Zork: Solving a puzzle via verb-checking   7 comments

When people talk about brute forcing on an adventure, they usually mean trying every inventory object on every puzzle.

Really though, there are four methods of brute force:
1. Map-checking
2. Item-checking
3. Puzzle-checking
4. Verb-checking

#1 for instance could be trying every feasible action in each room, checking them off as you go. #3 could be taking every item to a particular puzzle and testing each one out, hoping for a result. Occasionally a game will have an extra category particular for the game (for example, trying to use the magic words from Adventure in every possible room).

#4 is often overlooked but is quite handy for old adventures: trying everything that might plausibly be a verb for future reference. In particular I tested out SLIDE and got

>slide
Slide what?

Hmm. What slides? I tried the rubber mat from the beginning of the game:

>slide mat
under what?

Oho! I was in the first room, so I tried DOOR:

>slide mat under door
There’s not enough room under this door.

I could tell where this was going: there’s a puzzle that’s appeared in approximately 500 (*) other adventure games where a key has for some reason been left on the other side of a keyhole. To its credit, this appearance would be the first. A puzzle that could have been ornery later — it’s not obvious from the room description it would’ve been possible — turned out to be quite easy forewarned with this knowledge.

(*) Only a slight exaggeration.

Posted April 25, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork: Base of operations   5 comments

For a game with large landscape, many puzzles, and a harsh inventory limit, it’s essential to have a base of operations: a place to stash everything. One of my stopping points in Zork is that it’s been hard to find one due to the thief (who will keep treasures, and take non-treasures and drop them off at random spots on the map).

A good base here needed, preferably:

a.) To be fairly central to the map
b.) To be a place avoided by the thief
c.) To have its own source of light, so I can give the (time-limited) battery powered lantern a rest while I rummage through items and decide my next move
d.) Be reasonably close to an egress to the surface, to make it easier to start filling the trophy case with treasures

I originally tried the Flood Control Dam #3, because it seemed to fit conditions a, b, and c and marginally d (it’s five rooms away from the Cellar and exit — once the trophy case has at least one treasure the trapdoor stays open) but I discovered that the thief does show up, he just doesn’t show his face; items in the room just start disappearing one by one.

After fussing about some more, I noticed the Temple was also well lit, and lo (peeking at David Cornelson’s map to see if I missed anything on my map) there’s a very good source for exit to the surface and the thief also avoids the temple.

So my progress has accelerated, although strangely I haven’t really solved much more; just I’ve been able to put together all the parts I’ve solved during individual runs on a continuous run.

Posted April 18, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork: Ports   22 comments

I’ve tried my third port of Zork (although I am sticking with the second for now), so I wanted to talk more about the versions that are out there.

To recap an earlier post, the mainframe version of Zork started in 1977 a language called MDL, was ported in early 1978 to FORTRAN and renamed Dungeon, and was renamed again (due to trademark dispute) to Zork before it went commercial.

The FORTRAN source kept getting modified by the original authors up until 1980 and the MDL source kept getting modified up until 1981. However, no versions from 1977 seem to exist (Dungeon 2.5, the earliest version number on if-archive, came from 1980) so any type of modern play will be several years removed from the original.

I started with ZDungeon, a port by Ethan Dicks into z-code based off the MDL source. It seemed like the sanest way at the time to get the closest experience I could to the original, and being in z-code means I get to play it with Gargoyle:

In addition to looking easy on the eyes, it means I can SAVE and RESTORE and even UNDO to my heart’s content.

However, David Kinder kindly pointed me to an interpreter called Confusion by Matthew Russotto that can play the MDL source directly.

Since I’m tackling this project (which I have yet to name or define or admit exists) very seriously on the historical angle, I switched over. I noticed, as David Cornelson mentioned to me, the behavior of the thief seems different: the ZDungeon port has the thief appear in more rooms (any room?) and there’s no warning from the sword about thief proximity. So that aspect of the switch made me happy.

However, I was still slightly unhappy. I was fine sacrificing font readability (which wasn’t too bad) but I was having issues with game saves: the original Zork only allows one save at a time. I could do some file-copy-wrangling to get around that, but when trying an intense amount of experimentation to got to be a hassle. Also not-fun: upon dying, RESTORE no longer works:

> e
Oh, no! You walked into the slavering fangs of a lurking grue.

As you take your last breath, you feel relieved of your burdens. The
feeling passes as you find yourself before the gates of Hell, where
the spirits jeer at you and deny you entry. Your senses are
disturbed. The objects in the dungeon appear indistinct, bleached of
color, even unreal.
> restore
You can’t do even that.

Even that I would have been able to hang on with, but twice I’ve gotten odd crashes like this:

>play violin
*ERROR*
“UVECTOR-PUT-TYPE-VIOLATION”
LISTENING-AT-LEVEL 2 PROCESS 1
Atom REP has neither LVAL nor GVAL

I don’t think Confusion has bugs; I have the feeling these are “authentic” crashes in the code.

I decided to try a Windows port derived from the FORTRAN source (I used the WinGLK one by Andrew Plotkin).

There are some definite port changes. There’s a status line, the “restore” command works when dead, and there’s been bug fixes above and beyond the original port. There’s also (according to the README file) this:

VI. Assorted Additions (up to oct-94)

Performed, to the best of my knowledge, by Robert Supnik. This includes
several puzzles, “lots and lots and lots and LOTS” of bug fixes, an
unsatisfying afterlife, and a Last Lousy Point.

I’m not sure what’s an “unsatisfying afterlife”, but additions of “several puzzles” definitely makes me nervous.

There’s earlier Dungeon ports I could try going back to, even straight from the 1980 source, but at the moment all this has me scratching over my head what a “best” experience of the original would be. Would the original authors really want bugs as part of the experience? I do get the sense that while the MDL source hasn’t gone through bugfix revisions the FORTRAN source sure has (1981, 1985, 1986, 1987, 1994, and 1998) and maybe I’m better off with problems fixed. Rather than HELLO, SAILOR crashing the game (as it does in Confusion) in the Plotkin port of Dungeon it does this:

>hello, sailor
Misplaced comma or conjunction.

Should that really be frowned upon? Am I playing for most enjoyable experience or maximum authenticity? Am I stalling by writing this post because yet another puzzle is stumping me?

At least I know the answer to the last question.

Posted April 12, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork: Playing off and on the computer   8 comments

(Heavy spoilers ahead.)

Last time I posted about my system for handling being stuck. I’ve become less stuck; here’s how my breakthrough went down.

First there was a puzzle I already remembered from Zork II:

Riddle Room
This is a room which is bare on all sides. There is an exit down.
To the east is a great door made of stone. Above the stone, the
following words are written: ‘No man shall enter this room without
solving this riddle:

What is tall as a house,
round as a cup,
and all the king’s horses can’t draw it up?’

(Reply via ‘ANSWER “answer”‘)

I remember being stumped at the time. I realize it’s likely an easy-in-retrospect thing, but WELL now seems obvious to me.

Moving on:

Circular Room
This is a damp circular room, whose walls are made of brick and
mortar. The roof of this room is not visible, but there appear to be
some etchings on the walls. There is a passageway to the west.
There is a wooden bucket here, 3 feet in diameter and 3 feet high.
> read etchings

                        o  b  o

                        A  G  I
                         E   L

                        m  p  a

I could tell it was the bottom of a well (being prepared by the riddle) but the message stumped me, and while I could enter the bucket I couldn’t get the bucket anywhere (too large to pick up, and the rope didn’t attach). I considered anagrams: “Amiable Goop”? “Go Amoeba Lip”? I thought maybe “EL” was part of “WELL” but I couldn’t make more out of that.

At this point I reached my Stuck post. After I made the lists of objects and puzzles, I printed them out so I could study them away from the computer. My playing has been very sporadic out of work/life necessity anyway; adventure games are interesting in they are one of the few computer game genres playable outside of the computer. It’s even possible to play while sleeping (I recall a few stories of people realizing a puzzle solve from a dream).

Cue several days later: in between wrangling a small child, I glanced over at my printout, and almost mystically (or at least remembering the Frobozz Magic Goop Company from the toothpaste tube, also on my object list), I saw:

                     Fr o  b  o zz

                      M A  G  I C
                       W E   L L

                     Co m  p  a ny

Ah-ha! Somehow this insight was difficult from a computer screen, but it came easily on paper.

However, I was still stuck. Here’s how my thoughts went: Does this help me any? I want to move the well. I guess the rope was futile, it’s a magic well. Maybe magic words or some such?

I jumped back in the game, entered the well, and tried all variety of “magic” words. No luck, but as I was trying things out the thought sprung up: Well, how would a magic well work? It’d come up when there was water, right? Don’t I have water?

> put water in bucket
There is now a puddle in the bottom of the wooden bucket.
The bucket rises and comes to a stop.
Top of Well
You are at the top of the well. Well done. There are etchings on
the side of the well. There is a small crack across the floor at the
entrance to a room on the east, but it can be crossed easily.
You are in the wooden bucket.
The wooden bucket contains:
A quantity of water

Ha-HA! What made this extra satisfying came immediately after:

> read etchings

                        o  b  o
                    r             z
                 f   M  A  G  I  C   z
                 c    W  E   L  L    y
                    o             n
                        m  p  a

This was truly a nice solve: not only did I unlock a bottleneck in the map, but there was a artistic confirmation of my original insight. I’m fairly certain the second insight would _not_ have come to me off the computer, even thought it was quite possible to finish the solve based on the project list. Somehow being in the world-verse led me to the correct revelation.

I’m hoping thinking more about my stuck-unstuck process will improve my overall puzzle solving, and lead to theories to solidify what I consider “good” stuck and what I consider “bad” stuck.

Posted April 9, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork: Stuck   14 comments

(Heavier spoilers than usual this post.)

First, a mea culpa: you do get a faint blue glow when the thief is nearby, but only in the original Muddle version and not in the ZDungeon port. It’s actually even more effective than the sword suddenly turning on: it happens more frequently, and it makes for more moments of teeth-gnashing.

Paul O’Brian writes that “Dungeon wants nothing more than to see you fail” and I’m feeling the burn. I’ve developed a system over the years to give myself the best chance of solving adventure games; unfortunately my rate of hintless wins has not improved much. The only games I’d call “substantial” I beat with no hints at all were Countdown to Doom (1982) and Anchorhead (1998). Other than the “Beginner” level Wishbringer (1985) I haven’t beaten any Infocom game without hints.

Still I find my system helpful, so I’m plowing ahead. First I list all my open puzzles:

Locked trapdoor
Locked grating
Bank of Zork
Round Room
Entrance to Hades
Circular Room mysterious message:

   o  b  o
   
   A  G  I
    E   L
   
   m  p  a

Getting out of Aragain Falls with treasure intact
Getting coffin out of Egyptian Room area
Passage full of ice
Handling light source at narrow passage near coal mine

Then I list all my objects I have access to:

coil of rope
nasty-looking knife
elvish sword
jewel-encrusted egg
brown sack
garlic
old leather bag, bulging with coins
set of skeleton keys
deceased adventurer’s useless lantern
rusty knife
shovel
small brass bell
book
burning candles
grail
pearl necklace
bucket (can’t carry)
portrait
painting
solid-gold coffin
coil of thin shiny wire
toothpaste tube (Frobozz magic gunk)
wrench
screwdriver
matchbook
guidebook
sharp stick
folded plastic / pump -> boat
trunk with jewels
crystal trident
jade figure
bracelet
shovel
bat guano

Then I sit and compare the two lists and a map until something pops in my head. For instance, on one previous list-comparison check I realized I’d used the boat on only one body of water but the map had more I could try. My current list of things to try:

try shovel everywhere
burning timber didn’t work; try burning other things that can remain lit while on the floor to get an extra light source near the narrow passage
try alternate exit from area after the narrow passage instead
mess around with bell / candles / books at Hades some more (bell does have some effect)
try ‘odysseus’ or ‘hello, sailor’ more places

This works best if there’s a puzzle where all the systems are clear and it is just a matter of fitting the pieces together (like the troll bridge puzzle from Adventure) and worst when there’s some sort of in-game experimentation required (like not realizing an item is magic — that trident looks suspicious — and there’s no clue except for in-game manipulation).

Sometimes I’ll miss the existence of items; for instance the fact the “grail room” had an actual Grail object that could be picked up strangely eluded me for some time.

Absolute worst is when I miss a map exit exists. This can happen to me even on good, sensibly mappable games (I missed going up from the kitchen in Savoir Faire for something like an hour). One of my standing stuck procedures is to double check every room on the map and try every single direction — even if the room description insists there’s only one way out — just in case.

Anyone have their own method?

Posted April 6, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork: Plot and story   3 comments

I didn’t discuss Adventure‘s overall plot at all; there’s a true sense that there isn’t one. The entire task was of collecting treasures and storing them, and the “endgame” was tacked on in the same manner as a late-80s arcade game.

That’s not to say there wasn’t a story — see for example my experiences in the maze playing a game of cat-and-mouse with dwarves — but this story was generated strictly from the system, and transplants into a (relatively) barren universe.

Zork is nearly the same setup — you’re urged to go find treasure, then set loose — but there’s something more textured about it. Entering the dungeon isn’t just a casual affair, but rather sinister. Upon going down the trap-door leading to the main part of the game:

The trap door crashes shut, and you hear someone barring it.

Or compare Adventure‘s pirate

OUT FROM THE SHADOWS BEHIND YOU POUNCES A BEARDED PIRATE! “HAR, HAR,” HE CHORTLES. “I’LL JUST TAKE ALL THIS BOOTY AND HIDE IT AWAY WITH ME CHEST DEEP IN THE MAZE!” HE SNATCHES YOUR TREASURE AND VANISHES INTO THE GLOOM.

with Zork‘s thief:

Someone carrying a large bag is casually leaning against one of the walls here. He does not speak, but it is clear from his aspect that the bag will be taken only over his dead body.
Your sword has begun to glow very brightly.

The thief doesn’t just come out of the shadows, he is just there. The sword, which previously signaled danger by steps with a “faint blue glow” when an enemy is close and a glow “very brightly” when reaching the enemy, goes straight from dark to bright. The thief just leans against the wall — no visible signs of action — and you don’t notice your valuables have been removed until after he is already gone.

Not only is there a texture of menace, but a feeling of background; setting as character, if you will. Wandering outdoors:

You are at the top of the Great Canyon on its south wall. From here there is a marvelous view of the Canyon and parts of the Frigid River upstream. Across the canyon, the walls of the White Cliffs still appear to loom far above. Following the Canyon upstream (north and northwest), Aragain Falls may be seen, complete with rainbow. Fortunately, my vision is better than average and I can discern the top of the Flood Control Dam #3 far to the distant north. To the west and south can be seen an immense forest, stretching for miles around. It is possible to climb down into the canyon from here.

The casual mention of Aragain Falls and Flood Control Dam #3 suggest a story behind the place that is meant to be unraveled. Both locations become important later. Compare with a roughly equivalent scene in Adventure:

YOU ARE ON THE EDGE OF A BREATH-TAKING VIEW. FAR BELOW YOU IS AN ACTIVE VOLCANO, FROM WHICH GREAT GOUTS OF MOLTEN LAVA COME SURGING OUT, CASCADING BACK DOWN INTO THE DEPTHS. THE GLOWING ROCK FILLS THE FARTHEST REACHES OF THE CAVERN WITH A BLOOD-RED GLARE, GIVING EVERYTHING AN EERIE, MACABRE APPEARANCE. THE AIR IS FILLED WITH FLICKERING SPARKS OF ASH AND A HEAVY SMELL OF BRIMSTONE. THE WALLS ARE HOT TO THE TOUCH, AND THE THUNDERING OF THE VOLCANO DROWNS OUT ALL OTHER SOUNDS. EMBEDDED IN THE JAGGED ROOF FAR OVERHEAD ARE MYRIAD TWISTED FORMATIONS COMPOSED OF PURE WHITE ALABASTER, WHICH SCATTER THE MURKY LIGHT INTO SINISTER APPARITIONS UPON THE WALLS. TO ONE SIDE IS A DEEP GORGE, FILLED WITH A BIZARRE CHAOS OF TORTURED ROCK WHICH SEEMS TO HAVE BEEN CRAFTED BY THE DEVIL HIMSELF. AN IMMENSE RIVER OF FIRE CRASHES OUT FROM THE DEPTHS OF THE VOLCANO, BURNS ITS WAY THROUGH THE GORGE, AND PLUMMETS INTO A BOTTOMLESS PIT FAR OFF TO YOUR LEFT. TO THE RIGHT, AN IMMENSE GEYSER OF BLISTERING STEAM ERUPTS CONTINUOUSLY FROM A BARREN ISLAND IN THE CENTER OF A SULFUROUS LAKE, WHICH BUBBLES OMINOUSLY. THE FAR RIGHT WALL IS AFLAME WITH AN INCANDESCENCE OF ITS OWN, WHICH LENDS AN ADDITIONAL INFERNAL SPLENDOR TO THE ALREADY HELLISH SCENE. A DARK, FORBODING PASSAGE EXITS TO THE SOUTH.

Vivid, certainly, but adding to the story in only an immediate sense and not in establishing persistent background details and the feeling of a world history.

Posted April 4, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork: Open spaces, painful geography   20 comments

Just to clarify a bit of history: the mainframe Zork I am playing was too large a game for personal computers at the time, so when it went commercial it got split into Zork I, II, and III. Since I have only played the commercial versions playing mainframe Zork is like a fuzzy mash-up of my childhood memories.

One thing I noticed (compared to modern games) with Adventure and now also Zork is that it is enjoyable to explore a wide open space; the feeling of world-immersion is strong even when many of the room descriptions are unassuming.

Clearing
You are in a clearing, with a forest surrounding you on the west and south.
There is a pile of leaves on the ground.

Modern adventures and interactive fiction tend to a tight room-structure, where no space is “wasted”. With a full enough immersion in the world-space I don’t believe rooms are necessarily wasted. I’ll catch myself on that by saying I have experienced many adventure games where wasted rooms are both meaningless and painful (the worst offender I recall is Time Zone) and I can’t mathematize why Zork is different; just the amount of space feels right.

Not all is perfect with the geography, though. There appears a willful desire on the designer’s part that if you entered a room by going east, going west will not get you back to the same room. I can’t tell you how many times I had to redo room connections while working on the larger dungeon map:

Mapping with Trizbort is nice in that I can shift around whole sections to get the geography right. For example, there’s a “South Reservoir” that later connects up with a “North Reservoir”, so it’d be nice to have them in direct line with each other, but I had originally drawn the two areas all the way across from each other on the map. Just about a minute of cut and paste and everything was fixed. I suspect the general relationship of geography is important for at least one puzzle so I want to get it right.

The first maze (there’s at least one more, and I hear there’s a third epic one later) has a cute tribute to Adventure. It appears the same intrepid explorer who got carried away by elves in triumph didn’t make it in the Zork universe.

Incidentally, I’ve been picking up very little in the way of objects. The inventory limit is brutal, but interestingly enough it is based on object weight rather than exact number. It’s a nice touch of world-modeling which isn’t duplicated that often (I think?) in this era. The upshot is I’ve hardly started on puzzle solving at all. I do remember Zork I well enough that some parts will give me no trouble, but my Zork II and Zork III are quite foggy. There’s a rotating circular room that is driving me nuts. I presume there’s some way to stop it? I have forgotten how.

Posted April 2, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Zork (original MUDDLE mainframe version)   20 comments

WELCOME TO ZORK

ZORK is a game of adventure, danger, and low cunning. In it you will explore some of the most amazing territory ever
seen by mortal man. Hardened adventurers have run screaming from the terrors contained within!

In ZORK the intrepid explorer delves into the forgotten secrets of a lost labyrinth deep in the bowels of the earth,
searching for vast treasures long hidden from prying eyes, treasures guarded by fearsome monsters and diabolical traps!

No PDP-10 should be without one!

PDP-10, picture by Michael L. Umbricht


I wanted to play the next adventure game chronologically after the Crowther and Woods Adventure, but the history of mainframe games (on monstrosities like the PDP-10) is so murky it was difficult to tell what should come next. For one thing, the mainframe games did not have “release dates” — they were works in progress and in some cases earlier versions were more well-known than later versions. If it took 4 years to write something should I be using the end date or the start date? What if the later changes were only minor things dealing with source ports?

Additionally, even with testimony from the original authors, memories are foggy about exact years (something Dennis Jerz struggled with in his Adventure article) and the early history of electronic games tends to the wildly inaccurate anyway.

Fortunately, in 1985 the New Zork Times published a “History of Zork” which not only mentions not only years but months, so aside from one exception (which I will place next chronologically) it’s fairly clear Zork came next after Adventure.

Specifically, in early 1977 after the Crowther and Woods Adventure had gone viral on the ARPAnet (precursor to the Internet) and a group at MIT (Tim Anderson, Marc Blank, Bruce Daniels, and Dave Lebling) got the urge to write their own game:

Dave wrote (in MUDDLE) a command parser that was almost as smart as Adventure’s; Marc and I [Tim Anderson], who were both in the habit of hacking all night, took advantage of this to write a prototype four-room game. It has long since vanished. There was a band, a bandbox, a peanut room (the band was outside the door, playing “Hail to the Chief”), and a “chamber filled with deadlines.”

This led through the summer and the fall of 1977 to progress on what was at the time called Zork, until Robert M. Supnik got the notion to translate Zork from MUDDLE to FORTRAN:

At any rate, shortly after the Great Blizzard of ’78 he had a working version, initially for PDP-11s. Since it was in FORTRAN, it could run on practically anything, and by now it has.

At the same time the name got changed from Zork to Dungeon because as Tim writes: “Zork was too much of a nonsense word, not descriptive of the game”. This held for around a year when:

Fortunately for us, a certain company (which shall remain nameless) decided to claim that it had trademark rights to the name Dungeon, as a result of certain games that it sold. We didn’t agree (and MIT had some very expensive lawyers on retainer who agreed with us), but it encouraged us to do the right thing, and not hide our “Zorks” under a bushel.

(He’s referring to TSR and its boardgame Dungeon! from 1975.)

So the name got changed back to Zork, and the rest is history which I’ll sum up via original logo and box art rather than words:

(Image from the Gallery of Undiscovered Entities.)

(Image from Ye Olde Infocom Shoppe.)

Just like Adventure there’s a bevy of ports, but I’m going to go with one based off the original MUDDLE code (rather than the FORTRAN rewrite) into Z-code by Ethan Dicks called ZDungeon. I compared it with the mainframe version on Twenex and it is accurate enough for me to be happy.

Actual gameplay will start in my the next post!

Posted March 31, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Adventure (350 points): Puzzles and concluding remarks   15 comments

Huzzah! and so forth:

YOU MARCH THROUGH THE HOLE AND FIND YOURSELF IN THE MAIN OFFICE, WHERE A CHEERING BAND OF FRIENDLY ELVES CARRY THE CONQUERING ADVENTURER OFF INTO THE SUNSET.

It was easier to finish than I expected. I’m not sure if I’m better at these things now or if my success is attributable to vague half-memories of how puzzles worked. In any case, it was a pleasant enough experience which still holds up as a game. There’s shadows of emergent behavior (as I discussed with the maze), a strong sense of environment (given many of the rooms are based on the real Colossal Cave, see this article for pictures) and the impression of a rational system beneath the workings of the puzzles (I discuss what I mean by this below).

WARNING: The rest of this post gives spoilers on various puzzles.

I did use the dynamic hints twice — once in the Dark Room and once at Witt’s End (a puzzle I had solved by luck, but I wanted to know how it worked).

Just to clarify how dynamic hints work, in the case of Witt’s End you end up in a maze with this map:

After wandering in circles for enough turns, there’s this message:

YOU HAVE CRAWLED AROUND IN SOME LITTLE HOLES AND WOUND UP BACK IN THE
MAIN PASSAGE.

YOU’RE AT WITT’S END.

DO YOU NEED HELP GETTING OUT OF HERE?

yes

I AM PREPARED TO GIVE YOU A HINT, BUT IT WILL COST YOU 3 POINTS.

DO YOU WANT THE HINT?

yes

DON’T GO WEST.

That is, keep going in any direction except west you like until by random chance (I can’t be certain without looking at the source, but I think it’s 1 in 50) you exit the room.

This is strikingly unfair, except that the place is purely optional except for the Last Lousy Point. Just outside the room is a magazine addressed to Witt, so dropping it at Witt’s End will get a single point and push the final score from 349 to 350 out of 350.

Perhaps I still ought to be somewhat upset, but given in context Adventure was tackled on mainframes as nearly a group game I can understand having one too-hard-to-get point to force the group to band together (like a difficult alternate reality game puzzle). Quoting from Tracy Kidder’s The Soul of a New Machine:

That isn’t the worst maze, however. You can get caught in Witts End and think that you’ll never get out. Some of the engineers at Westborough who had come close to mastering the entire game believed that the only way out of Witts End was to tell the computer you wanted to commit suicide — AXE ME. That worksl you get reincarnated shortly afterward. But you lose points; suicide isn’t the best solution.

What I find most puzzling is that they should have at least been able to figure out the problem from the dynamic hints. Were they playing a port that had them removed?

I am somewhat less forgiving of the endgame puzzle. If you’ve been following along my posts closely (vanity, etc.) you may remember from Crowther’s original this odd message in the source:

blast
BLASTING REQUIRES DYNAMITE.

Woods must have noticed the verb and decided to run it as a puzzle. The endgame section is divided into two rooms with the premise that it’s a stockpile for the various items from the adventure:

YOU ARE AT THE SOUTHWEST END OF THE REPOSITORY. TO ONE SIDE IS A PIT FULL OF FIERCE GREEN SNAKES. ON THE OTHER SIDE IS A ROW OF SMALL WICKER CAGES, EACH OF WHICH CONTAINS A LITTLE SULKING BIRD. IN ONE CORNER IS A BUNDLE OF BLACK RODS WITH RUSTY MARKS ON THEIR ENDS. A LARGE NUMBER OF VELVET PILLOWS ARE SCATTERED ABOUT ON THE FLOOR. A VAST MIRROR STRETCHES OFF TO THE NORTHEAST. AT YOUR FEET IS A LARGE STEEL GRATE, NEXT TO WHICH IS A SIGN WHICH READS, “TREASURE VAULT. KEYS IN MAIN OFFICE.”

Notice how the black rods have rusty marks and not rusty stars. (Rods with rusty stars are in the other room.) It turns out if you pick up one of those rods and type BLAST it will explode. How one can feasibly know this without checking a walkthrough I am unsure. (I checked a walkthrough 15 years ago when I beat this thing, and remembered the solution to this puzzle because of how unfair it was.)

Just so you don’t get the wrong impression, let me discuss two puzzles I liked, although one of them might be considered an edge case on the fulcrum of good puzzle-bad puzzle.

The troll and the bear. There’s a troll guarding a bridge who wants treasure to cross.

A BURLY TROLL STANDS BY THE BRIDGE AND INSISTS YOU THROW HIM ATREASURE BEFORE YOU MAY CROSS.
THE TROLL STEPS OUT FROM BENEATH THE BRIDGE AND BLOCKS YOUR WAY.

Already there’s a bit of a dilemma since maximum points require gathering all the treasures, but it’s possible to sacrifice a treasure here for the purpose of moving the plot along and simply be content with a lower score.

However, in another portion of the cave in the Giant’s Room there are a set of golden eggs with the magic words FEE FIE FOE FOO. Taking the golden eggs elsewhere and saying the four words causes the eggs to disappear; it turns out they teleport back to the Giant’s Room. So you can give the eggs to the troll and teleport them back to your possession, still claiming all the points for the treasure. Working this out was a lovely ‘aha!’ moment: I had puzzled out the egg behavior, knew I somehow had to outwit the troll, and the connection seemed perfectly logical on contemplation. This correlates with Andrew Plotkin’s own experience with the puzzle as a child:

Not too many days later, I solved a puzzle that he described. (How to get past the troll bridge.) I saw the elements, I saw how they could interact, and the answer was obvious. He tried it the next day, and it worked. That was it, for me. I knew this game was doing it right.

After the troll there’s an area with a bear and puzzle that was satisfying (for me) for reasons other than difficulty. Feeding the bear tamed at and you could TAKE BEAR and have it follow you; this came to me fairly immediately. The satisfying part for me was not the taming of the bear but the final result of taking the bear back to the troll:

THE BEAR LUMBERS TOWARD THE TROLL, WHO LETS OUT A STARTLED SHRIEK AND SCURRIES AWAY. THE BEAR SOON GIVES UP THE PURSUIT AND WANDERS BACK.

This brought me a glow of karmic contentment.

The Dark Room. I had a good experience with this puzzle, and it makes for a striking early example of “training” a player in a system.

Deep in the cave there’s a narrow passage:

SOMETHING YOU’RE CARRYING WON’T FIT THROUGH THE TUNNEL WITH YOU.
YOU’D BEST TAKE INVENTORY AND DROP SOMETHING.

In particular the lantern is too big. Going on in:

YOU’RE IN A SMALL CHAMBER LIT BY AN EERIE GREEN LIGHT. AN EXTREMELY NARROW TUNNEL EXITS TO THE WEST. A DARK CORRIDOR LEADS NE.
THERE IS AN EMERALD HERE THE SIZE OF A PLOVER’S EGG!

Going northeast:

IT IS NOW PITCH DARK. IF YOU PROCEED YOU WILL LIKELY FALL INTO A PIT.

So the problem involves getting light to the dark room. I originally thought perhaps the geography was such that one of the pits mentioned around the map connected on the way down, so throwing the lantern down there would get it to the right place. No luck.

Later I was thinking about how magic words worked. Early in the game there’s this iconic description:

YOU ARE IN A DEBRIS ROOM FILLED WITH STUFF WASHED IN FROM THE SURFACE. A LOW WIDE PASSAGE WITH COBBLES BECOMES PLUGGED WITH MUD AND DEBRIS HERE, BUT AN AWKWARD CANYON LEADS UPWARD AND WEST. A NOTE ON THE WALL SAYS “MAGIC WORD XYZZY”.

Saying XYZZY teleports the player back the building, but more importantly (and fortunately a strong temptation), saying XYZZY in the building teleports the player back to the debris room.

Later there’s a room with a marked “Y2” where a hollow voice says “PLUGH”. PLUGH in that room will also teleport back and forth to the building.

I already mentioned the FEE-FIE-FOO-FOE combination worked on the golden eggs.

Given the setup of a.) magic words used for teleporting and b.) magic words related to objects I decided “why not” and while in the room with the “emerald the size of a plover’s egg” I typed PLOVER. And lo:

YOU’RE AT “Y2”.

Typing PLOVER again teleported back.

I feel like I shouldn’t be pleased with this — it is decidedly unfair — I still appreciate that I managed the puzzle with a lateral leap of logic, and that there is a primitive sort of “training in a system” just like was recently discussed on this thread at Emily Short’s blog.

That’s all for now. Was this worthwhile reading? Should I do the same thing for other games? If you’ve played it, how does it match with your own experiences of Adventure? Let me know in the comments.

Posted March 27, 2011 by Jason Dyer in Interactive Fiction

Tagged with

Adventure (350 points): On the best maze ever   2 comments

The twisty maze of passages, all alike:

So, so much better. (See previous installment: Worst Maze Ever.) It was clear early on there was some sensible geography going on with bottlenecks and specific sections. Being forced to drop items like a breadcrumb trail is much more interesting than squinting at the difference between “twisting” and “twisty”.

But also, perhaps more importantly: both dwarves and the pirate are active. The dwarves will pop up at randomly and throw knives at you; you need to throw an axe at them to kill them (it may or may not hit on the first try). Their knives also have a random chance of connecting and killing you back.

THERE IS A THREATENING LITTLE DWARF IN THE ROOM WITH YOU!

ONE SHARP NASTY KNIFE IS THROWN AT YOU!

IT MISSES!

YOU ARE IN A MAZE OF TWISTY LITTLE PASSAGES, ALL ALIKE.

throw axe

YOU KILLED A LITTLE DWARF. THE BODY VANISHES IN A CLOUD OF GREASY
BLACK SMOKE.

The pirate will steal any treasure you might be holding and store it “deep in the maze” as he says.

OUT FROM THE SHADOWS BEHIND YOU POUNCES A BEARDED PIRATE! “HAR, HAR,”
HE CHORTLES. “I’LL JUST TAKE ALL THIS BOOTY AND HIDE IT AWAY WITH ME
CHEST DEEP IN THE MAZE!” HE SNATCHES YOUR TREASURE AND VANISHES INTO
THE GLOOM.

This led (for my game) to some fascinating tactical and story implications. My first time in the maze I entered by dropping from the stalactite (meaning I couldn’t get back or find an exit easily) but I also left behind the axe so I could have more room for items to drop to help make a map. This was a mistake: dwarves started popping up and chasing me around the maze. It turns out you can duck into a dead end and they won’t follow, but will still be waiting for you outside. So a story came out where I was playing a game of cat-and-mouse, ended ignominiously by running out a dead end and getting skewered by a pair of dwarves waiting for me.

My second attempt came from the West End Hall (where it was easy to go back) but things were still interesting because there aren’t enough “normal” items to drop in every room for mapping so I had to resort to treasures. Of course the pirate can at any time snatch those treasures. While collecting treasures for the maze I had to balance out the reward from having more items to use vs. the risk of wasting more time vs. the fact they might be snatched while in the maze anyway. Eventually I realized the geography was sensible and started moving items that were already present rather than bringing in new ones (perhaps the only good effect of the twisty maze of passages all different is that I came in paranoid the geography would be bizarre and didn’t realize this would work straightaway).

Having an actual goal (the pirate’s lair) helped both with motivation and with story. As I kept getting treasures snatched and battling my way through dwarves I felt a tangible sense of getting closer and closer to the source.

The (general) elimination of NE / NW / SE / SW led to a much faster traversal even with the extra story involved. The map above took about an hour to make rather than three hours. Also interesting is that the pirate’s lair requires a northwest exit, so is somewhat “hidden in plain sight” — by that point players may have stopped even trying those directions. However it is still fair because the pirate does explicitly state he’s hiding his treasure deep in the maze, so it makes a puzzle of sorts based on one compass direction.

Everything put together for the most satisfying maze experience I ever recall having in interactive fiction. I can understand a little more why authors at the time wanted to copy it, but most that followed didn’t have the interesting considerations mentioned above.

Posted March 25, 2011 by Jason Dyer in Interactive Fiction

Tagged with