Introduction

For my self-devised essay I have
chosen to discuss the topic of animation in games and more specifically how
procedural generated animation has impacted the way we go about designing a
game and how procedural animations have changed the way we go about animating,
most specifically animating characters; but also the advantages and drawbacks
of using this in the building of the game and how animators go about
implementing them in to the world. Most importantly it will go about explaining
the processes that go in to procedural animation and what makes it what it is. The
question in which I will be working around is: Where is procedural animation
taking us?

What is procedural animation?

Procedural animations can be used
for a range of different things, these can include animations such as particle
systems; the idea of procedural animations is that movements can be produced procedurally
which realise heavily on simulating physics.

”The
basic idea is that the moments of a character can be
generated procedurally. One of the most common technique to
generate animations procedurally, relies on simulating physics.”

Examples of these can include
things such as smoke, fire and water which is used in almost every game that
comes out today as well as the past and as they’re a rather big part of games, you
want these aspects to look and feel as good as they possible can. When it comes
to water, you can obviously choose to animate it by hand and go through the
long process of that or choose to use a simulation system that uses fluid
dynamics to help make it look a lot more realistic as well as cutting down the
production time a great deal. I will discuss more on this further on in my
discussion about procedural animation.

”Procedural
animation is used to simulate particle systems (smoke, fire, water),
cloth and clothing, rigid body dynamics, and hair and fur dynamics, as well
as character animation.”

However, procedural animations
doesn’t only help simulate particle affects, it also helps with materials such
as cloth and clothing which is used throughout your game as well as the clothes
your character or characters could possibly be wearing as well as things that
can be seen around the world such as curtains in houses, table cloths, banners
etc. The list doesn’t end there though, there is also mechanics which help with
rigid body dynamics, and this means that the movement and interaction of
objects such as solid and inflexible objects is also easier. Fur and hair
dynamics that help bring wild life and certain characters to life with how the
hair and fur reacts in the game will add a whole other level of how you look
and see them. As well as the animations for the characters, which is in my
opinion is the largest one among the list as it helps a great deal with the
animations for said characters.

”Procedural
animation is the latest and most immersive type of game physics. There are
no predetermined animations here. Instead, all of the characters and much of
the environment is continually responsive to in-game physics. That applies to
death animations, of course, but it also makes every other aspect of the game
more convincing, too.”

A great game that uses the process of
procedural animation is Overgrowth which focuses on only a couple key frames
per animation as they use a lot of in engine properties and components to help ”animate”
it. For example, both the walk and run animation only comprises of two key
frames each, the pass pose and the reach pose; how they made it look better was
rather simple as they were changing things such as the movement with the help
of a stride wheel, both walk and run have different sized wheels as they are
going at different speeds, this means that the character will never float and
means that his steps are always in sync with the ground. As well as this, in a
normal run animation you will have to take in to account the bobbing of the
head and the bounce of the character, with procedural animation and with
Overgrowth, as there were only two frames, this was added within the game and the
results can be just as good.

As most procedural generated
animation only focuses on a few key frames and not going very In-depth or
spending a lot of time hand animating in software such as Maya or Blender, you can
make a lot of changes to how they look and faze in to one and other and this
can be done rather swiftly.

Ragdoll Physics

For a lot of games when a character
dies it will go in to a detailed death animation which will have taken the
animator quite some time to make and somewhat realistic, however now days, with
the help of procedural animation introduces the idea of ragdoll physics which a
lot of game developers choose to use in their games when it comes to death. So
the idea behind ragdoll physics and why it is a good tool to use is because it
helps simulate how it would be when a real person or creature would react when
falling as well as interact with their surroundings; it helps add a lot more
realism to the world, especially when you or another character dies/plummets to
their death. In an example, for a first/third person shooter, when you shoot
your gun and kill an enemy depending on where you shot them would then result
in a different effect on how they will move. If you were to shoot them straight
in the shoulder, they would be thrown sideways in the direction.

In addition to this, as I said
earlier this saves a lot of time and money as you don’t have to add expenditure
on a death animation but also because for a more realistic look and with the
same effects, without these tools would be very difficult and near impossible,
not to mention time consuming. Seen from many games, especially older ones,
once you die you would end up finding your characters dead body would flop
around on the floor, it may even disappear or glitch through a lot of things.
Death was the same and often very unrealistic in a lot of games and could even distract
you from the game and even take out some of the immersion you get from it.

These ragdoll physics are achieved
by connecting each limb to certain joints that help to copy the movements that
a real body would have. This is helped by the use of rigid bodies and joint
constraints. The use of ragdolls can be accessed pretty easily now days in the
development of any game as most engines with some sort of ragdoll tool. For
example, unity comes with one called the Ragdoll Wizard which will turn a
humanoid model into a ragdoll very quickly.

”By
replicating the processes and constraints that govern the human body,
it is possible to approximate realistic behaviours.”

However, like everything else there
are also disadvantages, ragdoll physics included. The main issues that they
come with is that they can be very unpredictable and as seen from many games
they can end up doing a lot of weird, silly and amusing things; often being
very unrealistic in the process.  Despite
this, a lot of game developers still trust this method and continue to use and
implement these as it adds more realism and is often worth the mishaps that can
come with it.

The use of rigid bodies in
procedural animation

Rigid body dynamics is the movement
and interaction of different objects within the game; these will include solid
and inflexible objects. It can also be explained that a rigid body is a
polygonal or NURBS surface that is made to be an unyielding shape. During
animation, rigid bodies allow them to collide rather than pass through each
other. When it comes to using rigid bodies, most 3D animation and modelling
software will have two kinds, active and passive.

”In
any game, only certain physical effects are of interest. Rigid body dynamics –
the movement and interaction of solid, inflexible objects – is by far the most
popular kind of effect simulated in games.”

For active rigid bodies react to
dynamics- fields, collisions and springs, not to keyed in frames. However, when
it comes to passive rigid bodies, these can have the active rigid bodies
collide with it. This will mean that dynamics have no effect on it, but it will
let you key the translate and rotate attributes during animation. Yet, a rigid
body will not allow you to animate ad scale change or other deformations that
occur to said body.

”For
instance, to bounce a ball on a floor, you would make the ball an active rigid
body because it needs to fall with gravity and rebound after colliding with the
floor. You would make the floor a passive rigid body so it doesn’t careen away
from the ball when the ball bounces off it.”

When you implement rigid body
constraints, this will result in the restriction of motion of the rigid body. Usefully,
when animating in Maya when you constrain any object in your scene, the
software automatically makes that object a rigid body. Saving you the hassle of
doing so.

Furthermore, returning back to the
topic of ragdolls, as with most ragdolls, they don’t have any mobility or very
lack of. You can choose to connect them with joints however which will allow
them to walk or jump, but falling is still the outcome of both of these.

A popular example of this would be the
game Grow Home- A BUD’s Life, the main Character in which you play as does not
actually have any handmade animations, when its legs move its feet are put in
position via lines of code; as well as its limbs being in constant ragdoll
constraint. This gives for a lot of unique movement throughout the game, this
system works by the character trying to at all costs, keep his body above his
feet, allowing him to walk and jump across the map. As BUD’s body can’t freely
move around as it is connected via invisible springs, this means that when it
is attempting to move around, the top half of its body will inevitably wobble
around trying to gain balance and stay up. Each step the character takes is
planned by the system and attempts to successfully move it forward while still
keeping its body over its feet.

There have also been quite a few
other games that have attempted to use ragdoll physics on moving characters, another
one of these examples would be Drunk Fu: Wasted Masters. This game has you
flailing your character around the many maps trying to knock out your
opponents. Just by looking at the game, you can tell instantly that they use
the ragdoll physics with rigid bodies and it can make for a lot of funny
moments while playing with your friends. Unlike BUD, Drunk FU will almost
constantly have your characters on the ground struggling to find their balance,
but for this type of game it works and just adds to the fun of. This however doesn’t
work for most games as you will probably want your characters standing up straight
and walking in straight lines, not falling over every minute; that’s exactly
where rigid bodies help out.

”The
motion of rigid bodies can be modelled using Newtonian mechanics, which is
founded upon Isaac Newton’s famous Three Laws of Motion:

·        
Inertia:
If no force is applied on an object, its velocity (speed and direction of
motion) shall not change.

·        
Force,
Mass, and Acceleration: The force acting on an object is equal to the mass of
the object multiplied by its acceleration (rate of change of velocity). This is
given by the formula F = ma.

·        
Action and
Reaction: “For every action there is an equal and opposite reaction.” In other
words, whenever one body exerts a force on another, the second body exerts a
force of the same magnitude and opposite direction on the first.”

 

Inverse
kinematics

For procedural
animation inverse kinematics is a very important step in the creation of what
makes procedural animation and is used frequently. The system can be used for
quite a range of different tasks, however it’s mostly known and used for
humanoid characters. An example of this is when a character reaches for an
object in the distance, inverse kinematics makes it look natural. The way they
accomplish this is by specifying the object/target that the hand is reaching
for and the inverse kinematics will work out the most natural way to move the
joints to reach it.

”While
Grow Home and Rain World let the physics determine how the joints would move
when subjected to gravity, inverse kinematics forces them into a desired
stance.”

When it comes to rigid bodies, it
is very simple as all you need to do is tell it where the hands and legs need
to be and the in-built physics engine will do the rest of the work for you,
however this simple way of doing it works with non-complicated characters, but
most often ends up looking un realistic. This happens because the rigid body
normally only takes in to account the gravity and mass, nothing more. Inverse
kinematics takes in to consideration a lot more giving it the more realistic
look also thanks to IK. Normally you will need to switch between both IK and FK
to achieve this goal, however with inverse kinematics, you do not need to
define a whole pose of an articulated body as IK does that for you.

”Inverse
Kinematics is the inverse function/algorithm of Forward Kinematics. The Forward
Kinematics function/algorithm takes a target position as the input, and
calculates the pose required for the end effector to reach the target
position.”

Inverse kinematics is used in many
games helping the characters interact with the world smoothly. The game Shadow
of the Colossus for the PS2 uses this system for one of its most main game mechanics.
For the animations not only used on the character, but also those colossus
monsters you see within the game, these also use the IK system. This is because
of their sheer size, the movement needed to make it look good will require the
foot to just stay positioned to the ground when walking, with the IK system, and
this allows this to become a reality.

Due to a lot of landscape changes
in Shadow of the Colossus and even other games where the landscape isn’t just
flat and has bumps, hills etc. When it comes to walking or anything involving
the foots placement on that ground, the angle of said foot needs to changed or adjusted
to make sure it doesn’t just clip though and can look like. With this, it is
also needed that you change the height, just moving the angle of the foot will
result in it looking not so nice. Because of IK, this helps with calculating
the angle and the direction of each bone so that it can correspond with the
changes in landscape/environment. In Shadow of the Colossus, this is shown not
just with the large colossus’ but also with the character and even your horse.

An example of this is when your
character(s) are on a certain piece of terrain that is slanted, higher than the
rest or un-even, you will see that because of the IK mechanics in place, the
legs and feet react to the change to make it more like what would really
happen.

A big piece of software that has
helped push the advances with procedural character animation has a lot to do
with inverse Kinematics (IK). FinalK which is a great plug in for unity has
managed to help a great deal when it comes to animating bipedal characters. It
includes a range of different features in which helps it become an amazing piece
of software which helps a great deal with animation. One of these is a flexible
full body IK solver that is used for real-time procedural animation modification
and allows for a variety of remarkable animation effects.

Another feature is something called
grounder, this helps with foot placement and alignment correction and isn’t limited
to just bipedal characters, it can also be used on creatures such as spiders, quadrupeds
and even bots. For feet placement, this makes sure that the feet are positioned
and aligned to the surface. It also makes sure that it keeps the look and feel
of the animations the best it can. For games that have wall running or
climbing, this is perfect as it can even be rotated to allow for characters to
walk on ceilings and walls. I can imagine a game such as Titanfall would have
used something very similar to FinalIK.

For most if not all shooter styled
games, it will require an aim system. FinalIK has a feature which can produce stable
and natural looking targeting for your character animations weapons. This will
make the animations look a lot better but also cut a lot of time in half for
when it comes to targeting within the game as all you need to do is attack the
component, set the bone reference and you’re done.

When it comes to turrets and
mechanical arm animations, Final IK has components which feature rotation
limits which can be used by itself or other software such as CCD, FABRIK or the
aim element within this to create complex constrained rigs. Meaning that you won’t
have to go about the long process of rigging up a complex rig yourself, this
will do the heavy lifting for you.

Conclusion

From what we can see throughout
this essay is that procedural animation has been changing the way we are and
should be going about animation in game development projects, especially for
indie developers who don’t have the time or money to cash out on certain
animations or an external animator; this can be especially appropriate for when
they don’t have someone in house that has the skills or time to learn how to
animate. With this we’ve discovered all you need is a few key frames to get
started and a lot of the changes and work can be done in your game engine,
especially with all of the features and content that comes with it. I believe
that procedural animation should be used a lot more often in game development
for a lot of these reasons. However because of this and how somewhat simple it
has made animation, some people may say that this is taking us away from the
artistry that comes with animating. I can both agree and disagree with this, as
I said earlier a developer may not have the time or money to focus on things
such as the artistry, however for bigger AAA companies, they have the time and
money to focus on such things. 

x

Hi!
I'm Katy!

Would you like to get a custom essay? How about receiving a customized one?

Check it out