What makes a good puzzle?

10 views
Skip to first unread message

Kathleen Fischer

unread,
May 17, 1996, 3:00:00 AM5/17/96
to

As I stumble along learning Inform (and making good progress thanks to all your
helpful advice) I have my eye on the future ... ie, once I get past moving from
room to room, sharpening my pencils and planting herb gardens I would like to
add a puzzle two... so, my question is... what makes a good one?

I know I don't like completely illogical ones (like getting through the tree's
in one of the Zorks, don't remember which one), and I don't like overly
complicated ones (like the infamous bell/candle/book).

I loved the slug and dustbunnies (from one of the zorks). They were cute and
not too difficult. Is difficult good? Do your puzzles have to be hard to make
the game worthwhile?

But this is just what I liked. On the downside, these puzzles are generally
"silly" in nature and I was hoping for something a little more enlightened (or
at least real world).

So, back to the question... does any one have any opinions on what makes a good
puzzle?


Thanks for your time,
Kathleen

--
// Kathleen Fischer
// kfis...@greenhouse.llnl.gov
// *** "Don't stop to stomp ants while the elephants are stampeding" ***


Brad O`Donnell

unread,
May 17, 1996, 3:00:00 AM5/17/96
to Kathleen Fischer

Kathleen Fischer wrote:

> Is difficult good? Do your puzzles have to be hard to make
> the game worthwhile?
>
> But this is just what I liked. On the downside, these puzzles are generally
> "silly" in nature and I was hoping for something a little more enlightened (or
> at least real world).
>
> So, back to the question... does any one have any opinions on what makes a good
> puzzle?
>

There are no real guidelines for puzzle-making, as anyone will tell
you, but here are a couple of things to think about.

--In what way is the puzzle difficult? Does it involve lots of steps?(I
LOVE those, as long as you don't kill me for screwing up a step.) Does
it use a really strange diction? (Arrgh! I don't know what a XXX is, so
how can I possibly know what to do with it?) Is it tedious? ('A maze of
twisty little passages, all alike. Whoever thought that one up should
be SHOT. E-Mail me, whoever you are, so I can hunt you down like a
dog!)

--Does the puzzle involve doing something at a specific time? Under
certain conditions? If so, make sure these conditions show up often
enough so that I might see what's going on with them.

--Please don't kill me for walking North. Or South. Or Entering any
location, really.

--Only make me do math once or twice a game. (Interactive Fiction, not
Interactive Math Test!)

-- All of the above (Except the math, really) can be ignored if "Fair
Warning" is given.

--What is Fair Warning? Well, you're likely to have to rewrite several
puzzles at the playtesting stage before you can be sure...


A little presumptuous of me, but you asked for it.


Brad O'Donnell

--"My sig is just quotes of mine; makes it so
much easier, and no angry mobs mail me in
the middle of the night."

Cardinal Teulbachs

unread,
May 18, 1996, 3:00:00 AM5/18/96
to

Kathleen Fischer <kfis...@greenhouse.llnl.gov> wrote:

>I know I don't like completely illogical ones (like getting through the tree's
>in one of the Zorks, don't remember which one), and I don't like overly
>complicated ones (like the infamous bell/candle/book).

The problem is usually not that the puzzle is illogical. It almost
always makes some sense once you know the solution. The problem is
that many puzzles use references that are too obscure or fail to
provide enough clues so as to give a player of reasonable intelligence
and education a fair shot at solving them.

As for puzzles with purely random or brute force solutions (such as
ones requiring you to type in every number between one and a hundred
until you hit on the right one), I don't consider those puzzles.
Luckily, you don't see too many of that kind, though.

>I loved the slug and dustbunnies (from one of the zorks). They were cute and

>not too difficult. Is difficult good? Do your puzzles have to be hard to make
>the game worthwhile?

No. If the puzzle is going to be hard, there had better be a big
payoff at the end in terms of enjoyment--but I guess it depends upon
the type of game you're making, too. Some i-f works exist almost
entirely for the sake of the puzzles. They're designed for the crowd
that likes logic problems and reference games. Others are designed to
be more of a story, with puzzles thrown in mostly as roadblocks. These
are designed with the "literary" reader in mind. A hard puzzle might
be enjoyable to a fan of the first type of game, whereas it might do
nothing but frustrate a fan of the second type who just wants to get
on to the next chapter. Conversely, a bunch of easy puzzles inserted
into a complex storyline might only bore diehard puzzleheads who don't
really give a jot who killed Cock Robin, anyway. The point is: decide
who your audience is going to be and write for them. There's no doubt
a certain audience that would be happy with no puzzles at all, just as
there's an audience for Freefall, which is nothing but one big puzzle,
and just as there's an audience for every mixture in between. You're
writing the game; make it be what you think it should be.

>But this is just what I liked. On the downside, these puzzles are generally
>"silly" in nature and I was hoping for something a little more enlightened (or
>at least real world).

What? Silly a downside? Preposterous!

>So, back to the question... does any one have any opinions on what makes a good
>puzzle?

Make the elements of the puzzle as universally recognizable as
possible. Don't use things that require specialized knowledge to
recognize and understand. Puzzles involving bells, books, and candles
are better than puzzles involving hadrons and muons if, for instance,
the player must know a particular decay mode of the K-particle in
order to figure the thing out or get the joke (funny stuff, that).

Related to this: know the difference between "hard" and "obscure." The
K-particle puzzle is not hard, it's obscure. An example of the
converse might be if you took one of those logic problems you find in
crossword magazines and coded it up in a game, with the known
propositions given as clues. The puzzle would then quailify as hard,
in that it would require the player to mentally step through a tight
and lengthy logical argument, but it wouldn't be obscure, inasmuch as
everything necessary is given and no special knowledge is required to
work out the solution.

Make the puzzle consistent with the tone and environment of the story.
If you're writing a horror story, for instance, don't throw in a
comedy-gag puzzle. The dust bunnies, as they now stand, would be
inappropriate in a game like Theatre. Similarly, a soup can blockade
might fit well enough into a game like Zork, but it has no place in
anything pretending to some relationship with reality.

These are my big three, anyway; my two cents, as it were.

--Cardinal T

I mean, what the hell kind of villain thwarts the hero's
progress with soup cans in the kitchen pantry?
--Russ Bryan

Cardinal, I follow up your post in the hopes that some
day I too will be quoted in your sig.
--Matthew Amster-Burton

Hey! This isn't what I said! What'd you do with my
quote?
--Bonni Mierzejewska


Magnus Olsson

unread,
May 18, 1996, 3:00:00 AM5/18/96
to

In article <4nj98i$q...@paraguay.it.earthlink.net>,

Cardinal Teulbachs <card...@earthlink.net> wrote:
> Puzzles involving bells, books, and candles
>are better than puzzles involving hadrons and muons if, for instance,
>the player must know a particular decay mode of the K-particle in
>order to figure the thing out or get the joke (funny stuff, that).
>
>Related to this: know the difference between "hard" and "obscure." The
>K-particle puzzle is not hard, it's obscure.

...unless, of course, it involved figuring out the transition amplitudes
for that particular decay mode, in which case it would be both hard
and obscure. :-)

As has been stated before, it is very hard to postulate general rules
about what makes good puzzles and what doesn't. Here is one, though:
unless you're aiming at a very limited audience, any puzzle requiring
intimate knowledge of the theory of electroweak interactions is probably
not a very good idea. (Another one for your .sig, Cardinal? :-) ).

On a more serious note, anybody interested in what constitutes a good
puzzle should consult Graham Nelson's "The Craft of Adventure", as well
as Kevin Wilson's "Whizzard Guide to Adventure Writing". Both are
available from the if-archive and could be considered required reading.

--
Magnus Olsson (m...@df.lth.se)

Cardinal Teulbachs

unread,
May 19, 1996, 3:00:00 AM5/19/96
to

m...@marvin.df.lth.se (Magnus Olsson) wrote:
>...unless, of course, it involved figuring out the transition amplitudes
>for that particular decay mode, in which case it would be both hard
>and obscure. :-)

True. One must posit the existence of demijons in order to even begin
such a task, and we know what slippery creatures those are...

>As has been stated before, it is very hard to postulate general rules
>about what makes good puzzles and what doesn't.

Agreed. I'd say the wisest, most general rule of thumb a person can
follow is probably just to consider puzzles he/she has liked/disliked,
and then use that as a kind of template in creating and judging
his/her own. If a puzzle was enjoyable, reflect upon why it was
enjoyable; if not, reflect upon why not, and then either emulate or
avoid the things embodied in these newfound principles as appropriate.

>Here is one, though:
>unless you're aiming at a very limited audience, any puzzle requiring
>intimate knowledge of the theory of electroweak interactions is probably
>not a very good idea. (Another one for your .sig, Cardinal? :-) ).

Holy smokes! You guys are killing me with this signature thing <g>.
Can someone out there collect all these pithy i-f proverbs and publish
them as an e-book or something? I can't keep up. (What you said is
quite true nonetheless).

>On a more serious note, anybody interested in what constitutes a good
>puzzle should consult Graham Nelson's "The Craft of Adventure", as well
>as Kevin Wilson's "Whizzard Guide to Adventure Writing". Both are
>available from the if-archive and could be considered required reading.

Ditto. They're both very good.

Richard G Clegg

unread,
May 20, 1996, 3:00:00 AM5/20/96
to

Kathleen Fischer (kfis...@greenhouse.llnl.gov) wrote:
: So, back to the question... does any one have any opinions on what makes
: a good puzzle?

(Grin) You might better have asked does any one NOT have any opinions on
what makes a good puzzle - I think there are as many answers as there are
players - ranging from "more complex the better" to "no puzzles at all
thanks".

I get the most buzz from a situation where I can see early on what needs
doing but the actual "doing" is difficult.

Extremely mild spoilers for Stationfall, Day of the Tentacle,
Starcross and HitchHikers.



One of the best examples was "using the explosive" in Stationfall - which
needed you to obtain various objects - most of which were difficult
to get anyway but clearly on display. I hate the type of game where I
have no clear idea of what I'm trying to achieve next. Particularly
frustrating (and therefore enjoyable - to me anyway) is putting something
that is obviously necessary in clear view but unobtainable - the items
in the vending machine for example which you see almost as soon as you
get on the station but can't get until near the end.

A similar example was the "winning the beauty contest" in Day of the
Tentacle. Once you'd worked out that your goal was to win the contest
you needed to work out how to win the various sub-sections - it took a few
hours for me to actually manage to win it but it was SO satisfying once it
was done.

The infamous babel fish puzzle was similar (you knew you needed the
fish but actually getting it was a nightmare) but perhaps a bit more
arbitrary.

The type of puzzles which I think can be most irritating are those which
require specialist knowledge. Starcross was perhaps the worst culprit -
without some (admittedly basic) scientific knowledge you'd have no way
to work out the right way to get the atmosphere on (although you could
use trial and error). I worked it out pretty quickly but I could imagine
someone without basic chemistry really puzzling over all those dots.
A similar problem in the same game I actually did solve with trial and
error (ten columns - one of which will open the hatch the others will
lock it permenantly - they're described in such a way that you can work
out which one to press if you realise it's a solar system layout - to
my shame I didn't). I guess the Zork "Bell book and candle" puzzle
would also be covered by this since if you'd never heard the phrase "bell
book and candle" it would be illogical in the extreme.

--
Richard G. Clegg There ain't no getting round getting round
Dept. of Mathematics (Network Control group) Uni. of York.
email: ric...@manor.york.ac.uk
www: http://manor.york.ac.uk/top.html


Damien Neil

unread,
May 20, 1996, 3:00:00 AM5/20/96
to

On 17 May 1996 15:20:24 GMT, Kathleen Fischer <kfis...@greenhouse.llnl.gov> wrote:
>As I stumble along learning Inform (and making good progress thanks to all your
>helpful advice) I have my eye on the future ... ie, once I get past moving from
>room to room, sharpening my pencils and planting herb gardens I would like to
>add a puzzle two... so, my question is... what makes a good one?

Well, this comes down largely to personal preference. My favourite
puzzles are what I refer to as `system puzzles': the player is faced
with a situation, a set of actions which can modify the situation,
and a desired result. A perfect example is the Atomic Chihuahua puzzle
in Hollywood Hijinx[1].

The key to such puzzles is that they cannot be solved simply by finding
the correct object (lock-and-key puzzles), or by guessing the correct
action (guess the verb puzzles). The player needs to examine the
situation, experiment with different actions, and eventually understand
the nature of the problem at hand in order to solve it.

For a particularly evil[2] example of this form of puzzle, play Andrew
Plotkin's "A Change in the Weather".


>I loved the slug and dustbunnies (from one of the zorks). They were cute and
>not too difficult. Is difficult good? Do your puzzles have to be hard to make
>the game worthwhile?

Absolutely not! One of my favourite games is _Wishbringer_, which is
one of the easiest games ever made by Infocom. A puzzle need not be
difficult to be satisfying. An interesting concept or good writing
can make even the most basic of lock-and-key puzzles entertaining.
Consider the troll in _Wishbringer_ for a good example[3].

- Damien


[1] For those who have not played HH: this puzzle involves the manipulation
of a B-movie monster model contained within a glass case. There are
several buttons which control the model (causing it to walk, breathe
fire, and the like), and several hazards the creature must evade.

[2] I mean this only in the most complimentary sense, of course. :>

[3] Valiently avoiding spoilers. If you've played _Wishbringer_, you
know what I am talking about. If you haven't, I envy you.
_Wishbringer_ was the first Infocom game I ever played, and I was
not ready to enjoy it fully at the time...


Kenneth Fair

unread,
May 25, 1996, 3:00:00 AM5/25/96
to

In article <4nlep7$k...@news.lth.se>, m...@marvin.df.lth.se (Magnus Olsson) wrote:

>In article <4nj98i$q...@paraguay.it.earthlink.net>,
>Cardinal Teulbachs <card...@earthlink.net> wrote:

>> Puzzles involving bells, books, and candles
>>are better than puzzles involving hadrons and muons if, for instance,
>>the player must know a particular decay mode of the K-particle in
>>order to figure the thing out or get the joke (funny stuff, that).
>>
>>Related to this: know the difference between "hard" and "obscure." The
>>K-particle puzzle is not hard, it's obscure.
>

>....unless, of course, it involved figuring out the transition amplitudes


>for that particular decay mode, in which case it would be both hard
>and obscure. :-)
>

>As has been stated before, it is very hard to postulate general rules

>about what makes good puzzles and what doesn't. Here is one, though:


>unless you're aiming at a very limited audience, any puzzle requiring
>intimate knowledge of the theory of electroweak interactions is probably
>not a very good idea. (Another one for your .sig, Cardinal? :-) ).

Well, it seems there's at least three of us who'd be interested in the
puzzle. Now just throw in a good quark gluon plasma and I'm there!

--
KEN FAIR - U. Chicago Law | <http://student-www.uchicago.edu/users/kjfair>
Of Counsel, U. of Ediacara | Power Mac! | CABAL(tm) | I'm w/in McQ - R U?
When you go for a job interview, I think a good thing to ask is if
they ever press charges.

Reply all
Reply to author
Forward
0 new messages