Cryptographic puzzles

12 views
Skip to first unread message

David Fisher

unread,
Nov 6, 2005, 7:26:02 AM11/6/05
to
I was wondering which of the following "encryption" puzzles seem fair to
people (in increasing order of difficulty - matter of opinion) ...

* Mirror writing - "siht daer uoy nac"
* First letter of each word - "his old wooden apple bounced over - under -
through the hedge in some other neighbouring entryway"
* Numbers representing letters - "14 15 20, 20 15 15, 2 1 4"
* Caesar cipher, shifting each letter by a fixed amount - "xgta vtkema"
* Substitution cipher - each letter is represented by another one, not
necessarily in order (eg. "A" means "G", "B" means "P" ...). You need a
large chunk of text to work this one out, but it isn't impossible if you
know what to look for - "A" and "I" are the only possible one letter words,
etc.
* Others ? (Any examples from real games ? If so, don't forget the "spoiler"
warning ...)

Just wondering,

David Fisher


Bert Byfield

unread,
Nov 6, 2005, 9:58:44 AM11/6/05
to
> I was wondering which of the following "encryption" puzzles seem fair
> to people (in increasing order of difficulty - matter of opinion) ...
> * Others ? (Any examples from real games ? If so, don't forget the...
> David Fisher

To see a whole world of basic encryption techniques, look into:
http://www.cryptogram.org/

dgen...@hotmail.com

unread,
Nov 6, 2005, 1:33:56 PM11/6/05
to

You really have to ask yourself who your intended audience is; a
player who enjoys these sorts of puzzles will find the first sort too
easy.

There was a code of the second sort (provided as a clue) in the orignal
Zork I, which should have been fairly obvious, although I didn't find
out about it myself until reading something online very recently, more
than 20 years after I first completed the game.

Even caesar ciphers and substitution ciphers aren't very difficult
with free codebreaking tools (or without them, for that matter).
Graham Nelson included a damned enigma code in Jigsaw, so I don't think
that anything is necessarily off limits.

On the other hand, players who don't like code puzzles will find even
the simplest ones annoying.

Dave E

> David Fisher

solar penguin

unread,
Nov 7, 2005, 3:54:09 AM11/7/05
to
--- David Fisher said...

> I was wondering which of the following "encryption" puzzles seem fair
> to people (in increasing order of difficulty - matter of opinion) ...
>

> * Others ? (Any examples from real games ? If so, don't forget the
> "spoiler" warning ...)
>

Mild spoilers follow.

One classic Infocom game had a "different tastes and flavours represent
different words" code. Yes, really. You had to lick the message to
find out what it tasted like, and then decode that.

--
___ _ ___ _
/ __| ___ | | __ _ _ _ | _ \ ___ _ _ __ _ _ _ (_) _ _
\__ \/ _ \| |/ _` || '_| | _// -_)| ' \ / _` || || || || ' \
|___/\___/|_|\__,_||_| |_| \___||_||_|\__, | \_,_||_||_||_|
|___/
http://www.freewebs.com/solar_penguin/

** They will soon be revealed in a state of the solar system.

** A xenomorph may be tucked away in the art of preparing food for the
heart of the Two Trees, for example.


Richard Bos

unread,
Nov 7, 2005, 12:11:35 PM11/7/05
to
"David Fisher" <da...@hsa.com.au> wrote:

> I was wondering which of the following "encryption" puzzles seem fair to
> people (in increasing order of difficulty - matter of opinion) ...
>
> * Mirror writing - "siht daer uoy nac"

That's not even a puzzle.

> * First letter of each word - "his old wooden apple bounced over - under -
> through the hedge in some other neighbouring entryway"

This needs to be clued, or I'm not even going to notice that there is an
encrypted text in it.

> * Numbers representing letters - "14 15 20, 20 15 15, 2 1 4"
> * Caesar cipher, shifting each letter by a fixed amount - "xgta vtkema"

Both quite obvious and relatively simple. Especially with longer texts,
where the large numbers of Es in English (and Dutch, and German) is
helpful; but if the cleartext is readily recognisable, also with
passwords and so forth.

> * Substitution cipher - each letter is represented by another one, not
> necessarily in order (eg. "A" means "G", "B" means "P" ...). You need a
> large chunk of text to work this one out, but it isn't impossible if you
> know what to look for - "A" and "I" are the only possible one letter words,
> etc.

The frequency of ETAION SHRDLU is helpful here, too. Since it can be a
lot of work, I wouldn't use this as the only solution to a puzzle.

> * Others ? (Any examples from real games ? If so, don't forget the "spoiler"
> warning ...)

First letter of each line of a poem or inscription. This is more likely
to be noticed than the first letter of each word, but obviously needs
more text.
A couple of grids of letters that need to be put together (with each
supplying, say, a third of the letters).
As a special case of the Caesar, ROT-13; as a special case of the
Substitution, a reversed alphabet.

Richard

Nathan

unread,
Nov 8, 2005, 1:53:22 PM11/8/05
to
solar penguin wrote:

> One classic Infocom game had a "different tastes and flavours represent
> different words" code. Yes, really. You had to lick the message to
> find out what it tasted like, and then decode that.

Okay, I give up. Which game are you thinking of? I'm pretty sure it's
not really Infocom.

Andrew Plotkin

unread,
Nov 8, 2005, 2:11:35 PM11/8/05
to
(SPOILER)

Stationfall.

--Z

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

Nathan

unread,
Nov 8, 2005, 2:58:10 PM11/8/05
to

Andrew Plotkin wrote:
> Here, Nathan <nts...@netscape.net> wrote:
> > solar penguin wrote:
> >
> > > One classic Infocom game had a "different tastes and flavours represent
> > > different words" code. Yes, really. You had to lick the message to
> > > find out what it tasted like, and then decode that.
> >
> > Okay, I give up. Which game are you thinking of? I'm pretty sure it's
> > not really Infocom.
>
> (SPOILER)
>
>
>
>
>
> Stationfall.

Doh! I even remember that now!

Dan Shiovitz

unread,
Nov 9, 2005, 1:04:04 AM11/9/05
to
In article <NHmbf.126$WM....@nasal.pacific.net.au>,

David Fisher <da...@hsa.com.au> wrote:
>I was wondering which of the following "encryption" puzzles seem fair to
>people (in increasing order of difficulty - matter of opinion) ...
[..]

>* Mirror writing - "siht daer uoy nac"
>* First letter of each word - "his old wooden apple bounced over - under -
>through the hedge in some other neighbouring entryway"
>* Numbers representing letters - "14 15 20, 20 15 15, 2 1 4"
>* Caesar cipher, shifting each letter by a fixed amount - "xgta vtkema"

Nobody seems to have addressed the word 'fair' specifically, but this
seems like the main point, really. Fairness doesn't have anything to
do with difficulty, really -- it's more about "can the player work out
the procedure for solving this puzzle?" The "first letter of each
word" encryption puzzles tend to suffer from the most basic fairness
issue, where it's not obvious there's a puzzle here. Then when it's
clear that there's something to be solved, the other fairness issue is
usually whether the player has enough information about the approach
for solving it.

I think both Savoir-Faire and Christminster have in-game reference
books for solving ciphers. This both suggests an approach and tips off
the player that, hey, this puzzle probably needs to be solved by you
the player doing out-of-game work to puzzle it out, as opposed to
finding an in-game device to do it (although SF also provides an
in-game way to decipher -- an elegant solution for satisfying both
people who want to break the cipher by hand and people who don't).

All that said, there's also the issue of "suitability" or "audience
appeal" or "fun" or whatever you want to call it -- it's perfectly
fair to include a message encoded with a 1024-bit RSA key, but I doubt
many players are going to stick around to decode it. Under this
requirement, I think the general rule is that somebody with no
cryptographic knowledge beyond what's given in the game ought to be
able to solve the puzzle in an hour or two. Usually the way to do this
is to minimize the tedious grinding part of solving, and use a code
that requires guessing some key and is easy to decode once the key is
guessed (then there's a fairness issue with the guess-the-key puzzle,
but that's another issue).

>David Fisher
--
Dan Shiovitz :: d...@cs.wisc.edu :: http://www.drizzle.com/~dans
"He settled down to dictate a letter to the Consolidated Nailfile and
Eyebrow Tweezer Corporation of Scranton, Pa., which would make them
realize that life is stern and earnest and Nailfile and Eyebrow Tweezer
Corporations are not put in this world for pleasure alone." -PGW

Kevin Forchione

unread,
Nov 9, 2005, 11:33:50 AM11/9/05
to

"Dan Shiovitz" <d...@cs.wisc.edu> wrote in message
news:dks3gk$26p$1...@cascadia.drizzle.com...
<snip>

> All that said, there's also the issue of "suitability" or "audience
> appeal" or "fun" or whatever you want to call it -- it's perfectly
> fair to include a message encoded with a 1024-bit RSA key, but I doubt
> many players are going to stick around to decode it. Under this
> requirement, I think the general rule is that somebody with no
> cryptographic knowledge beyond what's given in the game ought to be
> able to solve the puzzle in an hour or two. Usually the way to do this
> is to minimize the tedious grinding part of solving, and use a code
> that requires guessing some key and is easy to decode once the key is
> guessed (then there's a fairness issue with the guess-the-key puzzle,
> but that's another issue).

This form of puzzle doesn't seem all that different from Mazes to me. I'd
suspect they'd get the same kind of reaction from players.

--Kevin


dave e

unread,
Nov 9, 2005, 11:44:55 AM11/9/05
to

You're right.

But, you know, I always take a certain comfort when I encounter a maze
or a cryptogram in a game, because I know that given enough effort I
WILL be able to solve it. I don't take the same comfort when I
encounter the other classical type of puzzle, in which the player is
presented with a box of chocolates, a violin, a stack of zorkmid coins,
and a dozen other random objects and has to figure out what set of
commands involving these objects (or some item not yet discovered) is
needed to get through an impassable door.

Dave

Paul Drallos

unread,
Nov 9, 2005, 12:26:56 PM11/9/05
to
> --- David Fisher said...
>
>
>>I was wondering which of the following "encryption" puzzles seem fair
>>to people (in increasing order of difficulty - matter of opinion) ...
>>
>
>
>>* Others ? (Any examples from real games ? If so, don't forget the
>>"spoiler" warning ...)
>>
>
In my game Dawn of the Demon I used a puzzle like this:

RAT + ROT = TOTO

The sum will make sense only for certain letter/number
equivalences.


Esa A E Peuha

unread,
Nov 9, 2005, 4:32:39 PM11/9/05
to
Paul Drallos <pdra...@tir.com> writes:

> In my game Dawn of the Demon I used a puzzle like this:
>
> RAT + ROT = TOTO
>
> The sum will make sense only for certain letter/number
> equivalences.

Well, that puzzle doesn't have a solution (T must be 1, so O = 2 and
A = 9, but then 2 * R would have to be 11).

--
Esa Peuha
student of mathematics at the University of Helsinki
http://www.helsinki.fi/~peuha/

Paul Drallos

unread,
Nov 9, 2005, 7:30:07 PM11/9/05
to
Esa A E Peuha wrote:

> Paul Drallos <pdra...@tir.com> writes:
>
>
>>In my game Dawn of the Demon I used a puzzle like this:
>>
>>RAT + ROT = TOTO
>>
>>The sum will make sense only for certain letter/number
>>equivalences.
>
>
> Well, that puzzle doesn't have a solution (T must be 1, so O = 2 and
> A = 9, but then 2 * R would have to be 11).
>

My mistake EAT + RAT + ROT = TOTO


Bert Byfield

unread,
Nov 9, 2005, 8:45:27 PM11/9/05
to

I can do those pretty quickly, but I expect lots of people find that
impossible. Cryptograms are similar. I would think that there should be a
"warning" up front about this sort of specialized puzzle, so as not to
alienate the liberal arts people.


Kevin Forchione

unread,
Nov 9, 2005, 10:52:56 PM11/9/05
to
"Bert Byfield" <BertB...@nospam.not> wrote in message
news:Xns9709D33D566C4b...@24.24.2.166...
<snip>

> I can do those pretty quickly, but I expect lots of people find that
> impossible. Cryptograms are similar. I would think that there should be a
> "warning" up front about this sort of specialized puzzle, so as not to
> alienate the liberal arts people.

Not to mention the kids who aren't in college...

--Kevin


Rikard Peterson

unread,
Nov 10, 2005, 3:56:08 AM11/10/05
to

"Bert Byfield" wrote in
news:Xns9709D33D566C4b...@24.24.2.166:

>> RAT + ROT = TOTO
>

> I can do those pretty quickly, but I expect lots of people find
> that impossible. Cryptograms are similar. I would think that there
> should be a "warning" up front about this sort of specialized
> puzzle, so as not to alienate the liberal arts people.

I have one of those in my WIP. It turned off one tester who had liked
the game a lot up until then, and I had even given an extra hint to
make it easy.

"Well, the trouble is, for me, it is decidedly not interesting. :)
I'm not interested in math exercises in games, but for now, I'm just
going to have to leave it- I've run out of energy to apply additional
effort ;\ (sorry)

"Of course, I should point out- I *have* enjoyed playing the chapter,
as you shall read... ;) "

I'm still thinking about what to do with this part of the game. (None
of the other testers said anything positive about this particular
puzzle either.)

Rikard

solar penguin

unread,
Nov 10, 2005, 4:39:19 AM11/10/05
to
--- Paul Drallos said...

Still not a unique solution: 241+541+531=1313 or 732+932+962=2626. Were
there other clues scattered in your game? Or was it meant to be open
ended?

--
___ _ ___ _
/ __| ___ | | __ _ _ _ | _ \ ___ _ _ __ _ _ _ (_) _ _
\__ \/ _ \| |/ _` || '_| | _// -_)| ' \ / _` || || || || ' \
|___/\___/|_|\__,_||_| |_| \___||_||_|\__, | \_,_||_||_||_|
|___/
http://www.freewebs.com/solar_penguin/

** Even crossbreeding was possible between elves and humans in the
forests of North America in the south.

** Peter is best known for its lurid descriptions of the dinosaurs.


Boluc Papuccuoglu

unread,
Nov 10, 2005, 7:48:51 AM11/10/05
to
On 10 Nov 2005 08:56:08 GMT, Rikard Peterson <trumg...@bigfoot.com>
wrote:

2003 IFComp Winning Entry Slouching Towards Bedlam had a similar
puzzle, but it provided an alternate solution.

ROT13 Text Follows:

Vafgrnq bs qrpvcurevat gur jnyy jevgvat ng bar cbvagvat, lbh pbhyq
fvzcyl beqre gur Gevntr ebobg/pbzchgre gb qrplcure vg.

Perhaps you could place a codebreaking machine,etc in your WIP, but to
use it the PC would have to solve a non-mathematical puzzle?

Boluc

Paul Drallos

unread,
Nov 10, 2005, 9:08:57 AM11/10/05
to
solar penguin wrote:

> --- Paul Drallos said...
>
>
>>Esa A E Peuha wrote:
>>
>>
>>>Well, that puzzle doesn't have a solution (T must be 1, so O = 2 and
>>>A = 9, but then 2 * R would have to be 11).
>>>
>>
>>My mistake EAT + RAT + ROT = TOTO
>
>
> Still not a unique solution: 241+541+531=1313 or 732+932+962=2626. Were
> there other clues scattered in your game? Or was it meant to be open
> ended?
>

Yes, there was one other clue which would select the prefered solution. The puzzle is the key to a 5 step sequence, so the solution containing only the digits 1,2,3,4 and 5 is the correct solution.

I alow would like to mention that one of my testers *loved* this puzzle and especially looks forward to any mathematical-type puzzle. So they don't alienate everybody.

Reply all
Reply to author
Forward
0 new messages