Shigeru Miyamoto,
Keynote Speech, Game Developer's Conference,
San Jose, CA., March
18, 1999
Good Evening! Thank you very much for your
warm welcome. I am very happy to be here.
The video game industry has been through
many crises, yet, has continued on as a strong and growing business.
Thanks to that, I am still making games, and I get to speak to
you about them today. I owe a lot to a number of people who have
been involved in this business -- and also to the newcomers in
this industry who have grown up playing Nintendo games and continue
to purchase them today.
Making use of this opportunity, I'd like
to speak about my own experiences for the past 20 years. If what
I tell you today can help games to consistently be fun to play,
it would make me happy. In fact, I like making games so much
that I would do it for free, but don't tell Mr. Yamauchi that!
And now, I hope you will do me a big favor and allow me to speak
in Japanese. My friend, Bill Trinen, will translate for me. There
are three main topics I'd like to speak about today:
Looking back at the history of game designing.
And what I think game designing should be.
Ideas behind creation of Zelda64
My policy towards future creations
Japanese translation begins at this point:
Until the time of Donkey Kong, which was
the first game I directed, programming and hardware engineers
were responsible for game design. Those were the days when these
engineers were even composing the music and drawing the pictures
themselves (resulting in rather primitive, now classic, games).
When I, as a graphic designer first became involved in game design,
I used to boast to myself that I was one of the five best game
designers in the world, since there were few designers with artistic
experience in game design back then.
Donkey Kong and other games gave birth
to a new trend in which video games had an accompanying story
for the first time, and the work of game designers came to include
drawing the pictures and writing the story. This trend continued
for approximately 10 years, during which time many designers
joined the industry , and professional music composers also began
to take a role in game design.
Then, due particularly to the success of
Dragon Warrior and the Legend of Zelda in Japan, there emerged
a new trend in which scenario writers took leading roles in game
design. At this time I was inundated by many designers-to-be
hoping to get their scenarios turned into games, and we also
saw popular scenario writers teaming up with music composers
in the hopes of churning out a game for the sake of business.
And in recent years, as I predicted, advancements
in technology have once again brought programmers and engineers
to the forefront of game design, and we are now in age in which
we cannot accomplish anything new without these individuals.
Let us look back once again. In the days
since I have joined the industry, we have also seen a qualitative
change in the nature of game play. Early on, the objective of
arcade style games was to see how many quarters we could get
users to drop into the machines. It was with this goal in mind
that we created the Donkey Kong series, but at the same time
we encountered problems with new types of games that contradicted
this model. One example is baseball, in which you must play 9
innings to complete a game, but at a quarter an inning one game
is too expensive, while from the point of view of the arcade
operator, players could get more play time for a 25 cent inning
of baseball than they could on other games. Another example is
our Mario Brothers, in which the players' objective was to defeat
their opponent, but the better they got at the game, the shorter
their play time lasted. Players had a hard time accepting these
types of game concepts.
It was at this time that I became involved
in the development of the Nintendo Entertainment System, which
offered us an environment outside of the arcade for which we
could create games. We put higher priority on developing two,
hand-held controllers rather than a singe, ultra-functional controller
to open the door to games like baseball and Mario Bros. With
the NES, the business of selling game play time transformed completely,
and evolved into the business of simply selling game play.
Throughout these changes I have maintained
the same style of game design. Although I am not an engineer,
I have always included in my designs consideration for the technology
that will make those designs a reality. People have paid me a
lot of lip service, calling me a genius story teller or a talented
animator, and have gone so far as to suggest that I try my hand
at movies, since my style of game design is, in their words,
quite similar to making movies. But I feel that I am not a movie
maker, but rather that my strength lies in my pioneering spirit
to make use of technology to create the best, interactive commodities
possible, and use that interactivity to give users a game they
can enjoy and play comfortably.
I feel that I have been very lucky to be
a game designer since the dawn of the industry. I am not an engineer,
but I have had the opportunities to learn the principles of game
from scratch, over a long period of time. And because I am so
pioneering and trying to keep at the forefront, I have grown
accustomed to first creating the very tools necessary for game
creation. I have seen both the evolution of hardware, and the
advent of hit software titles, but I have seen that not all the
ideas for new hardware development come from engineers and hardware
professionals. And I have seen more than a few examples in which
the idea for a hit title have popped out a simple conversation
with an ordinary person.
Recently, I have encountered many instances
in which we hit a brick wall in game development, and never quite
make it to completion. I believe many of you have at one time
or another found yourself in the same boat, and may have even
had no choice but to release a game to market in an incomplete
or unsatisfactory condition, much to your own mortification.
I am sure that each case has its own unique
cause, but I know that when (Nintendo's) game designers and producers
make their plans without a sufficient grasp of the technology
and engineering necessary to make their game, they will often
fail. Also, we may be frustrated to find that a game we are developing
never really becomes fun to play no matter how hard we try to
improve it. Recently, when new technology and exquisite graphics
are regarded as the core of a game's element of fun, it becomes
difficult to evaluate a game before launch, because those fun
aspects of it can only be judged after everything is in order
and the game is nearly complete. So the answer to the question
of how many more months of tinkering will produce a truly enjoyable
game depends upon a variety of newly emerging technologies. Until
the technology is complete, game designers are unable to evaluate
how enjoyable the game will be, and because that level of enjoyment
is dependent upon the level of completion of that technology,
there is no guarantee that the final game will be enjoyable at
all.
In my understanding, game designers are
solely to blame in these situations. But, on the other hand,
since there is no concrete definition of what a game designers
work entails, I cannot say that it really is the fault of the
designers. As I mentioned earlier, in the history of game development,
there was a time when designers were engineers who were less
capable of composing sound and creating exquisite graphics. Then
we had designers who were painters, but could not understand
the technology behind the games. They didn't know what they could
and could not do. How were they to express their ideas so that
they could be understood by programmers and realized by the CPU.
Later, scenario writers took the lead, but they, too, had a low
level of understanding of the technology.
Fortunately, because I have been a part
of the industry since its dawning, I have at last come to the
conclusion that the role of a game designer is to design a complete
game system by first comprehending the technologies that will
enable and realize that system.
I believe that design itself is one of
the jobs of a game designer. How will the ideas I have in my
mind be reproduced by the computer? How can the power of the
CPU be best allotted in order to convey those ideas to the user?
Will the players always be able to find fun and enjoyment from
this game? How can I bring my own constructions and expression
of ideas together with the technology that creates that new level
of enjoyment within the hardware and budget limitations placed
upon me? This is what I mean by design.
I consider games to be entertainment commodities
and therefore place great importance on user reaction. I realize
that I am one of those users to be monitored when I play a new
game for the first time. It is important to design the whole
game creation process so that users' needs will be reflected
effectively and quickly at the time of completion, and this is
why game directors must be deeply involved in the design process.
Accordingly, in our company, all designers
must go through technical training. Graphic designers make games
by comprehending the ROM and RAM memory maps. Which specialists
are taking the led in design depends on the current trends, but
this technical training is a basic requisite for continued success
in game design.
Let me offer you some examples. Suppose
a director presented the following game specification for an
action game. "The enemy shall randomly search and react
to the character." To randomly search may sound like an
appropriate specification, but a programmer cannot program this
as it is, and in the latter stages of development, the director
won't know what to begin with when trying to bring the game close
to its original concept. But what if the specification presented
was "The enemy shall change course based on character movement
once every 30 game frames, and one in three times it will randomly
select to progress in." This should be more easily programmed
than the previous example, we will know where to begin modifying,
if necessary.
In a best case scenario, you may be teamed-up
with talented programmers, and the programmers may be able to
make the game on their own and have it turn out to be quite enjoyable.
In such cases, I may no longer be necessary, and programmers
can make the whole game on their own. Honestly, I have been very
fortunate in that I always seem to be joined by a group of excellent
programmers.
The next necessary element of design is
skillful management of the memory map and accurate estimation
of the processing speed. When we make games for consumer game
consoles, it is important to take into account the limited ability,
processing speed and transfer rate of the console. For users
and company management who do not have a technical grounding,
it is taken for granted that the virtual world exists in the
game. This is a matter of course. If we see a man running and
there is a hill in front of him, naturally, he will run up the
hill. If a car is bearing down on him, we can guess that it will
hit him. We assume that the cars wheels will turn when the car
is moving and stop turning when the car stops. For the users
and management, these are the laws of nature, but they don't
realize that we are the ones who have created this virtual world.
When problems arise in the end, the ask Can't you do this? Does
the processing speed have to be so slow? Why do I have to wait
so long at this point? But at that stage of development, such
areas can not be fixed. When we were making games for the NES,
sounds in the game consumed CPU power, so in early development
we would include dummy sounds so that we could estimate the processing
speed of the final product. There were days when we were playing
Super Mario Bros. to the music of Excite Bike, and Mario's jump
didn't have its characteristic "boing," but rather
the rev of a motorbike. This kind of consideration is necessary
for taking the interface into account and in order for us to
focus our attention on the areas of the game that we have placed
our creative priorities on.
Also, in many cases, we miscalculate the
time necessary to prepare parts of the game that are often taken
for granted. The less experience a designer has, the easier it
is to make this mistake. But the amount of programs and sequences
which are unique to any given game are actually quite limited,
and represent less than a third of the total. Taking the example
of an RPG, there are countless features that are taken for granted,
such as the title and name registration screens, the dialogue
system, item select windows, and so on. I call these parts "labor,"
and I am always trying to design games which feature as few of
these parts as possible. Unfortunately, many game blue prints
contain a number of these elements, and it is often work on these
"labor" elements exceeding our time estimations that
lead to the troubles of games not reaching completion.
With this in mind, we can reduce this Labor
by creating an entirely new genre of games that no one has ever
thought of before, right?
Though I have talked mostly about the technical
aspects of game design, I would now like to talk about something
on the opposite end of the spectrum. We must not forget the importance
of human ingenuity and creativity in game design. Naturally,
it is new and unique expression of ideas that gives birth to
new games. Recently, I am very sorry to see that the uniqueness
of many titles has been dependent upon new technology and specialty
development tools, while the personalities of the creators have
been diluted. For me, game creation is like expression through
music. When I am working as a director on a game, while I always
try to hit upon new plots, I place great importance on the tempo
of the game and the sound effects. I feel that those directors
who have been able to incorporate rhythm and emotional stimuli
in their games have been successful. When I am holding the controller
and setting the tempo, I feel that my own, personal game is in
the midst of creation. I have never created a game that has been
of a level that I could be satisfied with. Understanding the
technologies is the requisite if we want to fully realize our
expression. Game designers are apt to boast of the technical
aspects of their games, and I, too, have fallen into this trap.
Speaking of my own case, I tend to highlight new technologies
when I am less confident about the new ideas I am putting forward
in the game, and later, I always regret doing this. It is important
for us to remember that technology can inspire new ideas and
help us realize those ideas, but it should do so from the background.
Next, I'd like to discuss a game that I
recently worked on, which you may or may not have heard of.
We started with about four or five different
teams, each working on basic experiments related to game design.
Scenario and Planning - both very necessary.
The team discussed the position of this title in the whole series,
and included myself and several dedicated script writers.
Link's action and 3-D improvements of
items found previously in the series: This team included myself,
Mr. Yoshiaki Koizumi who has worked on player characters since
the days of Mario, and the head programmer. Mr. Koizumi is here
in the audience this evening. Where are you? If you happen to
see him at the conference, say hello. He may share some useful
information with you.
A variety of Camera experiments were conducted
by the same team that handled Link's action plus several designers
and programmers. We worked on incorporating new methods, such
as background virtual boxes which we did not use in Mario, as
well as fixed camera modes like we used in the castle.
Another team worked on bringing the items
that Link touches and uses that lay at the basis of the game,
into the 3D world.
Motion capture production and tests for
the creation of the converter used in the Zelda's animation,
which was done by an entirely different team.
We also formed new teams according to our
needs. There was the
Sound team: Ocarina play and 3-D sound
Special Effects
The flow of time in the background and
background culling appropriate for Zelda. What was unique here
was not just what was visible in the background, but progressing
with design and integration of terrain specific sound and movement,
camera moves, and enemy data and tools effective for manipulating
this data.
We design the entire game by organizing
these small teams and conducting tests to confirm memory size
and processing speed. Let us recall Hyrule Field in the Legend
of Zelda. The characters that appear in the field all share RAM.
So, following the scenario, we have the great bird, Gebola, and
the Skeleton enemies, and then once the bird has left the field,
the Marathon Man appears. When Link rides the horse as an adult,
the only enemy to appear is the ghost. So the scenario is written
to match these types of design specifications.
I would like to add that depending on the
type of game we may start development with just the character's
movement, and of course not all experiments will be adopted in
the final game.
These are the technical aspects of Zelda's
development. Let me now talk about the concepts behind the game.
Through our experience of making this interactive media called
video games, we have uncovered a number of methods to stimulate
a player's emotion. One way is through the use of cinematic sequences.
As you know, seeing with one's own eye can give a more pronounced
effect than a player's imagination. With that, let's take a look
at some of Zelda's cinema scenes.
We have seen several examples of these
cinema scenes, but we positioned these sequences as only one
part of development. In Zelda, there are over 1 hour and 30 minutes
of cinema scenes. But the staff involved in creating these scenes
was only 3 individuals for most of the development process, and
in the latter stages, only 6 or 7. The reason behind using such
a simple process, as I am sure you have all experienced in the
workshop, is that there is a total limit on team energy. There
is a limit to the work a team can do, and there is a limit to
my own energy. We opted not to use that limited time and energy
on pre-rendered images for use in cinema scenes, but rather on
tests on other inter-active elements and polishing up the game
to create a product that players can enjoy and play comfortably.
Actually, I changed the scenario just a few months before completion,
and although the staff was rather unhappy about taking apart
something they had already created, we were able to make the
fixes in a short amount of time, so that it did not cause any
problems time-wise. It is not because the N64 doesn't have access
to a CD-Rom that we incorporated real-time movie processing.
Quite the contrary, to the greatest extent possible we were able
to make use of truly cinematic methods with our camera work without
relying on the kind of data typically used to make cinema scenes.
Our president, Mr. Yamauchi says that each
product has its own mission. We set interactivity as the mission
for our products. And thanks to that, we were able to experience
development that no one had experienced before.
I have talked about several different topics
this evening, but now I would like to discuss the future of game
design.
Shall I begin making Zelda for our next
generation hardware. At this point, the answer to that question
is no. The reason for my saying this is that all of the elements
for which Zelda has received so much praise for had already been
incorporated into the game more than a year before completion,
when I felt the game was not fun to play. I think that a lot
of the reasons that Zelda has been so praised are not related
the N64's level of expression, the unique camera systems and
auto-jump system, nor the gorgeous cinema scenes and spectacular
boss fights. It is true that some other team may realize the
level of expression that we achieved with Zelda, but of course
it will not be the exact same as Zelda. With improved hardware,
I can imagine Zelda having more detailed graphics and a quicker
response time, but when it comes to increasing the degree of
fun, I cannot be certain of that at this time. This is something
that I feel we as designers must reconsider.
Also, I want to constantly make efforts
to create new ideas. I want to propose new game ideas without
worrying about the headaches of management, such as inflated
development costs. Video games have become far more popular than
in the past, but I feel that we have just been repeating the
same events again and again in this unique market. Even with
Zelda I did not feel that sense of freshness that I had with
the original Super Mario Bros. I want to make efforts to convey
the charm of video games to the general public that is currently
outside the reach of the industry in which we do business. This
is because I really want to feel the unique zest of the entertainment
industry, where one simple idea can create an unexpected social
phenomenon.
Let's take a look at a game I am working
on now called Talent Maker.
What you are seeing now is a newly-born
Mario Paint, a new game in which you can create your own characters
by utilizing the Game Boy Camera, and you can make those characters
dance, and what have you. The scenes you have seen, including
one in the introduction, where made very quickly by my staff
using this software. Nintendo will also make efforts to create
new types of commodities by combining the Game Boy, the Game
Boy Camera, the Rumble Pak and others with the N64. In Japan,
we have already launched the N64 title Pokemon Stadium, which
makes use of the Game Boy Pokemon games, and we are selling Pikachu
Genkidenchu, an N64 game which employs voice recognition technology.
Soon we will introduce a new system, in which the game boy can
be used as a controller for the N64.
We have expanded this industry and welcomed
new users with innovative products that continue to surprise
us. At a time when we were all developing Mario style scrolling
games, Tetris was born thanks to a team that tried to make a
new product with game & watch style software. When we were
stuck on talk of the spectacular 3D graphics of Mario 64 and
racing games, we a saw huge hit in the form of Tamagochi - a
tiny key chain boasting pictures made up of no more than 10 or
20 dots. At that time, I thought that Mario 64 had lost to Tamagochi.
I want game designers to be the designers
who make technology their tools, and use it to express their
own individuality, their own unique-ness and their own rhythm,
as well as the entertainers who make this world a more enjoyable
one. It is with this extravagant hope that I wish to end my speech.
My friends, let us design unique, fun software
with new appeal. Let us take on new challenges so that the world
of gaming is not left behind as a separate, closed off world.
And in the process, let's see if we can't make a little money.