[MUD-Dev] Re: Bad Game Designer, No Twinkie! -- By Ernest Adams

J C Lawrence claw at under.engr.sgi.com
Wed May 20 11:19:53 CEST 1998


On Mon, 18 May 1998 17:46:56 -0700 
T Alexander Popiel<popiel at beldin.snugharbor.com> wrote:

> In message: <199805182342.QAA08300 at under.engr.sgi.com> J C Lawrence
> <claw at under.engr.sgi.com> writes:
>> "Bad Game Designer, No Twinkie!"
>> 
>> By Ernest Adams

> While I agree with nearly everything Ernest says, I'd like to take
> exception to one of his examples:

I disagree with almost the entire article if taken as an absolute.
While it is easy to come up with supporting examples for all of his
points (yes, many mazes are pointless), it is much more rewarding to
come up with examples which thwart his assertion (such as mazes which
are purposeful, or a token which causes a far distant event).  Its not
terribly difficult, and it prompts some *really* devilish ideas.

What I find much more interesting is how much of my early game design
work was, in retrospect, obviously in counter-reaction to the
assertions in his article.

>> Boring and Stupid Mazes
>> 
>> The original text adventure, Colossal Cave, had two mazes. One was
>> a series of rooms each of which was described thus: "You're in a
>> maze of twisty little passages, all alike."  The other was a series
>> of rooms described as, "You're in a twisting little maze of
>> passages, all different" (or "You're in a little twisty maze of
>> passages, all different," or "You're in a maze of little twisting
>> passages, all different," etc.). These were the prototypical boring
>> and stupid mazes.

> I believe that both of these mazes were incredibly well done (in
> sharp contrast to the hundreds of boring and stupid mazes that have
> followed).  

I never observed the pattern in those mazes.  What I do recall was
getting a "feel" for them such that I was sem-automatically able to
navigate thru and about them.  I seem to do that a lot with games: I
don't know how it works, but I can subconciously predict exactly what
will happen anyway.  It makes for a lot of dodging bullets in
shoot-'em'ups before they've actually been fired.

> Yes, appreciating these properties of the mazes requires fully
> exploring them, and then abstracting your perception of the mazes.
> A helping of graph theory adds spice to them.  I don't see this as a
> failing.

Barrier to entrence.

I've done three mazes which I particularly liked.  The most notable of
which was the Blue Grass Path.  Fortress Fract was just too damned
complex and provided too little runtime data to be playable, despite
being an intellectual delight.

<URL:http://www.kanga.nu:80/~petidomo/lists/mud-dev/1997Q2/msg00726.html>

--<cut>--

You would have loved the stuff I did for ShadowHouse:  

  The Blue Grass Path was a transdimensional path.  Very simple
really.  Pretty well in any room in the entire game which contained
the word "blue" in the description, or for which some form of
"blueness" was implicit, a player could enter "BLUE" as a command and
be automagically transported to the Blue Grass Path.

  The BGP itself was simplistic in representation: a path floored with
faintly glowing blue grass, soft to the touch but
unbreakable/untearable, surrounded with faintly blue fog, with the
path extending out of sight ahead of and behind you.  The "BLUE"
command took you on and off the path.  Once on the path the only
movement commands available were "FORWARD" and "BACK".  No other
directions were possible.

  The other side feature was that the BGP was geometrically
inconsistant.  Getting on the BGP at say the Armoury and then
immediately leaving it again would most likely land you somewhere else
entirely (if in the same place every time).  Additionally the BGP
itself was non-linear and was directionally sensitive.  Thus while
getting on the BG at the Armoury and then getting off immediately
would land you (say) at the Pigs Tough, getting on the BGP at the
Armoury and going FORWARD and then BACK and then off would quite
likely land you somewhere other than the Pigs Trough.

  ie the BGP was essentially a directionally sensitive maze where the
entrances and exits from any given node commonly pointed to different
locations.  

  The result, and the intention, was that the BGP was incredibly
annoying and tedious to map, but had the ability to both transport you
vast distances across the land almost immediately, as well as to
transport you to otherwise impossible or very difficult-to-reach
locations with minimal effort.  Thusly it was the most disliked,
complained, lauded and generally discussed feature of the game, while
simultaneously claming the majority ot the player's attention whilst
playing.  It was sort of an "I hate this bloody thing!", mixed with,
"Mein Gotte!  Look what it can do!".

  Fortress Fract involved retrieving the ever lost and ever wandering
Princess Julia and returning her to King Mandel.  Simple on the face
of it, but getting to Fortress Fract involved either solving a large
portion of the Blue Grass Path, or re-constructing the Human Powered
Catapult with the appropriate human skins (what it was tied together
with) and climbing aboard to be shot towards the castle (weight
dependant -- how much you carried determined where you landed, and
whether you lived or died upon landing).

  Once in the Fortress (which resided in the middle of an impassible
lake (monsters and vortexes -- thus the catapult or the BGP))
returning to shore was equally difficult.  

  The fortress itself was another variation on a maze, but this time a
maze where the player's actions mutated the maze at runtime.  The
basic pattern was simple: The floors of the fortress were movable. 
Whole sections of the building floors could be made to tilt such that
the floors became gently sloping ramps between other floors.  Thus
(say) you could stand on floor #5, pull a lever and walk east onto
floor #3, or west onto floor #7, or try and get off half way at floors
#4 or #6.  To make it more fun, typically the slope in the floor was
only detectable by dropping rollable objects and seeing which way they
rolled.  Similarly, the floors were not numbered or otherwise
identified, and were slightly difficult to distinguish from each
other.  The other side of which of course was that pulling the lever
(or doing whatever for that instance, different things moved the
floors in different places), or having some other player do it
elsewhere in the building might drop the floor above on you, or maroon
you in a vircle of ramps all of which lead to blank walls.

  The fact that other players wandering about could also move or
re-orient your floor while attempting to move theirs made things
interesting, as did the fact that moving some floors also moved many
other floors.  

  Note: "floors" as above typically meant a small section of the
building, tho occassionally did mean the entire floor of the whole
building (which was 64x64 rooms).

  Then, spaced about the periphery of the Fortress, and centrally
located, were towers, whose floors could be made to move up and down
ala elevators.  This was occassionally deadly -- such as standing on
the bottom floor of a tower and having the floor above dropped on you,
or visa versa at the top.

  Oh yeah, and the entire floor at each level was also formed like one
of those tile games (the one with the square tiles, one is missing,
that you slide the tiles about to re-organise a scrambled picture or
some such) with 8x8 room tiles.  Thus in various ways you could move
the floor tile that you were standing upon about the floor, (ie move
the hole) whilst also moving other tiles about (or have the tile you
were on moved by someone else).

  At the bottom of all this all wandered the lost Princess Julia.  At
the top in the Throne Room stood King Mandel.  Find Julia, persuade
her to follow you, re-catch her every time she wanders off, and get
her all the way back to Mandel, all while a dozen or so other players
were attempting the same (and thus moving the building about on you),
would earn enough XP to throw you almost a third of the way to wizard
(roughly 4 or 5 whole levels in one shot).  Oh yea, and there were
lots of other really nasty point opportunities in the process as well
-- such as reorganising a particular floor into a specific pattern, or
reorganising multiple floors so that particular tiles were above each
other, etc (you could see the ceilings and thus ID the floor tile
above you).

  Then there was Mobious Row, and the Mud Village, the Cheese Factory,
the White Oak Tree (with its elephant powered elevator and a 20+ room
Tarzan's tree-hut), etc.  

--<cut>--

The BGP had all sorts of nice features: It was inherently
unpredictable, deliberately unmappable, any observed feature of it was
liklely unrelatable to anything other observed feature, and incredibly
useful.  Best of all it didn't offer anything that couldn't be done
otherwise, thus there was never an mandate to either use or solve the
BGP.  Its only value was that it offered safe shortcuts -- if you
duplicated your motions on the BGP *EXACTLY* every time.  Yes, you
could use the Human Powered Catapult to get the Fortress Fract, but it 
took a lot of work and time for little reward in the process.  Taking
the BGP OTOH took single-digit seconds and zero effort once you knew
the route.

Mobious Row was more "fun", far more intrigueing, but less valuable
and little used in the game (nobody ever seemed to think of a way to
take advantage of it).

> The countless boring and stupid mazes that I've seen seem to be made
> by people who are trying to mimic the style of these two classic
> mazes, without noticing the higher level of abstractions available.
> (Either that, or they just took a standard drawn maze, dropped it
> onto a sheet of dual-lined paper, and made each little square a
> room.  Yawn.)

<nod>

--
J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...

--
MUD-Dev: Advancing an unrealised future.



More information about the mud-dev-archive mailing list