Originally posted by Creslin321

 

It also doesn't START being deterministic because there are trillions of factors. That argument goes both ways ;).

This is how I think about it...

You are conscience for a reason. That reason is that the human body has a...driver. Something making non-deterministic decisions for it. If you were completely deterministic like a machine, you would not need a driver, and would not be conscience. I think that the fact that you are conscience is evidence of free will...because without free will conscienceness would be vestigal. Cogito ergo sum.

All of those trillion factors deterministically result in a decision that, were those trillion factors fully observable and understood, you could predict with 100% accuracy.  Exactly like a computer program.

Also, "conscious" / "consciousness".

How can you possibly use motivations ("drivers") as a reason we're not deterministic?  Motivations are part of why we are deterministic!  They're part of the program!  Seek food.  Seek shelter. Seek a suitable mate.

As for free will, "will" is essentially the same as motivation.  You "excercise free will" to go to the delicious burrito place, but really it was a deterministic decision resulting from (1) having not had burritos in a while, (2) you're motivated by hunger, (3) past experience with burrito places (and this place in particular) serving delicious burritos, and a trillion other little factors.  Or maybe tomorrow you'll go to the sushi place instead because (4) you read some post on a forum saying that free will is a myth and you want to intentionally do something out of the ordinary to excercise free will.  Which would also be deterministic.

Even in games, AIs have drivers, motivations, desires, will.  Why did they attack the tank?  Because the situation made them want to.  Why did you choose the burrito place?  Same deal, just with a lot more variables and complex factors involved in the programming.

Originally posted by Axehilt

Professional AI programmers (people who've made multiple games and watched how players play them and react to various methods of AI) would point out that if the player doesn't perceive it it doesn't exist.

In this case Bob's AI gives out the same exact quest he could've just been hardcoded to do.  Most (possibly all) players won't know the difference, and if we don't know the difference the underlying logic is completely meaningless.

Time and time again when I read articles about AI the important keys to success are about giving the perception of intelligence.  The actual need for true intelligence isn't there -- in fact in many situations intelligent AI is detrimental to the fun of a game.

So while AI can certainly stand to be more game-like, less exploitable, and less perceptibly stupid, it's actually not a good idea to make it more intelligent for intelligence sake.  

This is interesting, I feel the op was trying to point out that what a good MMO needs is an AI that is believable as opposed to just more intelligent. I certainly agree with you that more AI is not a good idea, only because AI in games is not about making them smarter, but in fact making them less intelligent and 'believable'. It is really easy to actually create an AI that is pure rational logic, but it is not easy to create an AI who changes its mind in the middle of performing an action and "not" oscillating in the process.

The common example is, to program an AI to make every mathmatical shot in a game of pool is easy, to program that AI to miss a shot and it be 'believable', now that is hard.

I agree with the OP that I would like to also see more believable AI's in MMOs, and based on AI conventions and articles from aigamedev.net, I would venture that we are getting closer and closer to merging some academic AI with developers of game AI.

In the not too distant future, behavioral AI will become more apparent as 2010 and 2011 demonstrated nice leaps forward.

There's a lot of interesting rhetoric in this thread, but a lot of people seem to forget that AI programming is a logic puzzle.

You're better off spending this time finding a way to make it work, then trying to find the esoteric value of making an NPC more human.

Originally posted by Creslin321
Exactly.

You can't quantify feelings and you can't quantify conscienceness. You CAN quantify what is commonly referred to as intelligence though, and that's why why you can create it with a program.

I think it's funny that people have used their free will to argue...against the existence of free will lol :).

Also...some people have said that I'm arguing in favor of free will because I'm religous or something...I'm not. I'm arguing for free will because I'm a programmer lol :). I write programs...I know what they can do. Trust me, they can't have freewill.

You could literally replicate what a program does with a massive mechanical machine. Do you really think that a planet sized mechanism of gears and pistons could be self aware?

There wasn't free will.  It was the result of our environment, our past experiences/learnings (including the desire to see truths strongly presented and perceived falsehoods argued against), and a trillion other factors.  There was will, but it wasn't free, it was inevitable and deterministic.

The earliest automobile engineers probably claimed it was impossible to travel faster than 100mph too.  Although most of them probably knew enough about engineering to imagine that eventually it would be possible.

There is no "magic" to the brain or consciousness or feelings.  Any magic to it is simply a limitation of our current understanding (much like early automobile engineers.)  They work in deterministic ways and we simply have to fully learn how they work to then set to work on how to fully replicate it.

Originally posted by GTwander

There's a lot of interesting rhetoric in this thread, but a lot of people seem to forget that AI programming is a logic puzzle.

You're better off spending this time finding a way to make it work, then trying to find the esoteric value of making an NPC more human.

This is an interesting post, typical AI in games has been mostly Finite State Machine (FSM), it is still used today, in fact, many of the games today thought of to have fantastic AI are FSM based. This is an older concept of AI and not on par with academia AI. However, I would argue that even behavioral based AI will have some kind of emergent behavior. In that, they will act irrationally much like a human would.

For example, I wrote an AI program of a farmer who, on occasion, would get up in the middle of the night to go work in the field based on various parameters. I first thought it was a programmatic issue and I sought ways to curb his interest in sometimes working in the field at 11pm... but it was doing exactly what it was suppose to.

Humans are irrational beings, we do not make rational decisions and in many ways even with the obvious rational decision right in front of us, we still make the irrational decision... Like vacuuming at 10pm at night.

I will venture that implementation of realistic NPCs who behave in various manners would elicit some kind of emotional response from players which would encourage attachment... It is possible now, even in MMOs.

Originally posted by stealthbr

Quick question here: What do you guys think of NPC's behaving according to day/night cycles (NPC's going to sleep)? Do you think it would work in an MMO?

I would require it in an MMORPG of my choosing, but I am very picky these days.

Originally posted by Axehilt

Originally posted by Creslin321

 

It also doesn't START being deterministic because there are trillions of factors. That argument goes both ways ;).

This is how I think about it...

You are conscience for a reason. That reason is that the human body has a...driver. Something making non-deterministic decisions for it. If you were completely deterministic like a machine, you would not need a driver, and would not be conscience. I think that the fact that you are conscience is evidence of free will...because without free will conscienceness would be vestigal. Cogito ergo sum.

All of those trillion factors deterministically result in a decision that, were those trillion factors fully observable and understood, you could predict with 100% accuracy.  Exactly like a computer program.

Also, "conscious" / "consciousness".

How can you possibly use motivations ("drivers") as a reason we're not deterministic?  Motivations are part of why we are deterministic!  They're part of the program!  Seek food.  Seek shelter. Seek a suitable mate.

As for free will, "will" is essentially the same as motivation.  You "excercise free will" to go to the delicious burrito place, but really it was a deterministic decision resulting from (1) having not had burritos in a while, (2) you're motivated by hunger, (3) past experience with burrito places (and this place in particular) serving delicious burritos, and a trillion other little factors.  Or maybe tomorrow you'll go to the sushi place instead because (4) you read some post on a forum saying that free will is a myth and you want to intentionally do something out of the ordinary to excercise free will.  Which would also be deterministic.

Even in games, AIs have drivers, motivations, desires, will.  Why did they attack the tank?  Because the situation made them want to.  Why did you choose the burrito place?  Same deal, just with a lot more variables and complex factors involved in the programming.

We can run around with these arguments in circles for days I'm sure, but we're never going to get anywhere.  I will never convince you that free will exists, and you will never convince me otherwise because it's a philosophical argument.  There are no facts on either side that prove anything.

However, you seem to believe that the universe itself is deterministic in all aspects, and that is what you are basing your philosophy on free-will's nonexistance on.  That, I can argue against.

I don't know if you've looked into quantum mechanics, but basically it's the study of subatomic particles.  Quantum physicists have discovered that some subatomic particles act apparently "randomly" i.e, they are not deterministic.  The implications of this is that the universe itself can't be completely deterministic because the building blocks of matter are not.

For more information, check out the "Heisenberg Principle."  And these links:

Michio Kaku on free will and non-determinism:  http://www.youtube.com/watch?v=lFLR5vNKiSw

Quantum indeterminacy:  http://en.wikipedia.org/wiki/Quantum_indeterminacy

Originally posted by Creslin321

Originally posted by Axehilt

Originally posted by Creslin321

 

It also doesn't START being deterministic because there are trillions of factors. That argument goes both ways ;).

This is how I think about it...

You are conscience for a reason. That reason is that the human body has a...driver. Something making non-deterministic decisions for it. If you were completely deterministic like a machine, you would not need a driver, and would not be conscience. I think that the fact that you are conscience is evidence of free will...because without free will conscienceness would be vestigal. Cogito ergo sum.

All of those trillion factors deterministically result in a decision that, were those trillion factors fully observable and understood, you could predict with 100% accuracy.  Exactly like a computer program.

Also, "conscious" / "consciousness".

How can you possibly use motivations ("drivers") as a reason we're not deterministic?  Motivations are part of why we are deterministic!  They're part of the program!  Seek food.  Seek shelter. Seek a suitable mate.

As for free will, "will" is essentially the same as motivation.  You "excercise free will" to go to the delicious burrito place, but really it was a deterministic decision resulting from (1) having not had burritos in a while, (2) you're motivated by hunger, (3) past experience with burrito places (and this place in particular) serving delicious burritos, and a trillion other little factors.  Or maybe tomorrow you'll go to the sushi place instead because (4) you read some post on a forum saying that free will is a myth and you want to intentionally do something out of the ordinary to excercise free will.  Which would also be deterministic.

Even in games, AIs have drivers, motivations, desires, will.  Why did they attack the tank?  Because the situation made them want to.  Why did you choose the burrito place?  Same deal, just with a lot more variables and complex factors involved in the programming.

We can run around with these arguments in circles for days I'm sure, but we're never going to get anywhere.  I will never convince you that free will exists, and you will never convince me otherwise because it's a philosophical argument.  There are no facts on either side that prove anything.

However, you seem to believe that the universe itself is deterministic in all aspects, and that is what you are basing your philosophy on free-will's nonexistance on.  That, I can argue against.

I don't know if you've looked into quantum mechanics, but basically it's the study of subatomic particles.  Quantum physicists have discovered that some subatomic particles act apparently "randomly" i.e, they are not deterministic.  The implications of this is that the universe itself can't be completely deterministic because the building blocks of matter are not.

For more information, check out the "Heisenberg Principle."  And these links:

Michio Kaku on free will and non-determinism:  http://www.youtube.com/watch?v=lFLR5vNKiSw

Quantum indeterminacy:  http://en.wikipedia.org/wiki/Quantum_indeterminacy

Random is not free will. It doesn't really matter if something is perfectly random or deterministic because neither of those result in free will. This is the indeterminancy irrelevancy principle. Free will adovcates always bring that up like it means something.

Originally posted by Cuathon

Originally posted by Creslin321

Originally posted by Axehilt

Originally posted by Creslin321

....

All of those trillion factors deterministically result in a decision that, were those trillion factors fully observable and understood, you could predict with 100% accuracy.  Exactly like a computer program.

Also, "conscious" / "consciousness".

How can you possibly use motivations ("drivers") as a reason we're not deterministic?  Motivations are part of why we are deterministic!  They're part of the program!  Seek food.  Seek shelter. Seek a suitable mate.

As for free will, "will" is essentially the same as motivation.  You "excercise free will" to go to the delicious burrito place, but really it was a deterministic decision resulting from (1) having not had burritos in a while, (2) you're motivated by hunger, (3) past experience with burrito places (and this place in particular) serving delicious burritos, and a trillion other little factors.  Or maybe tomorrow you'll go to the sushi place instead because (4) you read some post on a forum saying that free will is a myth and you want to intentionally do something out of the ordinary to excercise free will.  Which would also be deterministic.

Even in games, AIs have drivers, motivations, desires, will.  Why did they attack the tank?  Because the situation made them want to.  Why did you choose the burrito place?  Same deal, just with a lot more variables and complex factors involved in the programming.

We can run around with these arguments in circles for days I'm sure, but we're never going to get anywhere.  I will never convince you that free will exists, and you will never convince me otherwise because it's a philosophical argument.  There are no facts on either side that prove anything.

However, you seem to believe that the universe itself is deterministic in all aspects, and that is what you are basing your philosophy on free-will's nonexistance on.  That, I can argue against.

I don't know if you've looked into quantum mechanics, but basically it's the study of subatomic particles.  Quantum physicists have discovered that some subatomic particles act apparently "randomly" i.e, they are not deterministic.  The implications of this is that the universe itself can't be completely deterministic because the building blocks of matter are not.

For more information, check out the "Heisenberg Principle."  And these links:

Michio Kaku on free will and non-determinism:  http://www.youtube.com/watch?v=lFLR5vNKiSw

Quantum indeterminacy:  http://en.wikipedia.org/wiki/Quantum_indeterminacy

Random is not free will. It doesn't really matter if something is perfectly random or deterministic because neither of those result in free will. This is the indeterminancy irrelevancy principle. Free will adovcates always bring that up like it means something.

Like I said...you can run around with these arguments for days.  There is no hard proof for or against free will.  It's just what you personally want to believe in.

If you have some hard proof against free-will...I would like to hear it ;).

Originally posted by Creslin321

Like I said...you can run around with these arguments for days.  There is no hard proof for or against free will.  It's just what you personally want to believe in.

If you have some hard proof against free-will...I would like to hear it ;).

In this case it does feel like a religious argument, where one person asserts something arbitrary and unprovable (either way).  Like god or the giant invisible flying spaghetti monster, they're concepts without evidence where the burden of proof is on the person insisting the're true to provide the evidence.

And like those other baseless ideas (ideas designed to be resilient) there's an element encouraging the idea-holder to misinterpret their observations as evidence of the idea's truth.

Originally posted by Axehilt

Originally posted by Creslin321

Like I said...you can run around with these arguments for days.  There is no hard proof for or against free will.  It's just what you personally want to believe in.

If you have some hard proof against free-will...I would like to hear it ;).

In this case it does feel like a religious argument, where one person asserts something arbitrary and unprovable (either way).  Like god or the giant invisible flying spaghetti monster, they're concepts without evidence where the burden of proof is on the person insisting the're true to provide the evidence.

And like those other baseless ideas (ideas designed to be resilient) there's an element encouraging the idea-holder to misinterpret their observations as evidence of the idea's truth.

Bascially I assume people who believe in free will are religious. I know that some atheists have problems with this. It seems weird to me that they would. I think its just that practically we cannot utilize the idea that the universe is deterministic to do anything wrt the behavor of living things, be they human or animal due to the complexity.
We can get diminishing returns on constantly pushing the borders of our understanding of factors, but we cannot expect to prevail in all instances.

Crime and acting out and teenagers being prime examples.

"Free Will" as a concept doesn't make sense when you talk about decision making. Assume humans have a soul, and that soul is involved in the decision making process. The decision making process is the same with or without a soul. Pick the best possible thing from a set of possible things. Most of the time, humans have a good idea why they chose the thing they chose, which means there were reasons why. Even when a person doesn't know why they chose red or like red, there's still a reason. Even if the reasons are based on ideas like 'good', 'evil', 'conscience', 'God said so', etc., they are still reasons. Those reasons determined the decision that was made. This doesn't mean Free Will doesn't exist, but you can't determine whether or not Free Will exists by using decision making as an example.

Or, since part of the definition of Free Will is acting at one's own discretion, then we all have free will because we are free to make decisions. The decisions made are all deterministic, but the fact that we can make decisions in the first place is free will.

Whatever.

It's easier and far more efficient to simulate something that looks intelligent than to create something that is intelligent. Especially in games. What if you created a true AI and it never did what you wanted it to? Everything hinges on the AI, and all it does is pick apples and kick dogs so your game is boned.

AI is cool, but it has to have some use and it has to do what you want...which limits the amount of intelligence it really has. If you don't add in random elements to the decision making, it will also become very predictable...which means it will be exploited by players.

Originally posted by Axehilt

Professional AI programmers (people who've made multiple games and watched how players play them and react to various methods of AI) would point out that if the player doesn't perceive it it doesn't exist.

In this case Bob's AI gives out the same exact quest he could've just been hardcoded to do.  Most (possibly all) players won't know the difference, and if we don't know the difference the underlying logic is completely meaningless.

Time and time again when I read articles about AI the important keys to success are about giving the perception of intelligence.  The actual need for true intelligence isn't there -- in fact in many situations intelligent AI is detrimental to the fun of a game.

So while AI can certainly stand to be more game-like, less exploitable, and less perceptibly stupid, it's actually not a good idea to make it more intelligent for intelligence sake.  

The funny part is that even if you had better AI, that doesn't mean it would lead to more variation in the resulting combat.