About Perimeter
Q1: What graphic tools were used during development
of Perimeter?
A1: The main graphic tools for Perimeter we
developed ourselves. The most important of them
is landscape
editor Surmap, which is in its third revision
right now. We use it to create all these unique
worlds. The editor has a wide range of tools
for creative real-time landscape modification
as well as editing of procedural volumetric
geo-textures. We also have developed our own
powerful editor
for particle-effects which allows us to create
various eye-candy effects.
Q2: What hardware 3D features does Perimeter
use?
A2: We use float textures, compressed textures,
T&L, pixel shaders 1.1, 1.4, 2.0 as well
as vertex shaders 1.1 and 2.0.
Q3: Can you give us some examples of how Perimeter
takes advantage of those features?
A3: In the game you’ll see very highly
detailed landscapes changing in the real-time
as well
as huge number of bumpmapped 3D objects. Everything
in the scene casts soft realistic shadows; objects
are reflected in each other while all of it is
surrounded by complex particle effects. The visual
details in the game are so high that the player
will forget about annoying polygons. Visually
Perimeter is much better than any other released
3D game.
Q4: What 3D features do you like the most and
why?
A4: At this point of time Occlusion Query is
a quite important feature for determining visibility
of point in 3D space, which helps with proper
rendering of flares. Also, pixel shaders become
more and more relevant for visual effects because
it makes no sense to just increase polygon
counts
any more (in Perimeter average polygon is 2
to 6 pixel each at 1024x768 resolution).
Q5: If you could ask for only one new 3D feature,
what would that be? How would you use that
in your games?
A5: We would ask for ability to read depth
buffer values into a pixel shader. This would
enable
creation of better volumetric effects such
as fog, clouds and water.
Q6: If you could sum up your view on ATI's future
in 3D market, what would that be?
A6: Right now ATI is holding a very strong
position. In a high end market price-to-performance
ratio
of Radeon cards is better than of Geforce FX.
And in the near future we don’t foresee
any problems for ATI because their Radeon 9700-9800
family has much more powerful pixel shader 2.0
engine than the equivalent Geforce FX cards.
In our opinion ATI confidently moves forward
towards undisputable leadership in the graphics
market. There’re number of signs confirming
that – leadership in terms of getting technologies
to market and the fact of Microsoft’s partnership
with ATI for developing Xbox 2. It all points
to a change of the leader in the graphics market.
Q7: Give us a brief history of K-D LAB.
A7: K-D LAB was established in 1995 when a
group of student-enthusiasts founded in 1992
decided
to start developing games professionally. In
1996 our game called Brainy Balls has won the
Second Price at the ENIX Internet Entertainment,
a worldwide contest of entertainment software
held in Japan. In early 1997 K-D LAB released
a game called BIPROLEX+ which enjoyed widespread
popularity on the Internet. The year 1998 was
marked by the release of our first big action/adventure
project called Vangers: One For The Road in
US, UK, Germany and other countries. In Russia
game
Vagers gained a lot of followers and received
a status of a cult. In the fall of 2001 we
finished arcade racing game called Spanking Runners
which
was published by our strategic business partner
1C. A year before that Spanking Runners was
one of the IGF finalists at GDC 2000.
In the summer of 2000 K-D LAB and 1C signed
a contract for the development of a strategic
project called Perimeter. The game is built upon
a fresh territorial gaming concept and specially
created for that terra-forming technology.
K-D LAB is located in Kaliningrad, the west-most
city in the Russian Federation, located on the
shores of the Baltic Sea.
Q8: What do you think is K-D LAB's 1 key advantage
above other developers?
A8: Our strongest advantage is unique gaming
concepts that we’re not afraid to bring
to the market. Because we’re a small independent
developer we’re able to maintain a very
creative atmosphere within our team, which of
course contributes
to the development of interesting gaming ideas
and unconventional solutions as well as overall
quality improvement of our games. Besides, a
small team size allows us to maintain flexibility
in terms of company structure and our project
to reflect the present situation. In the future
we hope to stay a small flexible company that
can be responsive to the market changes and daring
when it comes to redefining ourselves or our
games.
Q9: Tell us a bit about the people who developed
Perimeter.
A9: We’re situated in a relatively small
city and we’re the only company right now
that is professionally working on the game development
around here. Currently the team working on the
Perimeter project has 18 members, majority of
which have experience of working on completed
K-D LAB projects. Almost all programmers have
degrees in physics and/or mathematics. The core
of the team has worked together for 10 years.
Andrew Kouzmine is the inventor of the game idea
and a game producer. In 1998 the puzzle called
MOBL based on his original concept received Grand
Prix award at ENIX Game Software Contest II (money
prize of $200,000) as well as title of best “Artificial
Life Simulator”. The game design is by
Yulia Chapochnikova. It’s not that often
that you see a girl responsible for such a critical
part of the project. Also we started expansion
by attracting game developers from other countries
(mainly from the Ukraine, of course) to work
for us.
Q10: How do you go about creating a game as
a team?
A10: It’s hard to imagine how we would
do it without a good team. The times when games
were developed by lone programmers are gone.
Today we have a clear split of the responsibility
for different components of the project between
10 to 20 developers. All team members are in
constant communication with each other. To make
our jobs more efficient we use our own software
to track task delivery. Every work day begins
with overall discussion what has been done yesterday
and what needs to be done today.
Q11: How long did the development take for the
Perimeter?
A11: We started working on a prototype in the
summer of 2000 and spend just over half a year
for prototyping.
Currently we’re roughly 3 years into development,
which I think is pretty normal for development
of a new game from “ground zero”,
considering that it’s based on a new and
untested gaming concept that required development
of unconventional technologies. Besides, this
is our first RTS project, so at the same time
we have to correct numerous mistakes and learn
from them.
Q12: Tell us about Perimeter and how it came
about.
A12: Perimeter is a real time strategy game
in which standard RTS concepts are combined with
new abilities of conducting a territorial warfare
by means of landscape changes. The idea has
been around for quite a while, since the development
of our first big project Vangers. For that
game
we created a map editor called Surmap that
allowed us in real-time to manipulate complex
gaming
landscape through adding volumes, leveling
areas and other intricate operations. It was
then,
when the idea for a new game came about. We
wanted to make a game where the world would seize
to
be a static background and instead would become
one of the important game-play elements, like
a piece of clay that player has to change significantly
to reach his goals. This is how a new concept
based on development of free-shape areas surrounded
by impenetrable defense perimeter was born.
Q13: What new ideas / features did you bring
to Perimeter?
A13: First of all we implemented ability to
play on very detailed landscapes that could be
(and
should be) altered during the game. The player
levels the hills, fills in the ravines and
oversees different surface deformations and so
on. All
of this is happening at the very high level
of details, with realistic shadows and volumetric
textures. This way the world instead of being
static background takes on an active role in
the game, accumulating even the slightest transformations.
Comparing the game levels before and after
a
good fight tells you the story of the battle:
the bigger part of the natural landscape has
been leveled; everywhere you can see traces
of work by terra-forming machines, of military
clashes,
of underground monsters and etc.
The second noticeable feature is a force field
that player can create for protecting perimeter
of his base. Using accumulated energy, the cellular
system of generators can create wavy force field
dome over his base.
We also have added some interesting aspects
of unit control. In particular the player can
only produce 3 basic types of units, while all
other military types are created through nano-transformation.
This considerably affects the tactics since the
player can rapidly adapt his military power to
particular situation. And the list goes on…
Q14: What were the best and worst moments during
that time?
A14: There was one very unpleasant moment that
almost ruined development of Perimeter. In my
opinion it’s very enlightening. Half way
into the engine development, when we’ve
invested a lot of time and effort, we encountered
very low performance when displaying our detailed
landscapes. The FPS was dropping even when we
were just scrolling the map. There was a threat
that we essentially couldn’t make it work
on existing hardware at the desired level of
detail. The programmers made excuses that there
was no equivalent of this technology anywhere
in the world, and that it’s essentially
impossible to implement. We had to face the problem
and decide if it at all makes any sense to continue
working on the project. However Andrew Kouzmine
was a firm believer that it’s possible
to realize our concept. We made a decision to
use external experts to analyze the code. In
a short period of time they examined the code
and identified poorly designed areas. As a result,
the performance increased a couple of times,
the problem was solved and obviously the team
went through some staff changes. The point is,
there should always be somebody who is a strong
leader and who isn’t afraid to take chances
and experiment.
One of the best moments that we would like to
point out was the first test of Perimeter. In
the first couple of months of development we
were able to make a prototype and show it to
people who had absolutely no idea about the game.
We have received a tremendous amount of great
feedback and we were also assured that the game
has a chance to live.
However, the most exciting time of the development
(which is probably relevant to any game) was
when all the separate pieces that were for a
long time developed by different people, came
together and all of it became a real game. For
example, when we made multiplayer version we
couldn’t tear away our programmers from
their “child” because of how exciting
the game play turned out to be.
Q15: How do you decide which graphic features
to support in a game given the development
time of your titles?
A15: Our technical experts and managers are
always keeping track of innovations that all
the time
broaden technological horizons. Most importantly
for us it’s the development of Direct3D® since
the engine is using it. After that, through special
tests we investigate how well each novel technique
looks visually and how well it fits the engine.
The amount of work and time it takes to develop
the game itself, including the design of the
plot, game-play and the content is always greater
than the work on technology development. Because
of that, game design is more of the gauging factor
than technology when it comes to setting deadlines.
Of course, we won’t delay delivery time
just to add some new graphic capabilities, but
if it’s apparent that development might
take some time regardless, we could make decision
to invest some extra resources into adding more
contemporary look to the game.
Another factor that we take into account when
deciding on features is performance. We conduct
a fair amount of experiments with different technologies
and select the proper solution according to the
performance we get on the modern PC’s.
Q16: How closely have you been working with ATI
and how do you see that continuing going forward?
A16: We’re very happy with our cooperation
with ATI and I say it absolutely sincerely. Rarely
do you come across such a high degree of quality
and efficiency of developer support. I would
like to note the work of Guennadi Riguer from
ATI, who we first met back in 2000 at GDC and
who we’ve been in a constant contact with
since then. If there’re any issues, we
at any moment can count on experienced help in
the shortest period of time.
Second, we don’t see any problems with
hardware and ATI regularly supplies us with their
latest graphic cards. Presently almost everybody
at our office works exclusively on ATI Radeon
9700-9800. We’re very happy about this.
Third, we’re very glad that ATI included
Perimeter in a list of featured game titles.
This very positively reflects on the game image
and speaks of it as technologically advanced
and high quality project.
Actually, the choice of ATI as our partner isn’t
a coincidence. At the very beginning of the development
when we came across the technological hurdles
we found out that the prototype performed considerably
better on cards from ATI. Back then we realized
that ATI graphics solutions are much more promising
for Perimeter.
Q17: What is next for K-D LAB?
A17: We prefer to talk publicly only about
the projects that have been officially announced,
but we’ll make a small exception. Besides
Perimeter, currently K-D LAB is also developing
adventure/quest game based on our own tools and
development methodology called Automated Game
Building. However, right now our heads are mostly
preoccupied with putting the final touches on
Perimeter.
In terms of the new projects, we have a huge
number of ready to use concepts that should be
enough for many generations of developers. Most
importantly we’re planning to continue
developments in RTS genre as it appeals to us
the most. On the broader scale, in our projects
we try to combine new technologies with fresh
interesting gaming concepts and well developed
plot. On this foundation we plan to build our
future titles as well.
Screenshots
 |
The Base under defensive shield
is attacked by the Filth
(brute creatures of the world). |
| |
|
 |
The Base is under attack. Core-generator
blows-up. |
| |
|
 |
Rocket units attack Filth creature. |
| |
|
 |
One of the Frames (large city-mothership)
with a lately implemented bump-mapping technology. |
| |
|
|
