Announce: Miniventure #2 CATSEYE Available

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

My second in a series of Inform 10K adventures. This one was
BRUTAL to code in just 10K. I spend hours pairing it down and
pulling every trick I could think of (and a few I couldn't).
This one clocks in at just 22 bytes short of 10K and is really
packed tightly. I think it's a bit more complex than Moonglow,
but probably still solvable in under an hour.

http://www.gis.net/~daveber/miniventure

Comments, as always, are welcome.

--
Dave Bernazzani
Join the Commodore 32 Inform minigame contest:
http://www.gis.net/~daveber/minform/c32.htm

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

It took me more than half an hour to guess the verb to put the jar
through the cat-door.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

And it'd be nice if when I tried things like "tie yarn to y" there was
some kind of message, any kind of message, some indication whether I
had in fact tied it or not...

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Sorry, I should have added spoiler space on that first one. I'm up
kinda late here.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

(sent kill messages on those; hope it worked b/f I ruined anybody's
experience)

S
P
O
I
L
E
R
S
P
A
C
E

I found it particularly annoying that "tie x to y" didn't give any
response at all (except in the case where you tied the correct ends
together); I didn't know if I'd tied the stuff or not. Also, it took
me over half an hour to guess the correct verb to use to push the jar
through the cat door. You should change the library messages so that
it's easy to tell whether a verb is even implemented or not -- all
you'd have to do is use the "Can't" message instead of the "Huh?"
message.

It's an interesting idea to make a small game, but the 10k limit
apparently restricts the games to players who'd use the exact first
verb that you'd come up with, making it incredibly frustrating for the
rest of us. From what I can tell about the construction of the mInform
library, perhaps you'd be better off discarding the inform libraries
altogether and just using inform as an assembler; it looks like you had
to leave a bunch of stuff in that you're not using just because other
things depend on it.

Alternatively, you could take out all the multi stuff from grammar.h,
which should make a little more room for you to put in synonyms.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Daniel wrote:

> (sent kill messages on those; hope it worked b/f I ruined anybody's
> experience)
>
> S
> P
> O
> I
> L
> E
> R
> S
> P
> A
> C
> E
>
> I found it particularly annoying that "tie x to y" didn't
> give any response at all (except in the case where you
> tied the correct ends together);

That's a clear bug - I had a response in place but it wasn't getting called.
I will fix this for the next release.

> Also, it took me over half an hour to guess
> the correct verb to use to push the jar
> through the cat door.

I have 5 different syntaxes that work - the most common of which is PUT JAR
INTO <anything that resembles the door> or PUT JAR IN <anything that
resembles the door>. I don't use any uncommon verbs (see below) and try to
get reuse out of a common set of 7 verbs for most puzzles.

> You should change the library messages so that
> it's easy to tell whether a verb is even implemented
> or not -- all you'd have to do is use the "Can't" message
> instead of the "Huh?" message.

That I can probably do, I have to see where I make the decision as whether
the verb is understood. Part of the pairing down was a combination of
decision points where the parser gets in trouble. I should at least be able
to change out the message in this case - it shouldn't cost more than a
handful of bytes.

> It's an interesting idea to make a small game, but the 10k limit
> apparently restricts the games to players who'd use the exact first
> verb that you'd come up with, making it incredibly frustrating for the
> rest of us.

Alternatively, it restricts the verbs so that the same small set of verbs is
used for many actions. Like PUT for the jar and the door puzzle. You can
actually get by in both games with knowledge of only 12 simple verbs.

>From what I can tell about the construction of the mInform

Miniventures do NOT use mInform (a common misconception). The mInform
library is a semi-full featured library - retaining most of what makes
Inform great. mInform is 19K and as such is a general library for use by
folks that want a small game without the overhead of multi-object processing
and a streamlined library message system. The mInform library has been
released to the public (I will put it up on the if-archive soon, for now
it's only at my page) and is the one featured for the C32 contest I'm
running.

Miniventures use a deeply restricted parser that was originally based on
mInform but has morphed into something else. These games are more of an
internal project I thought would be fun - built using a library called
"microform" (which you can see from the header line of MOONGLOW or CATSEYE)
which is truly bare-bones at under 6K. I didn't release microform to the
public since it is amazingly restrictive and requires knowing some serious
tricks to make a short game fit in those constraints. Furthermore, to help
squeeze games down under 10K I'm using a custom built version of the Inform
6.15 compiler that removes the run-time error checking (~400 bytes) which is
only necessary during debug (so I use the normal 6.15 to build and test,
then the modified 6.15.1 compiler to get it really small for distribution).

> perhaps you'd be better off discarding the inform libraries
> altogether and just using inform as an assembler; it looks like you had
> to leave a bunch of stuff in that you're not using just because other
> things depend on it.

Not really. I have done my own parser and it's only 3K smaller but leaves
out a lot of things like good inventory management and other object oriented
features like before and after processing. The 6K library may be
bare-bones, but it really does contain a ton of processing that is hard to
replicate.

> Alternatively, you could take out all the multi stuff from grammar.h,
> which should make a little more room for you to put in synonyms.

It's well gone. Grammar.h only has 7 simple stock verbs (quit, take/get,
drop, inventory, examine, look, go/enter). A few "specialized" verbs (though
still common) are added for each game. How else do you think I've managed
to get these Miniventures into the shoe-string budget of 10K!

I greatly appreciate the comments and feedback. I will correct the TIE verb
to produce something useful for trying to tie other objects together.

--
Dave Bernazzani
Join the Commodore 32 Inform minigame contest:
http://www.gis.net/~daveber/minform/c32.htm

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Here, daniel_mcl@hotmail.com <daniel_mcl@hotmail.com> wrote:
>
> It's an interesting idea to make a small game, but the 10k limit
> apparently restricts the games to players who'd use the exact first
> verb that you'd come up with, making it incredibly frustrating for the
> rest of us. From what I can tell about the construction of the mInform
> library, perhaps you'd be better off discarding the inform libraries
> altogether and just using inform as an assembler; it looks like you had
> to leave a bunch of stuff in that you're not using just because other
> things depend on it.

I haven't been playing these -- I'm barely on schedule with the IFComp
games. However, it sounds like the 10k limit makes an interesting
programming challenge, but not a good design challenge. That is, the
minimalism doesn't produce a focussed playable game. (In contrast to,
say, a 55-word story limit, which can produce a micro-short story
which is as readable as a longer story.)

If I were trying to invent a "minimalist" IF form analogous to
short-short stories, I'd limit the amount of code *over the standard
library*.

(Of course this is hypothetical -- I haven't tried it, so I can't say
it'll work any better from the player's point of view.)

--Z

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
* Make your vote count. Get your vote counted.

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Andrew wrote:

> However, it sounds like the 10k limit makes an interesting
> programming challenge, but not a good design challenge.

For me, it has been both an programming challenge and an interesting design
challenge. The programming of these Miniventures has taken less time than
the design (7 or 8 hours of design work though some of that was suffered
through during Red Sox games, 3 hours coding - though Catseye was a little
longer). My longer IF Comp entry (IDENTITY) was the reverse - and saw about
15 hours of design time (maps, initial room descriptions, NPCs, puzzles,
core objects, etc) vs. another 60+ in coding (more than half of that in bug
fixes and other suggestions from beta testers). My second full game
(GAUCHO) is still in design but will likely top out at 15-20 hours as well.
Working with the size limitation does force the designer to think about the
structure of his game - about balancing a story with puzzles and other
interactive elements. I can't say that either of my games are particularly
stunning, but I think they have made me a better designer while only
marginally improving my Inform coding skills (though the parser pair-down
effort has given me good knowledge of the Inform libraries).

Trying to squeeze a game into 10K is a definitely a programming challenge.
Designing a playable interesting game with a little depth (very debatable
here!) may be a greater challenge. My Inform programming skills are coming
along, but my IF design skills need work. I'm still hopeful that a
brilliant designer will step up to the plate and design one of these
Miniventures so that it is challenging for the designer/implementer and
equally fun and challenging for player.

In any event, the Miniventures were not so much about creating the
equivalent of "short-short stories" as they were about seeing what could be
done with the Z-machine in 10K of code - putting them amongst the smallest
text adventure games in existence (there are several 2K and 4K games
available) and certainly amongst the smallest Z-code text adventure games
(along with The Dragon and the Troll).

The C32 contest is a separate entity (though I've perhaps muddle things a
bit). Here I was trying to write Inform games for the C64 and found that
the stock Inform library was simply too bulky to be effectively used on the
machine. Once I got the libraries reasonably reduced, it seemed that they
could be used by anyone that wanted a streamlined (both in terms of size and
run-time overhead) library for use with their Z-code games for whatever
reason. In an effort to promote the use of the library but more importantly
because it seemed like a neat challenge to have folks design with real-world
constraints in mind (much like the early pioneers of IF), the C32 contest
was started. I still think the contest will be worthwhile. It will be
interesting to see what folks can do given the constraints of the mythical
Commodore 32 Z-Machine.

--
Dave Bernazzani
Join the Commodore 32 Inform minigame contest:
http://www.gis.net/~daveber/minform/c32.htm

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

Perhaps you should include a list of the verbs that are actualy defined
in the game, then, so long as they don't give one of the puzzles away?
I thought some of the things I had to do in "Moonglow" were decidedly
counterintuitive, ...


S
P
O
I
L
E
R
S
P
A
C
E


for one thing, the command "enter door" to exit the orb was highly
nonintuitive, and it didn't really make sense to me why the alien
wasn't willing to take the gloves from you inside the orb and then exit
rather than just standing there baffling you, although the player
probably hits on the solution accidentally pretty soon. (The first
time, I thought he wanted me to bring him the broken glass or
something.) It also didn't make a whole lot of sense that the cat
couldn't carry the ball of yarn around in Catseye; since cats are used
to pushing around balls of yarn, I was *sure* that this would help in
the solution of a puzzle (this really bothered me -- obviously "push"
was implemented b/c of the refrigerator, but when I wanted to push
something under a door or something I had to use "put" instead; I spent
a long time trying variations on "push x under y," "push x through y,"
"push x into y," "push x north," etc. When there are only 15 or so
verbs in the entire game, and you know that one of them is the most
appropriate verb for what you're trying to do, you're not very likely
to try the others out.)

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

daniel_mcl@hotmail.com wrote:

>Perhaps you should include a list of the verbs that are actualy defined
>in the game, then, so long as they don't give one of the puzzles away?
>I thought some of the things I had to do in "Moonglow" were decidedly
>counterintuitive, ...
>
>
>S
>P
>O
>I
>L
>E
>R
>S
>P
>A
>C
>E
>
>
>for one thing, the command "enter door" to exit the orb was highly
>nonintuitive,

Actually, GO DOOR works and since the door is described (with
sunlight streaming through) it seems reasonable. But I did add
EXIT ORB to Release 3 which has been out for a while due to
several email requests.

>and it didn't really make sense to me why the alien
>wasn't willing to take the gloves from you inside the
>orb

He wasn't interested in the gloves until he realizes that he
needs them to fix the damage when outside the ship. Though,
admittedly, that may be deeper logic than the game requires

>It also didn't make a whole lot of sense that the cat
>couldn't carry the ball of yarn around in Catseye;

I actually allowed it at first. Not sure why I didn't allow it in
the released version. It's not strictly necessary but may make
for a good red herring. Point well taken...

>"push" was implemented b/c of the refrigerator,
>but when I wanted to push something under a door
>or something I had to use "put" instead;

PUT or INSERT works. I will add PUSH/MOVE as well since it will
only require a few bytes as those are already implemented.

Thanks again for the feedback!

--
Dave Bernazzani
Join the Commodore 32 Inform minigame contest:
http://www.gis.net/~daveber/minform/c32.htm

 

[Dada]

Archived from groups: rec.games.int-fiction (More info?)

>> S
>> P
>> O
>> I
>> L
>> E
>> R
>> S
>> P
>> A
>> C
>> E
>>
>> I found it particularly annoying that "tie x to y" didn't
>> give any response at all (except in the case where you
>> tied the correct ends together);
Couldn't tie the yarn to the B. took a while befor I tried to tie tie B to
the yarn... That worked but doesn't seem gramatically correct...

Also in the fire place the hole, ash tray is not recognised. You should be
able to drop the spoon in the hole.

It's fun to play but sometimes too much guessing how to get it to do what
you wnat it to do. Sometimes you give up thinking the solution you thought
of must be wrong becouse you can't get it to work. Only after reading the
messages in this thread I understood what I wanted to do was correct and I
kept trying.

Dada.


Dada

 

[Guest]

Archived from groups: rec.games.int-fiction (More info?)

(serious spoilers below)

"Dada" <a@b.c> wrote:>
> S
> P
> O
> I
> L
> E
> R
>
>
> S
> P
> A
> C
> E
>
>

> Couldn't tie the yarn to the B. took a while befor I tried to
> tie tie B to the yarn... That worked but doesn't seem
> gramatically correct...

That is a bug. You can, in fact, tie the yarn to the spoon or the spoon to
the yarn in either order. It doesn't let you tie the yarn to the necklace
(but does allow you to tie the necklace to the yarn/spoon) which is wrong
and will be corrected.

> Also in the fire place the hole, ash tray is not recognised. You should be
> able to drop the spoon in the hole.

I assume you meant ash trap. I found a way to get back 20 or so bytes and I
can probably add "drop" functionality as well which will improve the "guess
the verb" issues that tend to be a problem with such a restrictive game.
I'll look into this.

I continue to appreciate all the feedback on these 10K Miniventures - both
here and in private email. I'm glad folks have enjoyed the games - it's not
an easy design-task given 10K bytes, but it's been very interesting. The
next Miniventure will be forthcoming later in November - and is (fortunately
for the rest of you!) not designed by me

I also want to stress that the upcoming C32 contest is not related to the
10K Miniventures. The 10K Miniventures have been more of a micro-game
experiment to see what could be done given 10K of Z-machine code -- and
given the severe limitations does end up feeling like a cross between a
Scott Adams game with a bit of early Infocom style. The C32 contest games
that are currently under development by stalwart designers offer a bit more
freedom from the severe limitations imposed by a 10K threshold and as such,
should be better suited to the IF community at large. These authors are
working hard to produce quality games under the guidelines and limitations
of the mythical Commodore 32 Z-Machine, and I hope folks will take the time
to download and try them when they are released on December 1, 2004.

--
Dave Bernazzani
Join the Commodore 32 Inform minigame contest:
http://www.gis.net/~daveber/minform/c32.htm