PM9/5/05
to
Here, James Mitchelhill <ja...@disorderfeed.net> wrote:
> [The level of quotes have got a little confusing, so... context: I have
> been arguing that asymmetrical connections on a map are confusing. Andrew
> has presented a couple of examples where, he argues, asymmetrical
> connections make more sense.]
>
> On Tue, 06 Sep 2005 00:21:04 +0000, Andrew Plotkin wrote:
> > If you're explicitly in a circular hallway, then diagonal directions are
> > a contrivance: the path *really does* lead straight east and west from
> > "North of the White House". (And starts to curve immediately.)
>
> OK... I guess the way I process directional descriptions in IF is a little
> different to some (most?) people. I tend to think of the direction given
> as a composite of the direction of the path to the next location. So, to
> me, "a path leads east" does not necessarily mean that I begin walking in
> that direction, but that the location in that direction is east of where I
> am now.

That is different from how I navigate in IF, yes. (I guess you're
thinking of it as "the *closest* location to the east", since there
will in general be several.)

> The diagonal directions *are* a contrivance, but they're one that I'm
> happy to indulge in. Besides, the whole directional system in IF is (in
> general) a contrivance.

Sure.

> Where we differ, I think, is the the level of
> abstraction we feel it should provide. I prefer the directional system to
> represent the position of locations at the global level, while you (and
> others) prefer it to represent the direction of exits at room level.
>
> Going back to your example, would this feel 'wrong' to you if it was
> implemented using diagonal directions, and would this break immersion?

If *only* diagonal directions are available, it feels a little wrong.
I could play the game, but I think I would wind up imagining the area
as a diamond-shaped corridor, in contradiction to the descriptive
text.

nbake...@charter.net

unread,
Sep 6, 2005, 12:42:51 AM9/6/05
to

Aw! You speak of the lost art of Text Adventures! That's why the
current offerings of the IF scene are free. Take a look at the grand
classics. In Crystal Caverns you would type "go building" or "go
valley" but if you forced the user onto the stupid compas/grid system
they would have to type "n, n, n, w, n, w, w, w, n, e, n, w, w" just to
get down into the valley and by then they turn off the computer out of
frustration. If you want the user to feel like they are playing a
game, then add some fancy three-D graphics and make a DOOM clone. If
you want your work to be regarded as "literature" then write a *story*
-- perhaps make use of the directions to choose different endings (If
you want to kill the mayor press "n", if you don't, press "s") -- your
primary focus should always be on telling the story and making it
*interactive* but this does not mean that you have to make it into a
game with a crazy obsession with directions, objects, and placement.

Nathan.

Reply all
Reply to author
Forward
0 new messages