 |
|
|
|
|
|
||
|
|
||
|
|
||
|
|
|
|
|
|
||
|
|
||
|
|
||
|
|
|
|
|
|||||||||||||||||
Add Forum to Favorites | Send Topic To a Friend | View Forum FAQ | Track this topic |
Last Thread Next Thread |
| Using Low Level Stories |
|
 |
MatrixCubed Member since: 8/4/2000 From: Hull, Canada |
|||
 Posted - 10/16/2001 8:08:53 PM |
||||
| Well, I'm not certain that this article is very coordinated. At the beginning, the author talks about Tetris' high level story, then proceeds to explain the rest of the article with views on RPGs, RPG-ish actions (burglary), and speaks of NPCs. Is this article about replay-value? If so, then the target game-genre (yes, I believe there are defined genres) must be defined for the article. I don't believe Tetris has a high-level or a low-level story; it's a puzzle game. If Monopoly, Game of Life, Poker, or Tetris had high- or low-level stories, then they wouldn't be strategy or luck games, they would be something different (and I'm not really certain how to define them if they did have these, because I don't think high- or low-level story games' rules apply to these games). Is this article about interactivity? If yes, then the article is probably targetted at RPGs (and possibly at FPSs that have RPG elements, like Deus Ex). Now we're getting somewhere. Such a game should always have more than one way to play through it (Ultima V is a great example, where you simply begin in the middle of the game map, and choose your own fate). And such a game should never impose extreme "rules" for reaching the end. This makes the game realistic, and much more interactive than being able to burglarize someone in the game. There must be a definition made, that explains the difference between "permitting a particular act (burglary), so that it seems the player has freedom, though the act is just another plot branch", and "creating the game's system in such a way that particular acts (burglary) are possible, and characters in the game react accordingly". Going back to Deus Ex, there is a point in the game where you can optionally kill Joseph Manderley, and other characters in the game respond to this. That's great, but does it really make a difference for the game's outcome? No. The reason? The act would probably have enforced a completely different outcome in the end. I suppose the only way that a game can be made replayable, is to (1) make "level"/"chapter" transitions seamless (a la Ultima IX), (2) offer many different ways to complete each individual level, in almost any order (with regard to whatever rules must be applied so that the game may progress in a logical fashion), and (3) offer many smaller sub-plots throughout the game that may or may not have a direct effect on the high-level story's outcome. MatrixCubed http://MatrixCubed.org  |
||||
|
||||
 |
Oluseyi Staff Member since: 5/14/2001 From: New York, NY, United States |
|||
 Posted - 10/16/2001 8:50:26 PM |
||||
| Project Ego from BlueBox or so (with some input from Peter Molyneux) purports to do all this article describes: freedom of choice, "Chinese Whispers", etc. I think the problem lies in the inherent linearity of the concept of a "story." To overcome this, the game should patch together the player's actions and their consequences to generate the resulting story. Provide an initial premise, an incentive to go out and do something, but don't require that anything actually be done. This is totally at odds with the prevalent notions of storytelling, but interactive media is not about storytelling; it's about story creation - the player creates a unique story every time she plays. This is why the most replayable games are those that have little notion of predefined story or linearity (Super Mario Bros. 3, which allows you to play the side games; sports games, which have no story other than what you make of them, etc) |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
 Posted - 10/17/2001 9:18:07 AM |
||||
| Hey, folks! Thanks for the feedback. Sometimes it's difficult for me to evaluate my writing without a certain bias. So I appreciate your comments. I really wasn't very clear about which game genres I was discussing. I was primarily concerned with RPGs, Horror, and similar genres which are story driven. To me, a low level story is something which really has no bearing on the outcome of the main story. A low level story includes actions which the player may take, and enjoy, but which will never have a profound influence on the game's high level story. Perhaps a better way to describe the low level story is to call it "what course the player took when completing the game". (Although that's hopelessly clumsy.) The fact that the low level story is independent of the high level story is very important. This allows the designer to increase replay value, while keeping a managable main storyline. Perhaps I should give an example. One of my favorite games for the Genesis was "Herzog Zwei". This was basically a RTS game. There was some sort of story to it. But what really made the game great was the freedom, the many, many ways in which the game could be played. Each time I played, I would do something different. One time I might launch a full scale attack with tanks. Another time I might gain territory slowly and steadily. The low level story in Herzog Zwei is really just the history of how the player won the game. Did he use infantry? Did he control the seas? How did he react when his bases were attacked? That is the low level story. It does not influence the main storyline. It just keeps track of what the player did. Maybe calling this a story isn't quite right. But I believe that paying attention to how the player conducted the game offers designers a chance to make a pseudostory out of those details. And I really think that this has good potential as a design technique. |
||||
|
||||
|
_002 Member since: 10/17/2001 |
|||
|
Posted - 10/17/2001 10:09:14 PM |
||||
| I like the article's concept, but it's a bit lacking in the implementation department. The idea about dynamic gossip is an interesting one. I've found myself thinking along those lines quite a few times. So far I have yet to figure out a practical way of implementing it, but for starters, gossip is based on who knows what. Logically this would be best implemented in something like an RPG with a world full of independent characters. Each of these characters could then have a knowledge base consisting of some kind of a list of knowledge bits that could be linked together. The linking bit is where I always get screwed up because in order to get the kind of flexibility you'd need to really get a realistic flow of conversation, you'd need a potentially limitless number of links on each memory node - very taxing on one's RAM, especially when you're talking about lots of bits of knowledge across at least several hundred entities. But hey, we're making games here, not doctoral theses. We can probably get away with using a limited number of links on each node. This might be a little easier to maintain, but could cause the characters to appear narrow-minded. Still that may be better than the walking, talking signposts that you see a lot of now. Obviously the whole concept of conversation comes naturally to a MMRPG where other players would be doing the dicussion - not a whole lot of memory-guzzling AI to be worried about. This is a far from complete technical treatment of the idea, does anybody have any thoughts? Tom |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/18/2001 2:30:44 PM |
||||
| Hey, Tom (_002). You're definitely right. The article included absolutely no implementation suggestions. Unfortunately, this is because I am only beginning to learn programming. So I really couldn't offer those details. But I think that an article which offers suggestions about how to implement this sort of system would be great. I don't have the neccessary background to do that. But if you find a practical way to implement this technique, perhaps you would consider writing an article about it. Or if you don't have the time to do that, I think it would certainly make a good thread in the programming forums. And thanks again for the feedback, everyone! Jonathon quote: |
||||
|
||||
|
Nimdok Member since: 8/28/2001 From: USA |
|||
 Posted - 10/20/2001 3:34:51 AM |
||||
| I enjoyed the article. Little details and side-stories can really bring a player into the game. They were all over the place in the classic adventure games. Stealing is fun, but experiencing the consequences is even more exciting. :D |
||||
|
||||
|
greggman Member since: 10/20/2001 From: Japan |
|||
|
Posted - 10/20/2001 9:56:21 AM |
||||
| Why are we concerned with replayability? Is it really a desirable thing? In movies it is because each time watched is another $9+ to view but such is not true with games (I'm not suggesting a purely business point of view) As a gamer, how many *story* based games have you wanted to replay? How many times after you finished reading a book did you immediately want to start reading it again? I think the suggestions in the article might be useful for making a ONE play game more involving for the player but I think there is alot of wasted thought on replayability. I don't believe it's something most players are looking for from a story based game. Spend your time making the first play fun. Make it take around 20 hours to finish without being tedious or frustrating and you've done something only about 2 games a year pull off. Time spent making the second play fun is probably arguably some time off not making the first play funner. Just my 2 cents. Gregg Tavares http://www.greggman.com |
||||
|
||||
|
Drizzt DoUrden Member since: 8/19/2001 From: Parsippany, NJ, United States |
|||
|
Posted - 10/20/2001 10:07:15 AM |
||||
| I understand the point of the article, and I really didn't find many problems with it. Look at Final Fantasy 7; I literately have 7 memory cards for PSX filled with FF7 blocks. Do you know why? Because the game is replayable. There is always something new, every time you play it. ( Not anynmore, for me.. I guess... but anyone who plays it can play at least 5 times and find 5 new things each time ) They are called "sub quests." John reffered ( excuse my programmer's spelling... if anything is spelt wrong ) to them as "low level stories," which might have thrown a few people off track, considering the fact that alot of people reading this had little expierience with writing stories ( not you guys, maybe, but I am sure there were a bunch of people who just read it to read it. ). A game should be replayable.. because the amount of money they cost nowadays is rediculous.... $50 for one game? One game I will play one time? I could go out and buy some new computer parts for development of my own games if I didn't buy anyone else's games. I must have around $600.00 worth of computer games and PSX games in my house, the only three I replay are Final Fantasy 7, Final Fantasy 9, and Metal Gear Solid ( though, there are no subquests really... but I like the "no radar" challenge, and I can never beat it, so I have to replay it.). I thought the article was good, and I will probably reffer to it some time in the future, right now... I am trying to develop a Baldur's Gate like story.... can someone write an article on that?  Good Job, John. By the way... I like your profile sig, did Bush really say that? "He who fights with monsters should look to it that he himself does not become a monster... when you gaze long into the abyss, the abyss also gazes into you..."~Friedrich Nietzsche Edited by - Drizzt DoUrden on October 20, 2001 10:08:05 AM |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/20/2001 11:18:36 AM |
||||
quote: This is a very important point. It is a wonderful experience for the player to be able to create his own story by influencing the game world. As you say, this is most common is games which have "little notion of predefined story or linearity". I would like to see more games which offer this sort of freedom and player empowerment, as well as a well written, top quality story like those found in the best RPGs. Jonathon quote: |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/20/2001 11:44:54 AM |
||||
quote: I hear what you're saying. But I have to say that I think it's sad that gamers aren't demanding more. I have played some wonderful RPGs. These games have featured good stories and good characters. But many of these games I haven't had any real desire to play through a second time. I think that's what you were getting at. But I also have to say that I don't like to spend $50 on a game I'll only play for a few days. quote: I agree with you. It's sad that so many games are not as good as they should be. (Of course, there are many reasons for that.) quote: Actually, this is something I was trying to address in the article. When a designer uses branching storylines, or similar techniques, that is certainly the case. It takes resources for each branch. But the idea of what I called the low level story is that it's actually separate from the main storyline. It's more of a game system. A typical RPG features a number of systems. There is usually a battle system, a character experience system, a travel system, a random encounter system. Think of the low level story as something like an "interesting little bits of story" system. Just like the battle system, or the spell system, it only needs to be created once. And once it is created, it applies to every time through the game. So it adds to the second play while adding to the first play as well. At least, that's the idea.... Jonathon quote: |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/20/2001 11:58:17 AM |
||||
quote: LOL!  quote: And what does that say? You've made a considerable financial investment in personal entertainment. Yet for all that cash, you've only received real value from three titles. Something's got to be done about this! quote: That and a whole lot more. Don't even get me started!  By the way, your profile quote is just scary. (Especially today.)Jonathon quote: |
||||
|
||||
|
_002 Member since: 10/17/2001 |
|||
|
Posted - 10/20/2001 4:12:56 PM |
||||
quote: Correct me if I'm wrong, John, but I think this guy didn't really get the point of your article. A sub-quest has to be manually entered into the game by a designer. The article was about game elements that allow a player to go on a side-quest without somebody writing the side-quest. The practical upshot of this is that with a bit of clever programming and world layout, a potentially infinite number of optional side-stories can be created. quote: I believe you're talking about the average price of a commercially produced game by $50, right? First off, the developer gets approximately $10-$12 of that, assuming the publisher is paying them well. The rest of it goes towards various other things that actually get the thing on the damn shelf so you are more likely to buy it. If you look around the net at games that are independently published, you'll find that a good deal of them cost a bit less than $50. Second, a $50 to write a commercial quality game? Perhaps maybe you are a virtuoso game developer, in which case disregard the following, but good code takes time to think out, good art must be inspired and drawn with the necessary skills, good music and sound effects require a vivid imagination and a trained ear, and of course there has to be a concept for the whole game that fits together on paper before you even begin creating content, otherwise you get a heaping pile of crap that took dozens if not hundreds if not thousands of hours to create. Many of these games cost hundreds of thousands, sometimes millions of dollars to create. $50 for a piece of the fruit of a team's blood, sweat, and tears is not asking a whole lot. Real people, lots of them, make those games you're playing. They have to eat too. Third, if you're spending $50 on a game that you're only going to play once, why the heck are you wasting your money?? Games get reviewed by people who play games and review them for a living. Read these reviews before you go out and spend your precious money on a game to decrease that risk of playing it once and never playing it again. When I spend $50 on a game, I expect the thing to be keeping me busy for at least a couple of weeks, otherwise I get very skeptical whenever the game's developer comes out with another game and says "oh you'll like this one, we promise." quote: $600, that's it? and that no radar thing you're talking about - that would be a low-level story element. Edited by - _002 on October 20, 2001 4:19:09 PM |
||||
|
||||
|
MatrixCubed Member since: 8/4/2000 From: Hull, Canada |
|||
|
Posted - 10/21/2001 2:16:21 AM |
||||
| Some thoughts: On reviews I would not trust sites that offer game reviews, for the simple fact that I would attribute to anti-"mob rules" attitude: the voting majority is *not* infallible. Trying games because they interest you is far less cow-herd-minded than trusting reviews. Compare your favorite games that have ratings less than 80%, to those others that have 90%+. On small story engines I still think it's a great idea, probably needs some more carving and sanding though. What would be an ideal situation is a game world large enough, detailed enough, that the game world itself offers a whole lot for the player, and "real quests" may be added as time goes on, by the game designers. So in the meantime that players have no "real quests" to follow, they might explore the game world, find secrets or follow small-quests, get employment with a farmer for a month or two, and so forth. Sort of like "Wing Commander: Secret Ops" but with a game world to play through "on the side". On replayability in general You also have on your hands, the element that there are (at least) two types of players: those that like to explore, and those that like to finish. The finishers are likely the ones that "Small Story Engines" are created for (because they usually are the players with the loudest howls for entertainment ); your explorers however (and these are the only players I would attempt to make a "good" RPG for) might catch most the Small Stories the first time through the game (assuming things like time limits or randomness are not in effect).MatrixCubed http://MatrixCubed.org |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/22/2001 9:06:14 AM |
||||
quote: Yes, that's exactly what I meant. quote: I agree with you on that too. Of course, designers at big companies aren't able to reduce the shelf cost of these games. But I do believe that $50 seems pretty expensive to many players. Many (like me) have limited resources, and really feel that $50! As long as the games are going to be expensive, I feel that designers really do have an obligation to provide as much value as possible in each game. It's true that I have purchased games which I only played one or two times, but were worth the price. On the other hand, I would like to be able to replay those games more often. And by using something like the system I was trying to illustrate, I hope that designers will be able to make more "one play" games "many play games". Jonathon quote: |
||||
|
||||
|
DarkDream Member since: 10/22/2001 From: Toronto, Canada |
|||
|
Posted - 10/22/2001 1:01:16 PM |
||||
| I found the article had some goodpoints, mainly that lowlevel stories can be useful and can add to the replayability, but it completly missed the mark when it was stated "the story can be developed with less work and fewer resources. Unless maybe your talking about a text only game? Unfortunately, any plan implemented will come at a cost. It will take lots of resources and the all important resource of TIME! How is "... the low level story is bound to be different each and every time without and effort at all on the part of the designer"???? Take your example of the socks. Socks do not magically appear in games, nor the room and drawers they come in. They are created in a 3d program and they must be textured, then incorporated into the game. What about Gildors house? Please explain to me how I'm going to break in to the house unless its already part of the game and that you might want to go there and break in? If your talking about stealing, where do you draw the line at what can and can't be stolen. I want to replace Gildors underware with pink silk panties as a joke. Can I do that? I want to frighten Gildor and put a horses head in his bed. How can I do that? Is there a butcher in town where I can find one. Or can I just kill a horse, remove the head and take it to Gildors house. If you promote your game as having low level stories how do you incorporate those ideas or do you make a list for the player to choose from? The cost is huge in manpower alone to incorporate everthing into a game to support every low level story in a game. Just think of more buildings, rooms, objects, sound files, textures, possible dialogue tracks etc....for every low level story. In your second example, the player and companions are transported 150 years forward in time to their homeland. Is that really a low level story or is that part of the main story. Do they have a choice to be transported, and where to be transported or can they only go to their home land? Do you see what I mean? I may be totally off the mark and missed something, and I appologize if I did. Please enlighten me |
||||
|
||||
|
jonbell Member since: 10/23/2001 |
|||
|
Posted - 10/23/2001 9:04:24 AM |
||||
| The ideas discussed in this article are totally unfeasible. So the player wants to go stealing items from a house. Fine that means that the programmer must write reams of code to support this minor act. Following this theme the player may next want to set fire to the house. Now the programmer must write all the code to support this. What next? Does the player want to move all the furniture around in the house? Well if he does there must be yet more code to support this. The writer of this article simply hasn't thought it though. He claims that supporting all the whims and acts that a player may or may not want to commit requires little effort. The fact is that the more a player can interact with an environment, the more code you must right to support each act. |
||||
|
||||
|
Nazrix Member since: 3/13/2000 From: Orlando, <s>USA</s>Tiberia |
|||
|
Posted - 10/23/2001 10:11:10 PM |
||||
quote: Actually, I know exactly what you mean and that is a great idea and one I and Wavinator are currently persuing.... Check out this thread A CRPG in development... Need help? Well, go FAQ yourself. |
||||
|
||||
|
Nazrix Member since: 3/13/2000 From: Orlando, <s>USA</s>Tiberia |
|||
|
Posted - 10/23/2001 10:13:53 PM |
||||
quote: I disagree. The idea is that you give the player a finite low-level set of actions and the player can mix & match using his/her creativity to achieve a goal. If what I'm thinking is what the article is trying to present, the stealing concept wouldn't be a special little coded situation by the programmer. It would be more atomic. A good analogy is how every molecule is made up of a limited set of atoms. So, stealing is really nothing more than "taking". Taking or picking things up is something that almost all games let you do. Stealing is just taking something that is not yours. So the extra thing you must add as a programmer in this case is the concept of an owner. The point is that things are on a low level and can be combined and used in different contexts to achieve things the player wants to do. Check out the above link which leads here: http://www.gamedeveloper.net/ubullboard/Forum12/HTML/000010.html It discusses this concept quite well IMO. A CRPG in development... Need help? Well, go FAQ yourself. Edited by - Nazrix on October 23, 2001 10:49:40 PM Edited by - Nazrix on October 23, 2001 10:50:44 PM |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/24/2001 9:23:41 AM |
||||
| Dark Dream and jonbell bring up a very important concern. How can a system like this possibly be implemented? It would seem that the game's designers, artists and programmers would need to create a complete, minute simulation of a real world. That is a very real concern. I should have pointed out that the idea is not to allow the player complete freedom. A system like this should not attempt to offer the same, infinite freedom one experiences in the real world. It is meant to be done on a much smaller scale. Consider the case of Gildor's lucky socks. This should be a fairly isolated event. The character Gildor is a major rival of the PC. So it would be reasonable for the designer to pay more attention to him. It could be assumed that at some point in the game, the PC will have access to Gildor's home. So attention could be paid to the objects present only in Gildor's room. In other words, the game only keeps track of whether certain items are stolen. Perhaps a few items in Gildor's room, and a few important items around the town might be kept track of. If the player steals something else, he won't hear about it at all. This is what I meant when I said that the game should not recite everything the player does. Only a few details should be kept track of. It should be a surprise to the player, but a pleasant one. Dark Dream asks about the PCs being transported into the future. This is a special case in which the low level story complements the high level story. And that means that the designer would need to pay special attention to this. The main story involves the time travel. But to enhance this experience, the game keeps track of a few of the things the PCs have done in the past. For instance, the game might keep track of how many times the party had fled from enemies while in the cave. Or it might keep track of which opponents the PCs had a difficult time with. Which characters were hurt the most? I think this would be easy enough to code. A few variables such as timesFled, mostDifficultEnemy, and charMostHurt could be updated after each battle. Then, after the time travel, those variables could be retrieved at a certain point. A good time would be when the PCs return to the town for the first time after the time travel. The high level story would force them into a particular meeting. Whatever the method, the PCs would eventually be in a particular place at a particular time. It is then that those variables would be recalled. Perhaps the PCs are asked to meet with a town official at the local inn. While there, they hear an old storyteller, who recounts the tale of the heroes. If Balrog was the character most hurt, he would be the character protrayed as the most heroic and invincible. Balrog the Fearless, who never fled in batttle (the opposite of the timesFled variable) is said to have singlehandedly waded through hordes of Ogres (mostDifficultEnemy). On the other hand, if the players were particular skillful, they might hear a very different tale. In this case, Balrog might be portrayed as a weak but crafty character with luck on his side. Although he was not overly brave (as the timesFled variable ==0), he somehow managed to bungle his way through the cave. But I have to confess that I am only beginning to learn Java programming. I began my first semester as a Computer Science major a few weeks ago. So there are doubtless certain implementation details which are beyond my expertise. It is certainly true that any system is going to require work on the part of designers, artists and programmers. But I think that it would be possible to create a simple version of a low level story system without too much effort. The important part is to decide which small, easily implemented details will add the most to the game. Again, I'm not a programmer yet, so please be gentle with me! Jonathon quote: Edited by - Jonathon on October 24, 2001 9:32:31 AM |
||||
|
||||
|
DarkDream Member since: 10/22/2001 From: Toronto, Canada |
|||
|
Posted - 10/24/2001 10:50:14 AM |
||||
| Now your getting somewhere Jonathon Ideas are easy, but writing articles, developing games etc is not easy, its hard work. With more thought, now your getting to the meat of your ideas and how to practically implement those ideas. The way the article was written would work well with any D+D game where free thought and ideas can be implemented on a whim but not easy for a computer game. I see where your going now. By keeping track of the players actions and implementing them into the game would add greatly to any game. Its the little touches in a game like this would really give a game some added life. quote: You hit the nail on the head here, in deciding what details to add. Getting it to work well in the game is another story in itself, but could work very well. Good luck on your programming. This discussion should continue with more of these ideas to add to a game. For example, in a Rpg what other things could be added or kept track of besides stealing or running away in battles that would add to the game. |
||||
|
||||
|
jonbell Member since: 10/23/2001 |
|||
|
Posted - 10/24/2001 4:32:29 PM |
||||
| Johnathon i feel that you have missed the point slightly. If as you suggest you start hard coding certain items and events such as steeling socks then the entire system you are discussing goes out the window. Your system sugests to me that as the player progresses, your program builds up a new story for each player. Hard coding details as you suggest does not follow your low level story system, it is merely another sub quest. It is a very good idea but not a new one. Trust me when i say that although it is easy to throw around ideas, implementing them effectlivly is very difficult and there is certainly no short cuts or easy ways. Using your example your 'mostdifficultenemy' variable would be pretty hard to determine in code and that would just be the start. Whats more the player probably wouldn't notice if what the old man was saying was reffering to your actions or was simply hard coded text. Look at Shenmue on the Dreamcast. It is probably the most accurate representation of the real world available at present and it is still pretty static and un-lifelike. There is still only a limited amount of interaction available and this is Sega's best attempt. I cannot emphasise enough that there are no short cuts to providing an interactive and emersive game world. Every single action that the player can perform must be coded. In my opinion it is impossible to provide the player with the amount of interaction that they want. For example in Shenmue when your mother tells you to come in for 11:00PM, i don't want to, i want to come back at 4:00am sneek into her room, slit her throat then set fire to the house. Why? Well why not When it comes to interaction the player always wants more than what you give them. |
||||
|
||||
|
Jonathon Member since: 2/15/2001 From: New York, USA |
|||
|
Posted - 10/25/2001 2:23:45 PM |
||||
quote: The example involving time travel and the cave would obviously require hardcoding. But the main idea behind the low level story is that it should require as little hardcoding as possible. In other words, the low level story system should be hardcoded, but the low level story itself should be created by the player. I think Dark Dream's last post gives a good description of this. quote: Yes. I was disappointed with Shenmue myself. I think this is because the expectations were unrealistic. At this time a computer cannot possibly offer a convincing simulation of reality, complete with all the freedom a player could possibly wish for. The idea behind a low level story system is not to attempt to offer complete freedom to the player. It is only meant to enhance the game by keeping track of a few, carefully selected details. quote: I believe it! Jonathon quote: |
||||
|
||||
|
_002 Member since: 10/17/2001 |
|||
|
Posted - 10/26/2001 6:01:59 PM |
||||
| While the original article does miss a great deal of implementation details, that does not mean it is not feasible. It seems these low-level story elements that are being talked about are really finite state machines interacting with each other. Interactions between a finite set of finite state machines will yield a finite number of possible events. The number of interactions will be a highly exponential figure due to the fact that you're talking about an n-dimensional hypercube of state values (one state value for each FSM). Obviously, hard-coding every single interaction between every single state machine is not a practical endeavour. However, finite state machines are made up of component states. The key to making this system work is having a set of common states that are hard-coded and then shared across multiple finite state machines. Left to itself along predetermined paths, the FSM set will follow a fixed path. Introducing a chaotic factor (i.e. a human player) will cause deviations from this path. What those deviations are depends entirely on the actions of the player. Consider a cup filled with water. It will sit there, ebbing and flowing by itself without any external forces acting upon it. Stick your finger in it, and the water will ripple and the water level in the cup will rise. Do the water interact with your finger? From a superficial point of view, yes. There was an interaction between the finger and the water. What *really* happened, at the most abstract level, was an interaction between a fluid an a solid. The same thing would have happened if you had stuck a pencil in a bowl of punch. The ripple and the level rise would occur just the same. The weird looks and isolated snickers you would receive at a party for doing that occurs at a different level of abstraction - social norms. Yes, there are a lot of things to account for, but it does not all have to be grunt work. The many complexities of this problem can be broken down into different levels of interaction. When something fitting Jonathan's description is completed, it will undoubtably have to be very big to be able to account for the numerous kinds of interactions that would be possible in a real-world environment - and that's going to be a very very big hypercube. A lot of the recent discussion has been about an RPG-style environment which revolves around interaction with artificial characters. Logically, this means a great deal of code focus on conversational artificial intelligence. Given that this site is mainly for independent developers, I doubt that anyone here has the resources to implement such a grand scheme in which stealing someone's socks or replacing their underwear with panties causes that person to lament to their friends the next day about the anomoly in their wardrobe. I'd estimate that sort of job would require at least several dedicated programmers well versed in psychology and sociology. I wish the best of luck to anyone who undertakes the task. Tom Edited by - _002 on October 27, 2001 9:07:48 PM |
||||
|
||||
| All times are ET (US) |
Last Thread Next Thread |
|
