Graphical IF, IF theory, and inform/tads
Mike Rozak
designed for amateurs. For awhile I have been THINKING about turning this
into one component of a larger interactive fiction authoring program. There
is a lot I can potentially do with IF (given enough time) because I have my
own modeller and renderer. (I also wrote my own text-to-speech engine
(http://www.mxac.com.au/m3d/tts.htm), and have experience with using speech
recognition.)
For many months I have been pondering the following questions:
1) Will an amateur IF author spend the time to not only learn IF, but to
learn 3d modelling? And will he/she be able to write an IF and create the
models in reasonable time? (After all, graphics take a long time to
produce.)
2) I have tried to produce a scale of "interactivity" for fiction in an
attempt to devise a good balance between plot and interactivity:
a) Linear - A book, movie, etc.
b) Nodal - Choose-your-own adventure books
c) Text/graphical adventure - Rooms, objects, characters, etc.
d) RPG/MMORPG - I label these as more "interactive" because while they lack
any plot (in part due to their level of interactivity) they don't constrain
the choices of the user beyond the game physics. Text/graphical adventures
often constrain the user's actions. (For example: I'm playing Uru now, and
really want to jump over some of the fences in the game. I'm not allowed to
because of story reasons. In a RPG/MMORPG I could jump over them, or maybe
even bash them down.)
Going from (a) to (d) results in an inevitable loss of plot, and an increase
in simulation. The more freedom the player has, the more difficult to
maintain a plot, and the more necessary simulation becomes.
Going from (a) to (d) also seems to increase the work load on the author.
Personally, I'm interested in something in between (b) and (c) because that
would provide a good balance between plot and interactivity. Perhaps due to
a mental block, I can't figure out a good infrastructure that would fit
between nodal and text/graphical adventure. (HTML pages are basically (b).
TADs/INFORM work well for (c)). Does anyone know what's in-between?
3) If I go down this path I will need to write an adventure programming
language. I have looked though TADs and INFORM, and several other scripting
languages. The basic OO architecture of TADs/INFORM seem to work very well
for IF. I would make some changes:
- GUI for modifying the code, enabling me to display object hierarchies,
created automatic help for objects, etc.
- Unicode, and string table built into the system so that localization is
easier
- Don't assume a primarily textual interface for playback like TADs and
INFORM do. Instead the IF playback interface would largely be 3d
visuals/animations, sound FX, and speech (recorded or synthesized).
(Personally, I hate reading text on a computer, especially all those damn
hand-written journals in Uru.)
- Don't assume purely textual input. It will need to handle text of course,
but also menus, click on objects, and speech recognition.
- Avoid 100% open ended commands. Instead, I'd query an object to enumerate
the commands it supports. I would display the top-4 commands (based on
context) in a short-list menu. The other 100-1000 commands would be
accessible by typing in keywords or scrolling through the list. Some
commands would only be visible if a specific keyword is typed in (to allow
for secret passwords, etc.) There are two reasons for doing this: 1) Take
advantage of a GUI since users expect to click most of the time, and 2) take
advantage of speech recognition, which requires a more limited syntax (CFG)
to improve speech recognition accuracy.
3a) Does anyone have any suggestions for improvements to existing languages?
3b) TADs/INFORM don't seem to handle conversations with NPCs very well. (The
AI functionality seems to be flexible enough.) The only alternative I've
found are solutions like Eliza and other chatterbots (which although
promising, would drive most users and authors insane) or a menued
conversation (as used in most RPGs/MMORPGs). I have also been pondering
search engine anlgorithms for accessing common knowledge, but this may not
work well. (Ex: Questions like "Who is the king?" should be known by every
NPC, although they will all give subtly different responses.) Does anyone
have any suggestions about this?
Thanks in advance for your feedback.
--
Mike Rozak
www.mXac.com.au
Mike Roberts
> For many months I have been pondering the following questions:
An excellent resource for researching these sorts of questions is the
groups.google.com archive of raif postings going back over the past fifteen
years or so. You might also be interested in Nick Montfort's just-published
book Twisty Little Passages (available at any on-line bookseller), and the
IF Theory book (www.iftheory.com), when it's published. (I don't mean to be
glib; it's just that none of your questions have answers that can be
summarized in a posting, if they even have answers per se.)
--Mike
mjr underscore at hotmail dot com
Mike Rozak
Two book references - Very helpful. (Amazon.com is getting more of my money
as I write this.) I found a few books about IF (Hamlet on the Holodeck -
kind of IF), MUD (one about to be released), and a few about MMORPG on
Amazon, but not Twisty Little Passages. I have also Emailed the IF Theory
publisher. Thanks.
Questions have answers that can't be summarized in a posting... the
questions really wouldn't be worth asking if they had simple answers. (Okay,
my last quote is a bit too philisophical.) None of my questions lead to
simple answers, so references to pre-existing materials is a good shortcut.
There's no point retyping what has already been said.
Thanks
--
Mike Rozak
www.mXac.com.au
"Mike Roberts" <mjrUND...@hotmail.com> wrote in message
news:wyTwb.22$wz....@news.oracle.com...
Quintin Stone
> 1) Will an amateur IF author spend the time to not only learn IF, but to
> learn 3d modelling? And will he/she be able to write an IF and create
> the models in reasonable time? (After all, graphics take a long time to
> produce.)
3D modelling is hard to do well and very time-intensive. I say this as a
guy who's done 2 FPS mods. However, a library of pre-built standard
objects (people, furniture, buildings, etc.) with a variety of skins might
encourage amateurs to try their hand at it.
> 2) I have tried to produce a scale of "interactivity" for fiction in an
> attempt to devise a good balance between plot and interactivity:
>
> b) Nodal - Choose-your-own adventure books
> c) Text/graphical adventure - Rooms, objects, characters, etc.
>
> Personally, I'm interested in something in between (b) and (c) because
> that would provide a good balance between plot and interactivity.
> Perhaps due to a mental block, I can't figure out a good infrastructure
> that would fit between nodal and text/graphical adventure. (HTML pages
> are basically (b). TADs/INFORM work well for (c)). Does anyone know
> what's in-between?
IF is really flexible and has a fairly big range in your scale. For
instance, there are some very linear text games out there, such as
Photopia and (I've heard) Ramses. Have you played these? Are they
somewhat near what you were thinking?
/====================================================================\
|| Quintin Stone O- > "You speak of necessary evil? One ||
|| Code Monkey < of those necessities is that if ||
|| Rebel Programmers Society > innocents must suffer, the guilty must ||
|| st...@rps.net < suffer more." -- Mackenzie Calhoun ||
|| http://www.rps.net/ > "Once Burned" by Peter David ||
\====================================================================/
Mike Rozak
news:Pine.LNX.4.44.03112...@yes.rps.net...
> 3D modelling is hard to do well and very time-intensive. I say this as a
> guy who's done 2 FPS mods. However, a library of pre-built standard
> objects (people, furniture, buildings, etc.) with a variety of skins might
> encourage amateurs to try their hand at it.
Difficult to do well - No matter what tool I make, modelling (or any
activity) will always be difficult to do well. I don't expect amateurs to
create professional-quality models. But they may not need to, since even a
mediocre model may be good enough to improve the IF experience. I'm not sure
of this though.
My 3d modeller has the following features, some of which help the difficulty
problem:
- Easy to create buildings (just a bit trickier to create than a House
design app like Punch Home Design, but more powerful)
- Easy to create landscapes (like bryce)
- It will come with a library of basic furniture and other goodies. Basic
furniture is also easy to create as long as there isn't a lot of detail.
- Easy to pass around object models between people, and then apply different
textures. If enough users are producing and distributing models, finding any
common object becomes a trivial matter, although distribution rights are
slightly trickier. (I suspect that most hobbyist model creators would be
happy for their models to be used elsewhere).
- Working on automatically generating people... this is tricky. There are
some apps out there that do this, especially for generating faces. Right now
I have sub-D surfaces and bones, and am currently adding easy-to-add eyes,
hair, teeth/jaw. Producing a model of a person is 4+ hours of work, which is
too much. Eventually I'll add automatic face generation, clothing sets, etc.
But, this takes time to program. Ultimately it'll be dial-a-person and then
pick an outfit. (One tricky aspect is that I also want users to be able to
build their own dial-a-X, so they could have dial-a-Klingon or dial-a-wolf,
etc.)
- Bizarre landscapes/structures like are seen in Myst - These are actually
tricky to impliment in my app because my code is designed to produce more
realistic landscapes. Interestingly, bizarre landscapes are easier to
generate with more traditional packages (like Maya) because a bizarre
landscape/structure is a 2D surface that has been organically deformed. My
app can do this too, but my app makes generating normal houses/lanscapes
even easier.
> IF is really flexible and has a fairly big range in your scale. For
> instance, there are some very linear text games out there, such as
> Photopia and (I've heard) Ramses. Have you played these? Are they
> somewhat near what you were thinking?
Thanks. I'll look into these.
Not having seen either, here's how I'd characterize my desire: Traditional
adventure games are a series of interactive puzzles with a plot thrown in to
glue them together. I want a plot/story with occasional interaction (not
necessarily puzzles) that enhances the plot/story. This is a vague and fuzzy
idea in my mind, so I can't explain too clearly.
I am thinking that maybe the story will progress linearly in most places,
perhaps like books-on-tape combined with a slide show of 3D images (since
animation is time consuming). Maybe the reader could branch off into
side-paths, or maybe choose which character to follow. The result is similar
choose-your-own adventure w/visuals and audio. Most books wouldn't allow
major plot-changing interaction on the user's part, but an author could code
this in.
However, at times I want more interactivity in order to bring the user into
the story more.
Here's an off the cuff example based on a well known book, The Hobbit. In
the scene where bilbo is lost in the caves, just before he meets gollum:
Stop the narrative and go into a more adventure-like mode of interaction.
Place the player in a maze (chiche) along with bilbo, wandering around
trying to get out. During the wandering bilbo will be making conversation,
drawing the player closer to his character. The fact that the user is also
lost in the maze enables him/her to emphasize more with bilbo's situation.
Ultimately, the maze will find a way of dumping the user into gollum's lair.
It could either resume to narrative, or the riddles could be solved by both
the user and bilbo. (Perhaps the riddle is spoken, and waits for the user to
type a response. If the user doesn't answer in time maybe bilbo whispers a
suggestion in the user's ear - which may be wrong or right. Maybe bilbo
answers some outright after 10 seconds, or maybe bilbo tells the user "I
don't know. What do you think?") The whole experience makes the user feel
like Bilbo's comrade, empathize with Bilbo's plight, and be drawn into the
story.
The user has to know from the start that he/she can't really influence the
direction of the story so that they don't get their expectations up though.
In one of the thousands of rec.arts.int-fiction messages that I scanned
through, one person mentioned their disappointment when they thought they
could change the plot, but were really on a well defined track.
By the way - If I wrote the software I'd want to to handle a wide range of
styles, including traditional adventure and choose-your-own adventure.
One aspect of TADs/INFORM IF that dsiturbs me is the institutionalization of
space (rooms), and objects. What's interesting to note is that in most
books, and even in real life, while every action takes place in space, and
people are always carrying objects (like clothes and jewelry) the space and
the focus aren't the focus of the narrative. Books usually gloss over
location/objects unless it's relevent to their plot. Real life people, while
aware of location/objects, likewise ignore them until they're relevent. When
space/objects are intitutionalized it encourages authors to emphasize them
in the fiction, and maybe discourages some of the more complex
relationships/states of the story/plot.
A choose-your-own adventure does not have this problem. However, it has
other problems (being too constrained, for example).
That's why I was wondering if anyone has come up with an in-between. My
archive seaches haven't revealed anything though.
--
Mike Rozak
www.mXac.com.au
M.D. Dollahite
>learn 3d modelling? And will he/she be able to write an IF and create the
>models in reasonable time? (After all, graphics take a long time to
>produce.)
Well, I can tell you that I for one have already taken the time to learn both.
However, I may be something of a special case, as I'm rather more fluent in
computer languages than human languages, and my 3D software of choice (POV-Ray)
uses a scripting language rather than a point-and-click system.
As for taking a long time to make graphics, that depends on the facilities your
modeler provides. For example, I was able to create a great looking natural
tunnel in just a few minutes using POV-Ray, by stringing blob components along
a spline path with a bit of random jitter. I just programmed povray to create
a bunch of blob components and slide them along the spline; if I'd had to
create and position each sphere by hand, it would have taken forever. So
basically, the more your modeler is able to build complex structures
automatically, the faster people will be able to construct scenes with it. And
I don't mean prefabricated objects, nor random scene generation; I mean being
able to offload busywork onto the computer, like in my example.
>2) I have tried to produce a scale of "interactivity" for fiction in an
>attempt to devise a good balance between plot and interactivity:
>
>a) Linear - A book, movie, etc.
>b) Nodal - Choose-your-own adventure books
>c) Text/graphical adventure - Rooms, objects, characters, etc.
>d) RPG/MMORPG - I label these as more "interactive" because while they lack
>any plot (in part due to their level of interactivity) they don't constrain
>the choices of the user beyond the game physics.
I have to disagree with this scale a bit. First of all, your distinction
between text adventures and RPGs on the basis of being able to interact with
things (ie. jumping over the fence) is more a shortcoming of IF authors and
authoring systems rather than the medium itself. In the past, few IF authoring
system have provided much in the way of generic behaviour, and authors were
often too busy with more important things to program in random action that have
no relation to finishing the game. However, the next generation of IF
authoring system, such as TADS3, has begun to solve this. T3 has a much more
sophisticated and simulation-oriented world model than any IF system I've seen
before (indeed, more than any game maker of any kind), and has a lot more
built-in behaviour for trivial activities that aren't part of solving the game.
Second, saying that RPGs don't have storylines in simply ridiculous. Now, I
know most of TSR's games (which seem to make up quite a large portion of the PC
RPG market) are pretty thin in the plot department, but TSR is stuck in the
days of pen-and-paper games played by socially dysfunctional college students
in their parents' basement. I dare you play almost anything by Square-Enix and
say it doesn't have a story as rich (or more often, richer) than any Hollywood
movie (Chocobo's Mysterious Dungeon notwithstanding). Once again, if an RPG
lacks storyline, it's the fault of the designer, not the medium.
>- Unicode, and string table built into the system so that localization is
>easier
TADS 3 already does that. Though the string table is moreso in the script than
the system, but then, in TADS 3 *everything* is in the script. The engine
isn't much more than an interpreter.
>- Don't assume purely textual input. It will need to handle text of course,
>but also menus, click on objects, and speech recognition.
Don't forget, that underneath all the menus, icons, buttons, and dialog boxes,
MS Windows, even to this day, is still a text parser at heart. Same goes for
MacOS and Linux. So just distill all input down its command-line equivalents,
like all the major graphical operating systems do.
>- Avoid 100% open ended commands. Instead, I'd query an object to enumerate
>the commands it supports. I would display the top-4 commands (based on
>context) in a short-list menu.
TADS 3 can do that, too, and TADS 2 had some ability to do it, limited by the
hard-coded parser. That's what the "verify" methods are for -- asking an
object what verbs will work on it (and in T3, how likely they are to accomplish
something).
>3b) TADs/INFORM don't seem to handle conversations with NPCs very well.
Again, check TADS 3. Massive improvements in the conversation engine.
>a menued
>conversation (as used in most RPGs/MMORPGs). I have also been pondering
>search engine anlgorithms for accessing common knowledge, but this may not
>work well.
> (Ex: Questions like "Who is the king?" should be known by every
>NPC, although they will all give subtly different responses.) Does anyone
>have any suggestions about this?
Check the TADS websites technical articles section, an article about
programming dynamic characters in TADS 3. The second page of that article is a
general discussion of various conversation systems, and is not TADS-specific.
You'd probably also find some of the T3-specific stuff on the third page
interesting, as TADS 3 does have something related to that "common knowledge"
thing you're talking about.
>>
One aspect of TADs/INFORM IF that dsiturbs me is the institutionalization of
space (rooms), and objects.
>>
Most IF makers do this because that's how IF has traditionally worked.
However, you should remember that all of that is part of the standard library,
which can be easily replaced with a different system if you have the skill to
write one.
When you say you want something in-between text adventures and choose-your-own
adventure books, I think what you really want is simply IF. Most of the
problems you've complained about either stem from a lack of experience with the
medium or are solved by newer authoring systems.
Mike Rozak
your
> modeler provides. For example, I was able to create a great looking
natural
> tunnel in just a few minutes using POV-Ray, by stringing blob components
along
> a spline path with a bit of random jitter.
Half of my brain agrees, but the other half says "Yes, but it would have
only taken 20 seconds to type a description for the cavern while it probably
took a couple minutes to build the cavern and texture it. Not to mention
stalagtites/mites, lighting, etc." I suspect the extra work is worth it, but
I like to hear other opinions than my own, (such as your agreement in the
possibility of modelling) to know if I'm delusional or not.
> I dare you play almost anything by Square-Enix and
> say it doesn't have a story as rich (or more often, richer) than any
Hollywood
> movie (Chocobo's Mysterious Dungeon notwithstanding). Once again, if an
RPG
> lacks storyline, it's the fault of the designer, not the medium.
I haven't come across anything by Square Enix. I'll have to look for them -
although I think they're mostly console, which I don't have. My experience
with RPGs has (sadly) been a minimal story and mostly hack-and-slash
(Neverwinter nights - which you mentioned, dungeon siege - even less story
than Neverwinter nights, and I never even bothered with Diablo). Morrowind
had more of a story, but I got annoyed with the conversation system (the
menus didn't fit in with the game mechanics, and recorded speech or even TTS
would have been better than 8-point text).
> Check the TADS websites technical articles section, an article about
> programming dynamic characters in TADS 3. The second page of that article
is a
> general discussion of various conversation systems, and is not
TADS-specific.
Just read this. Tads3 looks to be more powerful than the other systems I've
read about
Something still seems wrong with using OO programming for conversations. I
can't quite put my finger on why I'm concerned, but I think it's because
conversations tend to be more "fuzzy" in their logic/rules/behaviour.
Somehow I don't think OO is the best to deal with this. Not that OO couldn't
handle anything, it just might be easier using a different method. I don't
know what that method is though, and haven't seen anything better. I do know
how speech recognition works and have seen some experimental NLP systems
(extracting sentence meaning, as opposed to just parsing); database,
training sets, and probabilities are used extensively, something that OO can
do but which is not built into the structure of the language.
For example: I can image a system where a user plays the IF on-line. When
the user talks to a NPC, an administrator (the author), who is also online,
would have the possibility to usurp the NPCs conversation algorithms. While
the NPC is usurped by the author, the system logs the conversation. At some
later point, it chugs away and magically improves its conversation
algorithms based upon the transcription of the conversation between the user
and the author (who is playing the NPC). How this could learning could be
done, I have no idea. Speech recognition, text-to-speech, and even some NLP
systems magically learn information from training sets.
A simpler system would be able to log the conversations that beta testers
have with NPCs and somehow create a large database that the author can
choose to populate with resoponses. I wouldn't be surprised if tads could do
this.
> However, you should remember that all of that is part of the standard
library,
> which can be easily replaced with a different system if you have the skill
to
> write one.
Agree, and I've been thinking about doing this. The OO architecture is very
flexible. The "if you have the skill to write one" bothers me, not because I
lack the skill, but because other potential authors might. My mind starts
whirring and figuring out the most common modalities (standard IF vs.
choose-your-own adventure vs. ???) that potential authors might want, so
that I can build a system that can more easily accomodate them.
--
Mike Rozak
www.mXac.com.au
Adrien Beau
>
> I don't expect amateurs to create professional-quality models.
> But they may not need to, since even a mediocre model may be
> good enough to improve the IF experience. I'm not sure
> of this though.
I don't think it would improve my experience, I've seen too many
quality models to notice flaws right away and to be distracted by
them. With the overall quality of 3D images steadily improving, I
think this is the reaction most people would have.
I admit however that the quality of illustrations doesn't matter
as much as their consistency (in design, in style, in color,
etc.) and their number. It is likely that after a while I would
get used to them and enjoy them. Rhem is a typical example: it is
poorly modelled and uses low-resolution images (according to my
standards) but you get used to it and then enjoy it.
> Traditional adventure games are a series of interactive puzzles
> with a plot thrown in to glue them together. I want a plot/story
> with occasional interaction (not necessarily puzzles) that
> enhances the plot/story. This is a vague and fuzzy idea in my
> mind, so I can't explain too clearly.
Here are two excellent games that correspond to your description:
Photopia and Photograph. They take less than a couple of hours to
play, you should try them.
> Here's an off the cuff example based on a well known book, The
> Hobbit. (...) The whole experience makes the user feel
> like Bilbo's comrade, empathize with Bilbo's plight, and be
> drawn into the story.
It wouldn't work for me. Maybe it would work on kids.
--
spam....@free.fr
You have my name and my hostname: you can mail me.
(Put a period between my first and last names).
Adrien Beau
>
> I haven't come across anything by Square Enix. I'll have to look
> for them - although I think they're mostly console, which I
You can play old Square and Enix games on console emulators. They
indeed tend to have better plots than the typical PC RPG, even
though they all tend to use the same recipes over and over. I
have just played (the easy, american version of) Final Fantasy
IV, and was really pleased with the plot.
> My experience with RPGs has (sadly) been a minimal story and
> mostly hack-and-slash (Neverwinter nights - which you mentioned,
> dungeon siege - even less story than Neverwinter nights, and I
> never even bothered with Diablo). Morrowind had more of a story,
> but I got annoyed with the conversation system (the menus didn't
> fit in with the game mechanics, and recorded speech or even TTS
> would have been better than 8-point text).
Dungeon Siege was directed by someone that values gameplay far
more than storyline. His previous game, Total Annihilation, also
had a poor storyline, but excellent gameplay: it is by far my
favorite real-time strategy game, a genre that I mostly dislike.
Diablo is based on Nethack: the purpose is to roam a dungeon and
gain experience, level, and cool items, and to kill all the
monsters there are. The paper-thin storyline is just icing on the
cake.
Have you tried Arena III: Morrowind? I have watched a friend
playing it, and am very impressed with that game. I think it is
the closest thing to a "real" role-playing game: there are plots,
subplots, many side-quests. As you gain power, you gain real
influence on the people that surround you. You can betray people
and organizations (there are many of them), you can have your own
castle built, you do not have to go for the main quest. The
universe is vast, complex, and its physics are rendered pretty
well. ... A real time sucker it is. ;-)
Quintin Stone
> I haven't come across anything by Square Enix. I'll have to look for
> them - although I think they're mostly console, which I don't have. My
> experience with RPGs has (sadly) been a minimal story and mostly
> hack-and-slash (Neverwinter nights - which you mentioned, dungeon siege
> - even less story than Neverwinter nights, and I never even bothered
> with Diablo). Morrowind had more of a story, but I got annoyed with the
> conversation system (the menus didn't fit in with the game mechanics,
> and recorded speech or even TTS would have been better than 8-point
> text).
As much as I can't stand Square games, they're definitely not lacking for
story. (FF8 and ChronoCross have soured me on Square forever.) The bulk
of CRPGs only have a story as an excuse for linking together fight after
fight. Dungeon Siege is one of them. Neverwinter Nights, Icewind Dale,
and Baldur's Gate are slightly better. Morrowind does have a story,
though one that's easy to ignore. Two of my favorites, Fallout and
Fallout 2, have a story much more central to the game.
/====================================================================\
|| Quintin Stone O- > "You speak of necessary evil? One ||
|| Code Monkey < of those necessities is that if ||
|| Rebel Programmers Society > innocents must suffer, the guilty must ||
|| st...@rps.net < suffer more." -- Mackenzie Calhoun ||
|| http://www.rps.net/QS/ > "Once Burned" by Peter David ||
\====================================================================/
Quintin Stone
> Second, saying that RPGs don't have storylines in simply ridiculous.
> Now, I know most of TSR's games (which seem to make up quite a large
> portion of the PC RPG market) are pretty thin in the plot department,
> but TSR is stuck in the days of pen-and-paper games played by socially
> dysfunctional college students in their parents' basement. I dare you
> play almost anything by Square-Enix and say it doesn't have a story as
> rich (or more often, richer) than any Hollywood movie (Chocobo's
> Mysterious Dungeon notwithstanding). Once again, if an RPG lacks
> storyline, it's the fault of the designer, not the medium.
TSR doesn't really exist anymore. D&D is published by Wizards of the
Coast. Frankly, the amount of plot and story in any table-top RPG depends
entirely on the scenario and the Game Master running it.
Andrew Plotkin
> I dare you play almost anything by Square-Enix and say it
> doesn't have a story as rich (or more often, richer) than any
> Hollywood movie
This is that "damning with faint praise" that I hear so much about?
--Z
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
* Make your vote count. Get your vote counted.
Rikard Peterson
news:U3ixb.29011$aT.1...@news-server.bigpond.net.au ...
Adding graphics does not make anything better. Adding graphics to IF
is not an improvement, and a movie is not an improved book. I hope
that you (and everyone else in this group) agree on this, but I
wanted to have it said anyway. I play graphical adventure games. I
even *create* graphical adventure games! but I don't think one kind
is better than the other. (If I had to pick one, it'd be text only.)
At the same time, I think you're selling the graphics short. A
standard library of graphics sound far from interesting to me and I
don't see the point of the dial-a-person thingys. Why? I'm afraid
you'll end up with something lifeless and pre-fabricated. Graphics at
its best can sort of add another dimension, but I don't see human #42
with a number seven nose adding anything and would much rather see
James described as the nervous young man with "a smile that is not
quite strong enough to reach his eyes".
Of course, these are just my incoherent thoughts, and I do wish you
luck with your project, if you go through with it. It'll be
interesting to see how it turns out.
Rikard
Mike Rozak
> is not an improvement, and a movie is not an improved book. I hope
> that you (and everyone else in this group) agree on this, but I
> wanted to have it said anyway. I play graphical adventure games. I
> even *create* graphical adventure games! but I don't think one kind
> is better than the other. (If I had to pick one, it'd be text only.)
Agree; there are times when I want a book and times when I want a movie. I
suppose I'm pondering the question... if amateurs could make a gaphical
adventure, would they, and would other people play it? It seems like some
people think yes, and others think that the graphics don't add anything, or
because they won't be professional quality and would detract from the
experience.
> At the same time, I think you're selling the graphics short. A
> standard library of graphics sound far from interesting to me and I
> don't see the point of the dial-a-person thingys. Why?
Take a look at http://www.facegen.com/. I'll admit that all the faces look
too perfect, but a large variety is possible. Plus, once the face is
generated you can always hand-tweak it in the normal 3d editing way. You are
correct that the "default/automatic" system always gets used more than
hand-tweaking, and it does result in less variation in the models/scenes.
Some people in the 3d modelling world complain that Poser
(http://www.curiouslabs.com) is so easy to use that many artists just use
Poser models (of people) and never create their own. Those people that use
poser have the counteraguement that it lets them spend time on other aspects
of the artwork instead.
From another post:
> I don't think it would improve my experience, I've seen too many
> quality models to notice flaws right away and to be distracted by
> them. With the overall quality of 3D images steadily improving, I
> think this is the reaction most people would have.
Yes, I'm definitely concerned about this problem. I thought of one very
hackish solution. Use image-post processing to make it look like a
hand-drawn cartoon (called cartoon rendering), or a watercolor or oil
painting (like photoshop). These techniques hide bad modelling/texture and
apply a style to the rendering. They can also be used to augment the
emotional content - such as long strokes to indicate movement, etc. The
artistic look also given the author freedom to make a sparser image (aka:
easier/faster to create), since tradionally painters/illustrators will
paint/draw the main character, and make scribbles to indicate the background
without actually detailing the background. But, I can easily imagine getting
sick of seeing computer-synthesized water colors all the time too. The
painting simulations completely fail when applied to animation because they
change signficantly every frame. (There are ways around this, but they're
not pretty.) See http://www.curiouslabs.com/article/archive/127/?df=1 for
examples.
By the way, thanks for the feedback. It's giving me a lot to think about.
I've been spending lots of time following up people's links, and have come
up with some new ideas/issues.
--
Mike Rozak
www.mXac.com.au
Rikard Peterson
news:5ewxb.30027$aT.2...@news-server.bigpond.net.au:
> I suppose I'm pondering the question... if amateurs could make a
> graphical adventure, would they, and would other people play
> it?
To that question, the answer is without a doubt 'yes': They already do.
I got the impression that you are unaware of those games - take a look
at http://trumgottist.com/crowsnest/ holds the most complete index of
such games and game engines that currently exist.
Of course, what you are doing will be different, and while I'm not
convinced it's a good thing (particulary the graphics library part), it
will be very interesting to see the result.
Rikard
Mike Rozak
--
Mike Rozak
www.mXac.com.au
"Rikard Peterson" <trumg...@bigfoot.com> wrote in message
news:Xns9441C6865FD6Btr...@130.133.1.4...
M.D. Dollahite
>Half of my brain agrees, but the other half says "Yes, but it would have
>only taken 20 seconds to type a description for the cavern while it probably
>took a couple minutes to build the cavern and texture it. Not to mention
>stalagtites/mites, lighting, etc."
I once read that a picture is worth a thousands words, and a word is worth a
thousand pictures. I know from experience that this is absolutely true. There
are some things that you just can't describe properly with words, and other
things that just can't be shown in a picture. To really have a full range of
expression, you need equal helpings of both.
>I haven't come across anything by Square Enix. I'll have to look for them -
>although I think they're mostly console, which I don't have.
Yes, they tried making PC ports of a few of them (FF7 for one), but that didn't
generally go over well... mostly because PC gamers have evolved such radically
different tastes from console gamers -- many of the complaints expressed by
reviewers were more about the console-style UI than the game itself.
As someone else pointed out, the NES and SNES games can be run on emulators,
and a lot of them are better than some of the newer ones anyway.
>I
>can't quite put my finger on why I'm concerned, but I think it's because
>conversations tend to be more "fuzzy" in their logic/rules/behaviour.
Yes, but you have to balance realism against not going insane trying to program
all the variables.
>While
>the NPC is usurped by the author, the system logs the conversation. At some
>later point, it chugs away and magically improves its conversation
>algorithms based upon the transcription of the conversation between the user
>and the author (who is playing the NPC). How this could learning could be
>done, I have no idea. Speech recognition, text-to-speech, and even some NLP
>systems magically learn information from training sets.
I've seen some academic attempts at this kind of system. In practice, they
tend to be destroyed rather quickly and permanently when some jerk gets it in
his head to type in a bunch of nonsense.
-----------
"Quintin Stone st...@rps.net "
"TSR doesn't really exist anymore. D&D is published by Wizards of the
Coast. "
Well, they still put the TSR logo on everything. Obviously I'm not a big fan,
so it makes little difference to me.
"As much as I can't stand Square games, they're definitely not lacking for
story. (FF8 and ChronoCross have soured me on Square forever.) "
...FF8 and CC were pretty sorry, as Square games go. FF9 and 10 were
considerably better, and I hear that FF11 (aka FF Online) is the first MMORPG
with an actual storyline. (Alas, I have not the finances to play MMORPGs.)
"Adrien Beau spam....@free.fr"
"Diablo is based on Nethack: the purpose is to roam a dungeon and
gain experience, level, and cool items, and to kill all the
monsters there are. The paper-thin storyline is just icing on the
cake."
And before Nethack there was Gauntlet. Truly, almost every game in existence
today can trace its roots back to a classic arcade game, and in most cases the
only thing that's really changed is the graphics.
R. N. Dominick
> And before Nethack there was Gauntlet. Truly, almost every game in existence
> today can trace its roots back to a classic arcade game, and in most cases the
> only thing that's really changed is the graphics.
Rogue, the grand-daddy of all roguelikes, dates from 1980. Gauntlet was
released in 1985.
M.D. Dollahite
>released in 1985.
Before my time, I guess; I was born in '80 myself. Just more proof of my
point: almost all games are merely remakes of games that came before.
R W
> story. (FF8 and ChronoCross have soured me on Square forever.) The bulk
> of CRPGs only have a story as an excuse for linking together fight after
> fight. Dungeon Siege is one of them. Neverwinter Nights, Icewind Dale,
> and Baldur's Gate are slightly better. Morrowind does have a story,
> though one that's easy to ignore. Two of my favorites, Fallout and
> Fallout 2, have a story much more central to the game.
The official single-player games that come with Neverwinter Nights -
the original game especially - aren't very good. Neverwinter Nights is
supposed to be more of a roleplaying system than a game in itself.
There are plenty of really good user made games for it, and that is
where the game ireally comes alive, particularly when played
multiplayer with a Dungeon Master.
For a good story driven roleplaying game on the PC, I hear that
Planescape: Torment is supposed to be very good, despite being so
little heard of.
I don't really count games like Dungeon Siege and Diablo as
roleplaying games, they're really fantasy-action games, which use RPG
as a term interchangable with fantasy, which it's not.
Michael Coyne
> For a good story driven roleplaying game on the PC, I hear that
> Planescape: Torment is supposed to be very good, despite being so
> little heard of.
I concur. I'd never heard of it until I saw it at the local electronics
superstore for $20 with a $20 rebate, thus free.
So I bought it for the staggering sum of $2.79 (taxes, you know), and
was amazed. Great story.
Michael
Michael Bechard
Yeah, really great game. Good, good story-driven gameplay.