Archived

This topic is now archived and is closed to further replies.

Hikeeba

Hero Morphing and Ending Goblin Genocide

Recommended Posts

I was listening the Molyneux lecture on Gamasutra, which is something you really should listen to I think. Anyway, one part was discussing the way your hero would change in Project Ego. Such things as an overdeveloped right arm if you went around carrying a battle ax, or becoming more lean if you snuck around and stabbed people in the back. So, it dawned on me about using a similar system for experience. Instead of gaining points for ruthlessly slaughtering the slaughterable, why not gain experience based on what you do? If you do sneak around a lot, you should become better at it. If you swing swords around, you should become better at it. Carry rocks around and become stonger. Run, jump, etc and become faster and more agile. If you're a magic user, burn down whole forests or torch grasslands while practicing your fireballs. Just an idea. Not very fleshed out, but I wanted to get it down before I forgot. I wanna' ride on the pope mobile. Edited by - Hikeeba on December 19, 2001 1:41:42 AM

Share this post


Link to post
Share on other sites
How about if we are honest and admit the truth: characters _do not_ improve. They can have stats, they can have skills and all that, but whatever happens, in but one week or two (the length span of a game) these will not change at all. What can change is what the world thinks of the player character and how the PC thinks of himself (fame / confidence).

And while at that, how about we be honest and admit another truth: wearing the best armor or wielding the finest weapons will not make that much difference.

And while at that, how about we finally accept that 10 goblins will eat alive the meanest hero any time day or night.

Do these, and then worry about adding _real_ gameplay to the game.

Share this post


Link to post
Share on other sites
Quote Hikeeba:
"Instead of gaining points for ruthlessly slaughtering the slaughterable, why not gain experience based on what you do?"


Dude, that ideas been around for ages. Daggerfall immediately comes to mind, but there are a heck of a lot more aswell. I wouldn''t spend ages refining this system to work with every single stat, instead I''d choose some key points like walking speed or strength or things that can improve in a reasonable amount of time like Diodor said. Can you imagine someone becoming more intelligent in 2 months?

Share this post


Link to post
Share on other sites
quote:
Original post by Diodor
How about if we are honest and admit the truth: characters _do not_ improve. They can have stats, they can have skills and all that, but whatever happens, in but one week or two (the length span of a game) these will not change at all. What can change is what the world thinks of the player character and how the PC thinks of himself (fame / confidence).

that''s not completely true. if you spend all afternoon swordfighting for two weeks, you can bet you''d be better at if afterwards!
quote:
And while at that, how about we be honest and admit another truth: wearing the best armor or wielding the finest weapons will not make that much difference.

you have apparently never wielded the Dark Spork of Chaos!
quote:
And while at that, how about we finally accept that 10 goblins will eat alive the meanest hero any time day or night.

not if he has the Dark Spork of Chaos!

--- krez (krezisback@aol.com)

Share this post


Link to post
Share on other sites
I''m not talking about skill points, practice sessions, and levels or anything like that. Unless I missed something, the closest any game has come to this is skill points. I''m not talking about skill points though.

In the real world, if you run a lot you get faster. If you fence, you get better. You don''t kill things until you gain so many experience points and then go spend the practice sessions you earned at your guild hall.

Ok, you know how you trained your creature in Black & White? He got better at stuff after repetition. I''m talking about using something similar to that for character advancement. You control your character, and he gets better at swinging a sword just by doing it. He gets more accurate by swinging for a target, a tree for example. Get better at throwing fireballs just by throwing the damn things. Get more accurate by throwing them at a target. Get better at backstabbing by backstabbing. Better at sneaking by sneaking.

No levels. No skill points. No practice sessions. Sort of an experience neural net for lack of a better way to explain it. Not a tree, but a system where repetition results in your character being more effective.

I wanna'' ride on the pope mobile.

Share this post


Link to post
Share on other sites
quote:
Original post by Diodor
How about if we are honest and admit the truth: characters _do not_ improve. They can have stats, they can have skills and all that, but whatever happens, in but one week or two (the length span of a game) these will not change at all. What can change is what the world thinks of the player character and how the PC thinks of himself (fame / confidence).

And while at that, how about we be honest and admit another truth: wearing the best armor or wielding the finest weapons will not make that much difference.

And while at that, how about we finally accept that 10 goblins will eat alive the meanest hero any time day or night.

Do these, and then worry about adding _real_ gameplay to the game.


This has got to be the shallowest post I''ve ever seen. I think Hikeeba is latching on to the new wave of games that actually start to take advantage of all this processing power now at our fingertips. In Project Ego, not only does your character "evolve", but the entire world does to. If you leave your house as a kid and come back as an adult, things will have changed drastically. Also like Hikeeba said, Molyneux is probably using a spinoff of his Black and White AI to accomplish this feat, which is entirely possible. I think we''ll be seeing a lot of games that start to use these "evolving" features as time goes on. It presents a way for the player to personalize his character not through any set means (like leveling up with strictly defined abilities) but by the actual actions he performs in the game. So now it''s not just about fighting monsters to get enough skill points to level up, but fighting monsters to improve your swordsmanship, accuracy, strength.... all these general attributes. Levels will be a thing of the past. I think it''s cool.


_________________________________________________________________

Drew Sikora
A.K.A. Gaiiden

ICQ #: 70449988
AOLIM: DarkPylat

Blade Edge Software
Staff Member, GDNet
Public Relations, Game Institute

3-time Contributing author, Game Design Methods , Charles River Media (coming GDC 2002)
Online column - Design Corner at Pixelate

NJ IGDA Chapter - NJ developers unite!! [Chapter Home | Chapter Forum]

Share this post


Link to post
Share on other sites
If I understand what you''re talking about, I''ve seen this system in older games like Jagged Alliance. In those games, you could do things like improve your aim through repeated target practice. (I think Ultima Online was also like this.)

I think that experience based on action has promise, but I''m not sure how you''d properly scale challenges. If I practice again and again, I might be able to easily beat all but the top challenge / foe.

Also, one thing repetition often fails to capture is difficulty of what you''re practicing against. I know from martial arts that if I practice only against beginning white belts, I''m not going to improve very much. So maybe it''d be more realistic to award greater experience the higher the challenge the player tackles.



--------------------
Just waiting for the mothership...

Share this post


Link to post
Share on other sites
I think there is something to Diodors post, while you do improve a little, in real life the change is nothing near what rpgs use nowadays.
Equally with equipment, a well-balanced and sharp sword will make a difference, but it will not be a deciding factor if a good swordsman fights a mediocre one.

Then there are a lot of gameplay problems, if you get better by doing things you will find that your players spend a lot of time hitting / climbing trees just for the sake of doing it. And endlessly climbing trees is not fun.
If you limit it to "real" experience only, then the player has lost most of the control over his characters development, so either way you are in a difficult situation.

Share this post


Link to post
Share on other sites
On the subject of repetetive action to practice a skill: I had a great deal of "Fun" by setting the speed in quest to glory as high as it would go, and maxing out my throw skill and strength in short order hucking rocks... That way I avoided the non-fun of dying while I explored the rest of the game.

A good game is still good even when there is a way to make all but the hardest parts of it easy, so if the rest of the game stacks up well, don''t lose any sleep over making the stat maxing too easy. A lot of players like that sort of thing anyway.

Better too easy than too difficult, in nearly every case.

Visit my web site.
"I came, I saw, I coded."

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
Daggerfall did this. You did not go up levels and gain "skill points" to spend on different skills. Instead you skills would improve as you use them. If you run a lot, running improves. Swim a lot and swimming improves, etc. Many other games do this as well. (In Daggerfall you could also go up levels, but that only affected basic stats, not skills)

Also, I wouldn''t get too excited about the claims a developer makes about a game they are working on. At this point Project Ego is really is Project *Ego*, 100% vaporware hype. I''m getting sick of people claiming their AI is going to change the world, or that their environments are going to be "truly interactive," blah blah blah.

Now when I read these claims I take them with a bucket of salt. If someone says weilding an axe makes your arm bigger and that sneaking around makes you thinner, I will assume those are the two of the very few main ways your character will physically change. I don''t use my powers of imaginiation and dream up other cool ways my character can grow. Just as when someone makes a game where you can choose the "thief" class, I will assume that the thief classs is useless, as it almost always is.

JM

Share this post


Link to post
Share on other sites
quote:

Original post by Gaiiden

I think Hikeeba is latching on to the new wave of games that actually start to take advantage of all this processing power now at our fingertips.



There's no connection between processing power and the evolving system. ADOM (a Rogue-like RPG) used such a system for all skills, spells and fighting abilities, and it worked on a 33mhz 386 just fine.

quote:

In Project Ego, not only does your character "evolve", but the entire world does to. If you leave your house as a kid and come back as an adult, things will have changed drastically. Also like Hikeeba said, Molyneux is probably using a spinoff of his Black and White AI to accomplish this feat, which is entirely possible.



I doubt the AI in Black and White had the strength to simulate human interactions. I'd rather expect a scripted evolution based on lots and lots of rules :
if (brother.killed) goal = revenge;
if (goal == revenge) action = learn_to_fight;
if (action == learn_to_fight) strength++;

quote:

I think we'll be seeing a lot of games that start to use these "evolving" features as time goes on. It presents a way for the player to personalize his character not through any set means (like leveling up with strictly defined abilities) but by the actual actions he performs in the game. So now it's not just about fighting monsters to get enough skill points to level up, but fighting monsters to improve your swordsmanship, accuracy, strength.... all these general attributes. Levels will be a thing of the past. I think it's cool.



While I agree this is an evolutionary step, it's not a lot more than that, and it doesn't address the real problems of the RPG genre. It is still the same shallow gameplay: perform repetitive action so that a certain stat goes up so that the player can pass certain barries (quests, monsters and the like) so that other stats go up (wealth, new weapons, etc.) so the player can perform another repetitive action that increase some other stats even more and so on and so forth. And for things not related with the main goblin killing gameplay, it may even be worse. Killing goblins is at least mildly fun, as opposed to sawing shirts or cooking. I'd rather kill 1000 goblins to learn to cook than boil 1000 cauldrons of goblin stew.

[edit]
Thank you Hase, you beat me to it.
[/edit]

Edited by - Diodor on December 19, 2001 5:09:18 PM

Share this post


Link to post
Share on other sites
I think the boredom of pure repetition can be handled. If they want to go off and climb trees all day, they can. But, if you had miniquests of a nature that couldn''t help but increase experience while at the same time rewarding the player, you could keep them entertained and advancing.

For scaling the challenges, the constantly evolving world would handle that. Molyneux mentioned competing heroes in Project Ego. I think that same thing could be applied to monsters and everyone else in the world. Practicing over and over wouldn''t do much. The player could train until he''s the fastest sword in the land, but everything else would be too.

What if you had two parts to character advancement? The technical knowledge and the applied. Practicing on trees may make him fast and he may be able to cut in a gnat in half, but what about where to hit an orc? Giving the player precise control over where you''d strike would lead into a VSIM like mess and take the emphasis away from where it should be. So, what about having the character evolve from actual experience? As he interacts with different creatures, he figures out what to do with them. He figures out where to hit an orc to drop it fastest. The player might still aim for the torso, but the character adjusts to hit the most effective spot in the torso.

What and where it gets done would be player controlled, but the tiny little details would go down to the character.

Yes, it''s just an evolution and not a revolution, but I think it''s a step in the right direction.

As for hype, after Black & White, I think the guy can do anything short of create something from nothing.

I wanna'' ride on the pope mobile.

Share this post


Link to post
Share on other sites
An idea system that I was working on:

The concept of "levelling" is gone... skills improve only through use. So, if I use my sword, I get better at it. However, there is also a "level median" on enemies and practice items.

So, say that I''m using a practice dummy for sword fighting... when I''m below the level median, I get more "skill experience" for hitting it. When I pass the median, I get less and less experience for hitting the dummy (meaning that the value of hitting a training dummy will degrade over time, eventually getting to zero use -- granted, a training dummy would only be useful to a low number of skill levels).

Enemies work the same way -- when I first get into a sparring war with a goblin, it will be worth more skill experience than later on (eventually, destroying a goblin with my sword would get me nothing for my sword skill). However, if this same person suddenly picked up an axe or a mace, it would be worth a lot more experience to that particular skill. To stop people from "bottom levelling" all of their skills (meaning, going into a goblin area, and getting all of their combat skills to a certain level), I plan to have a finite number of enemies in the game OR make it so that all combat skills are grouped for the level median (meaning that the average of all short-ranged combat skills are compared to the level median, as opposed to each individual skill -- either you can become much better at one skill against a horde of a single type of enemy, or moderately better with multiple skills of the same "class").

Ranged skills (bow/sling) would need to be handled slightly differently. Even if I''m shooting a slow goblin, if I''m shooting that goblin from 200 yards out, it should be worth SOMETHING. So, the distance would be a factor, as well... even though I''m not getting points for the goblin itself, I''d get points for the distance (until that distance becomes easier for my archer, anyway).

Magic would be a combination of multiple things: casting, range, scope, and effect would all come into play. Casting the spell to begin with should be worth something, maybe towards that "class" of spell, range would of course play a factor, scope would be the number of targets the spell was intended against, and effect, such as acheiving the desired effect against the target would all need to be considered when giving the exerience for that particular spell.

As a side note, even missed attacks (any type) should get SOME experience (not as much as contacting, but still). If I swing a sword at something, and miss, I''m still learning for next time, just not as much as if I actually hit.

Wizardry 8 does a reasonably good job of handling this issue. They still do levelling up, but character skills grow through use. It''s still not perfect, but it''s a fun game (which is really more important than anything else, no matter how purist many of us try to get)

-Chris

Share this post


Link to post
Share on other sites
quote:
characters _do not_ improve. They can have stats, they can have skills and all that, but whatever happens, in but one week or two (the length span of a game) these will not change at all.


I''m not sure what games you''re playing, but the RPG games that I play don''t have some weird 7-14 day span, they have several months or years span.

As for not improving, I joined the army. In 8 short weeks I went from 140 pounds (at 6'' tall), relatively weak, incapable of shooting a rifle beyond very basic skill to 180 pounds, maxxing out the PT test and able to score a near perfect in the rifle marksmanship test. People very clearly _do_ improve and they do so very quickly.

Maybe you have problems learning, but most people in higher stress situations improve very quickly and I can''t imagine a higher stress situation than combat.

quote:
how about we finally accept that 10 goblins will eat alive the meanest hero any time day or night


Yes, in a ring with restricted movement, 10 anythings will eat any person alive. In a relatively open environment though, a highly experienced fighter will defeat several opponents and if the difference in skill is high enough, he will do so while taking very little damage himself. I have witnessed a bouncer take on 6-8 people at once and win, leaving the bar with only a fat lip.

There have also been situations where a very small number of people have taken out several times their number while taking very low casualties. This is due to discipline. Goblins aren''t known as the most disciplined race in the fantasy genre and as a result, all will be seeking the glory at the same time and probably interfere with each-other''s attempts. Look through history, you''ll find that your statements are very far from what history has taught.

quote:
I doubt the AI in Black and White had the strength to simulate human interactions. I''d rather expect a scripted evolution based on lots and lots of rules :
if (brother.killed) goal = revenge;
if (goal == revenge) action = learn_to_fight;
if (action == learn_to_fight) strength++;



I seriously doubt that the AI is nearly so pitiful. I don''t think that Molyneux is the second coming or anything like that, but he isn''t doing scripting that is barely better than what I could write back on my C=64.

I guess I''ll go on to the actual meat of the topic now.

quote:
What if you had two parts to character advancement? The technical knowledge and the applied.


How I''m dealing with this is to give a small increase in skill and a bonus to experience gained toward that skill for a certain period of time (to represent that training alone can''t confer full understanding of what is taught, actual application in a real situation is what cements the trained abilities). The reason for the time maximum is that if you don''t use it quickly enough, you will easily forget it and be back to square one, but with a few extra tricks and a bit more intuition.

It''s entirely possible to do it the way that you''re talking about, but the skill system that I''m using takes up enough space on the DB server as it is right now. Giving every skill 2 ratings would just slow it down that much more.

quote:
it will be worth more skill experience than later on (eventually, destroying a goblin with my sword would get me nothing for my sword skill). However, if this same person suddenly picked up an axe or a mace, it would be worth a lot more experience to that particular skill.


Yes and no. One thing to consider is that part of what you learn when fighting a particular type of opponent isn''t how to use your weapon (though that is the larger portion), it''s how the opponent moves and reacts. If you use a sword on goblins until I''ve defeated 200 of them and then switch to a mace, the goblins will still have the same tendencies in attack and defense. A lot of that knowledge will still work with a mace. Does that mean I''ll learn to use the mace that much more quickly or just that I know more about the way critters move? Realistically, no matter how unskilled King Arthur is at using a mace, he won''t learn anything by fighting them though, he knows too much about fighting and they will be an almost painfully easy opponent, even with a weapon that he is unfamiliar with.

You could propose knowledge specific to critter types, but that''s also somewhat impractical due to database performance issues. Even on a local machine, it has to keep track of a large amount of data and the way that it will be doing that is to use a local database. Adding more data to keep track of will simply slow down the game and it will actually get worse and worse the longer you play the game.

quote:
I think the guy can do anything short of create something from nothing.


I would say that is a mistake. I don''t think that anyone should think that someone else can do any more than they personally can. He''s just a man. Perhaps with relatively good ideas and an ability to get other people that can implement his ideas well, but still just a man. I don''t think that he still does a lot of programming, just producing. As a result, he''s not doing much in and of himself, he''s simply telling other people what he wants to see and gives them an idea on how to implement it, but it''s the people under him that are really making it happen.

Some people do, some teach, some lead. The key is knowing what you are good at and learn to do the others. My best skill is to teach (I''ve been told that I can put anything into terms that a moron could understand), but I''m trying to learn to lead better and most definitely to get off my butt and do. Molyneux is just a very good leader who is somewhat good at doing and possibly a bit better at teaching. Don''t think that he can part water because he didn''t write 2 million+ lines of code for B&W and he won''t be writing 3 million+ for his next game either, a team of other people will be.

Share this post


Link to post
Share on other sites
quote:

Original post by solinear

I''m not sure what games you''re playing, but the RPG games that I play don''t have some weird 7-14 day span, they have several months or years span.



And they do that by artificially increasing time by as much as 10 times.


quote:

As for not improving, I joined the army. In 8 short weeks I went from 140 pounds (at 6'' tall), relatively weak, incapable of shooting a rifle beyond very basic skill to 180 pounds, maxxing out the PT test and able to score a near perfect in the rifle marksmanship test. People very clearly _do_ improve and they do so very quickly.



8 weeks real life would equal around 10 hours of gameplay. I''d be quite bored if I''d have to have my character shoot targets and do pushups for 8 hours just to increase his weight and accuracy. I wouldn''t place such boring exercises during game time, but during pauses in the game time (like in MM6 - leveling up would also use one week''s time - training time)

Aside from that, you started from very low, so you could have learned fast. If you spent another 8 short weeks learning the same things, you woud have evolved but a fraction of this. If you spent another 8 weeks yet a smaller fraction.

Do I need to even mention that you didn''t have to actually kill someone to learn those skills?f

quote:

Maybe you have problems learning, but most people in higher stress situations improve very quickly and I can''t imagine a higher stress situation than combat.



They may well do so, but some things do not improve. All the training in the world won''t save anyone from dieing when shot in the head, or stabbed in the wrong place.

quote:

Yes, in a ring with restricted movement, 10 anythings will eat any person alive.



This is actually good enough for me, as it would by itself be a huge change to RPGs combat systems.

quote:

In a relatively open environment though, a highly experienced fighter will defeat several opponents and if the difference in skill is high enough, he will do so while taking very little damage himself. I have witnessed a bouncer take on 6-8 people at once and win, leaving the bar with only a fat lip.



Hmm, a bar fight is still a simulated kind of fight. Noone wants to kill the enemy, noone wants to risk getting seriously hurt. If the 8 would have rushed the bouncer, all at once, hitting him with chairs and throwing bottles at him...

Coming back to goblins, and assuming they are willing to risk their lifes to frag the hero
- if they find an archer, they can storm him - he won''t be able to shoot them all - and rip him apart because of his lack of close weapons support.
- if they met a heavily armored knight, they could move faster than him, suround him, jump on his back, dodge his slow sword movements, imobilize him and stab him to death.
- if they met someone with less armor and no ranged weapons, they could stone him to death.

quote:

I seriously doubt that the AI is nearly so pitiful. I don''t think that Molyneux is the second coming or anything like that, but he isn''t doing scripting that is barely better than what I could write back on my C=64.



You couldn''t write a million lines of script on a C64. I refuse to believe that in a game where all the player does is kill stuff the NPCs will have an AI that can do more than react in a scripted fashion to the few actions of the player (kill, steal, uhm.. I forget)


quote:

You could propose knowledge specific to critter types, but that''s also somewhat impractical due to database performance issues. Even on a local machine, it has to keep track of a large amount of data and the way that it will be doing that is to use a local database. Adding more data to keep track of will simply slow down the game and it will actually get worse and worse the longer you play the game.



You can divide critters in categories. Put goblins, kobolds, gnomes in one categ, lions, tigers, dagger toothed cats in another, and so forth. Then you''d have a single knowledge attribute for each category.


quote:

I would say that is a mistake. I don''t think that anyone should think that someone else can do any more than they personally can. He''s just a man. Perhaps with relatively good ideas and an ability to get other people that can implement his ideas well, but still just a man. I don''t think that he still does a lot of programming, just producing. As a result, he''s not doing much in and of himself, he''s simply telling other people what he wants to see and gives them an idea on how to implement it, but it''s the people under him that are really making it happen.



Communist

Share this post


Link to post
Share on other sites
quote:
Original Post by Hikeeba
So, what about having the character evolve from actual experience? As he interacts with different creatures, he figures out what to do with them. He figures out where to hit an orc to drop it fastest. The player might still aim for the torso, but the character adjusts to hit the most effective spot in the torso.


Now that would be cool. Not only would the character be able to evolve to a more precise level of development, but the player will not be troubled with directly manipulating the character to attain this level, so all the MM freaks can go to Hell, muahaha. Apply this on a broader scale and you got a nice system.
quote:
Original Post by Diodor
(Original Post by solinear
I'm not sure what games you're playing, but the RPG games
that I play don't have some weird 7-14 day span, they have
several months or years span.)

And they do that by artificially increasing time by as much as 10 times.


Hee hee, I was about to razz you along those same lines when I read a bit more carefully. I still wasn't sure if that was what you mean tho.
quote:
Original Post by crouilla
So, say that I'm using a practice dummy for sword fighting... when I'm below the level median, I get more "skill experience" for hitting it. When I pass the median, I get less and less experience for hitting the dummy (meaning that the value of hitting a training dummy will degrade over time, eventually getting to zero use -- granted, a training dummy would only be useful to a low number of skill levels).


But let's not forget that after a long time of disuse, skills can be lost. Let's can the term "zero use" for "Zero additional skill development". Your term implies that we no longer have a need for it when that's not true, as your character should continue to practice on it to maintain his level of skill. So from that we derive that over time, skills degrade in the game if they are not practiced. The more skilled you are, the less you have to practice, but you still have to every once in a while.
quote:
Original Post by solinear
I would say that is a mistake. I don't think that anyone should think that someone else can do any more than they personally can. He's just a man. Perhaps with relatively good ideas and an ability to get other people that can implement his ideas well, but still just a man. I don't think that he still does a lot of programming, just producing. As a result, he's not doing much in and of himself, he's simply telling other people what he wants to see and gives them an idea on how to implement it, but it's the people under him that are really making it happen.


Hey! Don't be knocking my buddy Pete! Seriously tho, even if there is a talented team behind a game, it don't mean squat if their leader doesn't bind them together with a common vision. By leading a team in production you are doing plenty, beleive me. Molyneux is quite a visionary, he'll get it done.

_________________________________________________________________

Drew Sikora
A.K.A. Gaiiden

ICQ #: 70449988
AOLIM: DarkPylat

Blade Edge Software
Staff Member, GDNet
Public Relations, Game Institute

3-time Contributing author, Game Design Methods , Charles River Media (coming GDC 2002)
Online column - Design Corner at Pixelate

NJ IGDA Chapter - NJ developers unite!! [Chapter Home | Chapter Forum]

Share this post


Link to post
Share on other sites