Artificial Intelligence is a term that has been up for debate the moment computers came into existence.
Basically it’s a decision that needs to be answered by the philosophers.
Regardless, artificial intelligence can be divided into two categories:
Strong AI
Weak AI
Strong AI
Strong AI tends to use complex algorithms that give it the ability to act
differently in different situations.
It’s AI that can learn and adapt to any situation.
In a sense it has a mind of its own and can make independent decisions.
It has the ability to learn and adapt as a human can, therefore it is basically human.
Examples of Strong AI:
Google DeepMind AlphaGo
Google program where AI taught itself to play and master 1980’s arcade classic games
Weak AI
This is the AI we are used to now-a-days.
It can simulate human behaviour, but it does not learn and adapt.
It’s a system or program that is pre-programmed to act out in a certain way.
Examples of Weak AI:
iPhone (Siri)
Amazon (Alexa)
Google.com (Search Results)
Self-driving cars
IBM Watson (Jeopardy)
Deterministic vs Non-Deterministic AI
When it comes to game programming, there are two types of AI you will be dealing
with:
Deterministic
Non-Deterministic
Deterministic AI
Deterministic AI is the simplest form of AI to implement in games.
It’s creating a system or object where behaviour is predictable.
Due to the static nature of the implementation, this makes it harder to program and design as both game programmer
and game designer must come up with all possible player scenarios.
An example of deterministic AI is our chase/evade AI in our current Pong game.
Non-Deterministic AI
Non-Deterministic AI is one of the hardest things to implement in game programming.
This is because we have to write our code in such a way that the system or object is capable of learning and adapting
to the player.
Since Non-Deterministic AI is adaptable to the player, this means our AI systems behaviour is unpredictable because every players behaviour is unique.
An example of a non-deterministic AI is neural networks.
Goals of AI in Games
There are many purposes for AI in video games, however the most important reason for game AI is to
give life to our game characters
I also want to touch on the idea of varying intelligence of your game characters in your game.
Just like in real life, you most likely want to vary the intelligence of your game characters.
Does a human have the same intelligence as a zombie?
Does a robot have the same intelligence as a primitive alien?
Game AI as a Competitor
One reason to implement AI in your game is to create a competitor or obstacle for your player.
This is the most common type of AI that is implemented in a majority of games.
In our current pong game, we use AI as a competitor to challenge the players' skill in Pong.
The point is to create frustration for the player but not too much for the player to quit.
AI to Guide Player Game Making Choices
Another use case for AI in video games is to force the players to make decisions.
In the game PacMan, the ghost AI forces the player to:
Avoid the ghosts
Plan movement around the ghost
Sparingly use powerups to scare the ghost away
Basically the player is not allowed to move around freely.
On top of that the AI in the game, due to constantly chasing you also add tension.
AI to Guide Player Emotions
The last, and most overlooked use case of AI is that we can use AI to guide player emotions.
One simple strategy of this is to use AI to add and relieve tension in our games.
The best example of this is the Xenomorph in the game Alien: Isolation.
The Xenomorph adds tension to players when they are roaming the same room as the player.
The Xenomorph can also release tension to the player when it leaves the room.
In this sense if you wanted to add moments of tension on the player, add the Xenomorph; if you want
to relieve player tension, the remove the Xenomorph from the room.
Summary
I have only touched the surface on AI game design, and hoping to have inspired you to go and think of your
game objects and characters as more than pixels on the screen.
Transcript
Hello, Godot Tutorials is not sponsored by or affiliated with the Godot game engine. In this episode, we will take a look at the following strong and weak A.I., along with deterministic versus non-deterministic A.I. in game programming, the goals of using A.I. and games using a as a competitor,
such as in our
Pong game, using A.I. to guide player game making decisions such as Resident Evil Village and using it to guide player emotions such as alien isolation. Before we begin, though, we must first answer a question. What is artificial intelligence? However, we can break this down into an even simpler question, which is what is intelligence? And this is a question for philosophers. And it really is up for debate over artificial intelligence does have a litmus test.
And that question is basically how close is it to human intelligence? And we can break this down into two categories, strong AI and weak AI. Let's go ahead and take a look at strong and weak.
A strong
AI tends to use complex algorithms that give it the ability to act out differently in different
situations. Basically, it's
able to learn and adapt. You can think of strong as having a mind of its own in which it can process and make independent decisions.
As a matter of
fact, learning and adapting is pretty close to human intelligence. I would argue it is human nature that humans don't learn everything and or anything instantly. It takes time, patience and lots of practice along with iteration. Now, strong AI is basically the opposite of weak and weak A.I. relies on human interference. Basically weak AI is pre-programmed
to do
a certain task and that's a matter of fact, weak. I can only simulate human behavior. It does not learn and or adapt to new environments or situations. And this is again, because it is preprogrammed to act out in a certain way. So let's take a look at some examples of weak A.I. We have Siri from iPhone, we have Alexa from Amazon. We have Google. Com search results, self-driving cars and IBM. Watson, innocent Syrian Alexa are pretty smart
in what
they can do and what they have been able to achieve. And although it responds to your voice like a personal assistant would, it cannot give you any recommendations outside of the questions
us, for example, let's say
us, Siri, that you want to see a movie and all the movie theaters are close. Well, Siri doesn't give you alternative recommendations, like a personal assistant would, for example, recommending to rent or buy a movie that is similar to the one you wanted to watch in theaters.
We can take
this to the next level with Google dot com search
results. Now, when
you type into Google, Google will recommend results to you based on your search history, along with other information it has on you, such as your age, your gender and more importantly, your location,
because
different locations tend to want different
results. Now, all
of these things are quite complex and how they are created and how they give you back information. But regardless, these are all considered weak because again, these are
preprogrammed and they do not
learn and or adapt to you outside of information and already has on you. Now some examples of strong AI I is hard to find
because strong
I can take many things into consideration, but ultimately strong is determined
by whether or not it is able
to adapt and learn from different
situations. Two examples I
have in mind. The first is Google deep mind also go where it taught itself the game. Go and beat the number one player in the world in I believe twenty sixteen.
The second is coming from
Google again, and it basically is a program where and I taught itself to play and master nineteen eighties arcade classic games. As a matter of fact the I taught itself strategies that were not originally programmed, nor did the researches at Google expect the to come up with certain strategies for certain games. Now to give you an idea between strong and weak AI, which is up for debate, let's talk about chess. So there are two ways and I can handle playing chess against a human player.
The weak
A.I. version of playing chess is to calculate two billion possible moves and then picking the best
choice. This is
different. Then a stronger. Playing chess as strong as I will
actually calculate
the opponent's tempo move patterns, they shall tell signs such as whether the opponent is stressed or calm, perhaps even taking in nervous tics from the opponent. And then from all these considerations, it will pick the best possible move against its unique individual opponent.
And the more the
strong plays against the
opponent, the better
the stronger.
I will calculate
its next move based on stimulus
provided
to it by the opponent human
player. Now, regardless,
there is a little bit of confusion between strong AI and artificial general intelligence.
When people
say strong AI, what
they sometimes
mean is ajai, which is just basically that an autonomous agent is able to learn and understand as a human can. And basically this type of A.I. is what you see in sci fi movies
such as
iRobot, and to test for whether or not an autonomous agent has achieved artificial general intelligence. There are many tests you can apply to it. The one I like the most is something called the coffee test. And the coffee test is basically a test where you tell the autonomous agent to make you a cup of coffee. And like a human, it has to find all the ingredients and appliances needed to make coffee. And if some appliances and or ingredients are
missing, the autonomous
agent needs to use the appropriate substitutes.
For example,
if you are missing a coffee machine, the A.I.
may have to adapt
by using hand drip coffee method. And if you're missing soy milk, if you want milk in your coffee, then the autonomous agent must substitute it with an alternative that is healthy, such as
almond or
macadamia milk. Moving on, let's take a look at deterministic and non deterministic A.I. in games. So deterministic A.I. is basically behavior that is predictable. It is static behavior. And this means that the programmer and or designer must come up with all the possible player scenarios
and calculate
against it. One example of deterministic A.I. is the simple Chase and the Vade A.I. implementation we have in our current Pong game. This is the opposite of non-deterministic A.I., which is behavior that is unpredictable
and basically non-deterministic
just means that our A.I. is capable of learning and adapting to the
player. You can think
of this as neural
networks, A.I. implementation, and you could
consider, for example, the game alien isolation. As non-deterministic as the alien learns to adapt to you the player. Regardless, there's only one goal when we add A.I. into our games, and that is to give life to our characters.
And we have
many different characters in games such as humans, robots, zombies, aliens and so forth. And as a matter of fact, ask yourself this question. Are all game characters equal in intelligence? Does a human have the same intelligence as a primitive alien? Does a robot that can calculate two billion calculations per second have the same intelligence
as a zombie?
And this is a question you need to answer for yourself. And as a matter of fact, I would say that giving different variations of intelligence to your game characters add variety and richness to your game content. Now, as a matter of fact, I want to segway into something called user profiles.
For example, in
marketing companies, they create user profiles or customer profiles depending on who you are asking. And basically they use profiles to help create strategies to market against their target audience.
And user
profiles are used to help visualize what these marketing companies are marketing
to. And they
also use pictures to aid
as well.
But in our example, we will not use pictures. And as a matter
of fact, marketing
companies
use profiles
because individuals are unique.
So let's take
a look at our two user
profiles. In a sense,
we have two character profiles that are basically the same in every regard when it comes to the name age, gender, job and income. However, there is a difference between their locations. And this is important to note.
And the reason
for this is that one user profile is located in California, while the other user profile is located in Memphis, Tennessee. And it should become apparent to you why this difference?
Matters, and that
is because someone
making and in this
case, someone making one hundred twenty thousand dollars who lives in California does not have the same purchasing power as someone making one hundred twenty thousand in Memphis, Tennessee. And that is because the purchasing power
of someone
making money in Los Angeles is weaker when compared to someone making the same amount. In Memphis, Tennessee, and this is because for a marketing
company convincing
someone to let go of their money in Los Angeles, California, may require a different approach when compared to someone located in Memphis,
Tennessee. Now, in a sense,
we can use the same strategy they use with marketing companies when it comes to creating our game character profiles specifically for our
eyes. Let's take a look
at an
example. In this case, we
have a game. More specifically, a pong game. And even though we create a character profile for a Pong game, for example, and a guy named Janet Species', human age eight, what we really care about is whether or not it is hard or easy to play against. And in our
case, our A.I., Janet has a pong
mastery of a beginner. Now, we can compare that to a game such as alien isolation, where we have quite a few more things to worry about than our A.I.M. Pong, and that's basically the species and the age. In this case, the species is in endl. Parisot toid alien, and that just means that it needs a living host other than its species to grow inside of during the early stages of development, ultimately killing its host when it has fully developed. And the age is also important for our alien isolation game because of the behavior of Xenomorph and the behavior of Xenomorph is that Xenomorph are designed specifically to kill species that are weaker than themselves.
And in a game such as alien isolation, where a single alien is hunting you down and it is actually quite important to know the age of the alien.
And let's go ahead
and take a look at why age matters to us when we have our game character profiles for games such as alien isolation.
In this case, when we create our game character profiles, let's split them up into two different sections, one for the child and one for the adults. And when you actually create the age and characteristics of each, you'll see that they differ depending on who we are targeting with our character profile.
In this case, we have
an age which is a child. And as a matter of fact, even though it's a kid, its behavior is still to seek and kill species that are weaker than that. However, because our xenomorph is a
child, you will
in fact notice here that the characteristics
change
instead
of something such
as being violent, aggressive and sneaky or characteristics for our child. Xenomorph is cowardly, shy and scared. And the reason for this is that a baby xenomorph has the behavior of something that wants to survive. And this is because of the vulnerability of the child.
Xenomorph a baby
xenomorph is weaker than an adult human
and so a baby
or in this case, the child xenomorph in the face of an adult human has the natural instinct to run and
hide. As a matter of
fact, let's compare this to the adult xenomorph. The adult xenomorph has the characteristics of being violent, aggressive and sneaky, and the reason for that is because our adult xenomorph are stronger than a human adult. And so in that sense are adult. Xenomorph is seeking to hunt down and kill humans. And even though the behavior of an adult xenomorph is the same as a child xenomorph because of the age which determines the size and the height of
our alien
will basically change the characteristics of our alien and in this case, the characteristics of our A.I.. If we were to create an A.I. or a child
xenomorph suddenly in
the game alien isolation, you will be dealing with an adult xenomorph which has the characteristics of being violent, aggressive and sneaky. So let's go ahead and take a look at some use cases of A.I. in video games.
There are many use
cases, however, I have broken them down into three separate
categories. The first use
cases to create obstacles or competitors in our games. Some examples can be found in Mario games and our current Pong game that we are currently creating or almost finishing. The second category is to drive player game making decisions. Some examples can be found in Pacman and the newly released Resident Evil eight village game. The third category is to drive player emotions, and some examples can be found in Bizshark three infinite and of course, alien isolation.
Let's start with obstacles. In this case, in a game like Mario, we have simple pattern movements. For example, our enemy ace just simply move left and right, and the goal of our enemy is to act as an
obstacle and
basically force our player to decide what actions it wants to
take. For example,
over here, we have the choice.
Of getting our mushroom, but it requires the player to head down and face off against two enemy
eyes, and maybe this is
something the player wants to do and maybe not. However, if the player chooses to avoid a confrontation with our enemy, I can simply. Go over them and avoid our enemy a completely and let me go ahead and circle that so you can see the choices.
And so
in this
case, having
an enemy, a forces our
player to
come up with decisions it wants
to take.
In this case, this mushroom looks like a one up for life. And so if the player is in desperate need of life, it needs to actually go through the enemies in order to
get it if the player knows
that the one up is
there. However, if the
player is not desperate for more lives, they can simply skip and avoid it. Moving on, another example of using A.I. is to simply have a competitor for the player. And this is made apparent in our current Pong game. The A.I. we implemented in our Pong game is a simple chase and a
Vade pattern
implementation. It can also be referred to as Push-Pull, for example, push towards the objective, pull away from the objective and in the case of the Pong game,
are enemy
A.I. simply moves up or down based on the position of the ball. And the goal of our A.I. is to act as an obstacle
to the
player achieving a
point. And the
only way the player can overcome this obstacle is to play the AI
now
since we created Chasten Aveda in our game. Let me go over. Chase of the chase of the A.I. consists of two parts. The first is decision to initiate the chase or evade movement B push or pull movement. And then the second part is affecting the chase or evade action. Now basically what determines the decision to initiate and how the game object will chase or evade is up to you, the programmer and or designer. And it also helps to have a game character, a profile to help us visualize what we need to get done in our Pong game.
Regardless, the simplest method of chasing and or evading is changing the game A.I. object's coordinates
based
on the coordinates of the opponent or pray that you want to be affected by or act upon. And so in this case, for our Pong game, our A.I. should act like a beginner and it will act like a beginner by aiming for the center of the pedal.
And then it's a matter of
fact, our pedal changes coordinates based on the balls y axis position. On top of that, because our A.I.
wants to give the effect of
being a beginner, we don't have to do anything complex
outside of our chase
and the Vaidehi
implementation. So the first
step is to initiate the
chase and our pong
game. And to do this is a very simple, if an else
statement, and that is if
our
ball is Y
axis aligned with the no chase zone, then the pedal does not chase
and or push. However, if
our ball
is
outside of the no chase zone of our paddle, then we need to chase for the ball along the Y
axis.
Not once we've initiated the chase. We need to actually affect the chase. We actually need to act upon
chasing
whatever it is we want to chase. And in this
case, in our pong
game, we have something called the variable chase
point, which has a
buffer and the chase point plus the buffer creates our chase zone. And what we want to do is move our chase point to approach the Y axis position of the ball or edge of the screen. And as a matter of fact, if our ball moves down or if our ball is below the no chase zone, then our player paddle chases down and if our ball is above the no chase zone, we move up and that's basically our game. We move down in depending
on whether or not the ball is
above or below our chase zone.
We also have something else we need to consider, and that
is we added
randomness to the chase position and we use the normal distribution to give
us random
chase point values. And the reason we did normal distribution is because we would like our
chase point to more
often be in the center of the paddle and approaching never to either the very top or the very bottom of our paddle. And so, in a sense, like a beginner or a paddle moves to aim itself
to the bow around the
center of the paddle. And we built it a little on purpose because since our profile was that of a beginner,
we didn't
want the chase point to be towards the edges of the paddle. And that is a game design decision you will have to make in your own games as well. Next, let's take a look at how we can use A.I. to guide player decision making. And in a sense, we can use A.I. to force a player to behave a certain way. And this is something a little more complex than an A.I. acting as an obstacle or opponent because it's a little more advanced and implementation and consideration to the player.
Now, let's take a look at the Peckman game in 1980.
The Peckman
game is a game where a player has to eat all the
dots and is
constantly chased by a
goose. And so when creating
an eye for a ghost, we need to ask yourself a very basic question. And that is what is a
ghost and in a sense,
a ghost? Although having the species of a ghost has the characteristics
of hunting an
entity, and in this case, the characteristics of our ghost in the Peckman game is to hunt particular locations, objects or
people. And the behavior
of a ghost, in my opinion, is one that is tireless, persistent and annoying. As a matter of fact, if we take a look at the
ghost in
Peckman, the ghost in Peckman force the player to one, avoid the ghost to plan movement around the ghost, basically head to areas that are not haunted by ghosts. And three, sparingly use power ups to scare the ghost.
And one thing to
keep in mind is that the ghosts in our Pacman game basically force our player or
limit our player to
certain areas of our game map. In a sense, the player is basically not allowed to move around freely. The player needs to take into consideration of where the ghosts are at all times.
And as a matter of
fact, ghosts that constantly chase you adds tension to the game. We will take a look at tension a little later. The next game I want to go over with you is Resident Evil Village. So let's look at how an evil village
forces
player decision
making now in the game
resident evil village. In a very specific part of the game, you have a group of unstoppable
forces hunting
you down. Now, let's ask ourselves another question, and that is what is a human vampire?
And you'll notice
that the similarities of our human vampires share the same characteristics as the ghost in the
Peckman game. And that could be
because of coincidence or simply that vampires and ghosts fit the same genre for the undead. However, we have something here that's different than the ghosts, and that is we are dealing with human vampires and humans by their very nature are intelligent. Humans are adaptable and reactionary creatures.
On top of
that, vampires are unkillable creatures up until the point they are killable, which is just our basic game and movie magic logic. Now in the game are human vampires know that you are in one of the rooms.
They just
don't know which one you are in and sort of like a human trying to find their car
keys, they will
persistently, tirelessly and unknowingly try to search for you and search everywhere until they find you. And so the vampires in the Resident Evil Village game force certain player decisions upon you, and that would include moving slowly and cautiously,
planning
ahead, such as looking into your map, quite often, being resourceful with your weapons. And of course, lastly, if caught by any one of these vampires, give you the feeling and the need to run away, considering that the vampires
are an
unstoppable
force.
Now, the reasons are simple. You move slowly because human vampires, like normal humans, will react to noises such as gunshots and running footsteps.
Basically the.
Human vampires forced upon you, the player, to move slowly because you need to scan the area for the creatures and the noises they make, the signs they give up. On top of
that, the human
vampires force players to look
at the map often
and see if there are escape routes and the likely scenario of getting caught
by one of them. And this
forces the player to plan for contingencies
and to think
slowly and
carefully. On top of that, you
need to be resourceful with weapons because, one, vampires can hear gunshots if within hearing distance. And therefore, when players reach certain enemies, they are forced to use
knives or evasive
tactics in order to avoid drawing the attention of the human vampires.
On top of
that, because the creatures are unstoppable forces until their designated boss battles, players who get cut by one of them have to run to safety, whether through a safe room in the game or by creating such a large gap between you and the human vampire, where the A.I. will be forced to check every room for the player on top of that. This means that the player cannot linger in one area for too long, and so it creates a sense of urgency onto the player.
And on a
side note, the height of one of the
human
vampires gives the emotional state to the player that the creature is really and truly an unstoppable force.
Speaking of adding emotions
to the player, let's take a look at
that. As a matter
of fact, we can use a light to guide player emotions. He can help players feel
emotions, and we can
use a simple strategy such as adding and relieving tension. The first game is Biersack three Infinite,
where our player
is accompanied by our ally and soon to be friend Elizabeth to escape our current predicament. And this forces another question on us. What is an ally? What is
a friend? And the answer
to that question depends on the profile
of Elizabeth, who is
one.
A human, too,
has the characteristics of
friendliness, and three,
has the personality
of a child and lastly,
refuses to kill or hurt
others. And so we
need to take into consideration
our
game character profile in this case, Elizabeth's
profile to come up
with an A.I. that fits
this
personality, this type of character. And so, in a sense, Elizabeth creates certain emotions on our players. For example, Elizabeth makes you feel safe. Elizabeth, like a friend, will guide you to where you need to go, basically leading you to the next objective. Elizabeth gives you the feeling of comfort in not getting lost and or alone. On top of that, Elizabeth adds a level of relief and difficult tasks.
For example,
when fighting big bosses or large groups of enemies, she aids in
battle, whether through her
powers or giving general
advice, sort of like telling
you to aim for the head if that is where the opponent's weakness is. On top of that, Elizabeth gives you the feeling of companionship. Elizabeth gives you the feeling that you are not alone.
Elizabeth makes
random comments about the environment
you are in, helps you feel that there is truly
another person working beside you. In this case, the Elizabeth II makes you feel that there is another person tagging along with you in your adventures.
And as a matter
of fact, in battle, Elizabeth will find items
to help you. And this is
because Elizabeth's character profile is to do no harm to others to avoid confrontation and refusing
to hurt other people. And as a matter of
fact, her eye matches her personality of doing no harm to others.
For example, Elizabeth
seeks to collect items and health potions for you in the middle of combat. She helps you by creating
rain, but it is
you that has to hurl electricity to fry your opponents to do harm to your
opponents.
Elizabeth can create bowls of fiery energy, but it is you, the
player, that has to hurl this to defeat
your opponent. That is not
Elizabeth hurting others.
It is you. And as a matter
of fact, her
decisions are geared towards her personality traits such as kind and caring. And it adds some sort of dimension to Elizabeth as well. For example, are you really a kind and caring person if you aid your friend
in hurting others? And lastly,
Elizabeth makes you feel
optimistic.
Her decisions, again, are generally geared towards giving you the player hope on. All your tests, whether by how Elizabeth handles enemies or how Elizabeth talks to you as a person and this leads into
tension and relief, Elizabeth
gives you a sense of relief when Elizabeth is there.
And in this case,
Elizabeth adds a sense of tension when a way to summarize when Elizabeth is with you, tension is released, when Elizabeth is away from you. Tension has been added. And as a matter of fact, tension and relief are good emotions to put
on the plane. It makes
them feel like they are inside a roller coaster of emotions with all the ups
and downs that the player
is experiencing throughout the game's journey.
Speaking of tension
and release, let's take a look at alien
isolation in the game.
Alien isolation players must avoid the unstoppable hunting force of the alien at all costs. And this gives us a lot of questions for a game with an A.I. as complex as the alien in the game, alien isolation.
The main question
is what is an end of parasitic alien? More specifically, we need to ask ourselves other questions. For example, what is a hunter?
What is a
primitive life form? And on top of that, how do you go about programming the answer to these questions? Regardless, the age of the alien gives the player the sense of urgency to escape their current predicament because the alien is both primitive and reactionary. And on top of that has the characteristics of a violent, aggressive and sneaky force whose behavior revolves around the desire to kill species that are weaker than themselves.
And sadly for you, you are
human and the alien
is hunting you throughout the game.
Let's take a look on the
emotions that the alien puts upon the
player, and we can
separate it into two different categories when the xenomorph is present and when the xenomorph is not present
and away, when the xenomorph
is
present, it adds
tension and nervousness to the player with its presence. It makes you doubtful when you commit to a plan or execute a plan you have.
As a matter of fact,
the
alien forces
nervousness on the player when the player wants to execute a plan or go to an objective, as the player never truly knows where the alien is. And the alien A.I. by
design, does not
know where you
are, but it knows the
area you are in. As a
matter of fact, if you want to
make your room a dangerous environment, just add the xenomorph. So in a sense, with the xenomorph present, tension is added to the player. A sense of urgency is added onto the player.
Now let's
take a different
look when the xenomorph
is not present. As a matter of fact, when the xenomorph is away, you feel the release of the tension that was previously added to you by the alien, which is called release. As a matter of fact, you are safe when you are
alone, which is a
completely different
take from the
Bizshark game where Elizabeth makes you feel safe when you are not alone, when Elizabeth
is with you back to the
alien isolation
game. When you are
alone, you can afford to make more noises. You can afford to make more errors in your decision
making because there is
no alien to punish you for bad
decisions. As a matter of fact,
I would argue that this
game makes you
feel safe when you are alone. And so if you want to make an environment, a safe environment and alien
isolation, just
remove the xenomorph. On top of that, the
alien, when added to
the environment, adds
pressure. For example,
in everyday life, in a game like alien isolation, where dark
is scary, you feel safe
when there is
light. However, when the xenomorph
is present, when the alien is present in your
environment, the
light in the environment that once made you feel
safe now works
against you. And that is because the xenomorph reacts to both noise and sight.
If it sees
you, it runs towards you. It cannot see you when you are hiding in the shadows. But if you are out with the
light, well,
let's just hope that the alien does not turn around. In conclusion, we went over strong and weak A.I., deterministic and non-deterministic A.I. in video games as well. Basically, two types of Ehsaan video games exists, one which is simple and predictable, deterministic A.I. and the other non-deterministic A.I. is unpredictable
and both are highly used
in games. One is not better than the other. As a matter of fact, I would say that deterministic A.I. is simple to create, simple to debug, simple to test upon, whereas non-deterministic A.I. is harder to implement test
in the book on top of that.
We talked about the goals of using A.I. and there are three goals.
The first is
using A.I. as an obstacle or competitor to the player. The second
is to
force certain game making decisions upon the
player. And lastly,
we can use A.I. to give player emotions, to force emotions onto the player. And in the case of both Bizshark three
and alien
isolation, we can use tension and the release of
tension to sort of create
a roller coaster of emotions onto our player. Anyway, thank you so much for joining me. Thank you for clicking the subscribe button and thank you for clicking the like button. I look forward to seeing you in the next episode. Have an amazing day.
