Somewhere around late June, 1998, SIMIS and Mindscape/SSI released Team
Apache (TA), a unique approach to a military helicopter simulation. The
emphasis was on "team," and managing your team was a simulation in
itself, much like the role-playing "simulations" that are popular with
Japanese console gamers.
In TA each member of your team had a series of aptitudes and
weaknesses. The challenge was to pair pilots and gunners with
complimentary skills and personalities and use them to their maximum
potential. This involves being sure that you can get them enough sack
time, deal with the occasional crisis, and still be able to field
enough able men and machines to accomplish your mission and hopefully
win the war. The outcome of specific missions and the progress of the
campaign in general has effects on not only the morale issues, but can
have a great deal to do with the player's status as a commander.
Team management even extended to your resources and crew chief.
Bringing back damaged birds every time you flew made the crew chief
owly. And if you had seriously damaged choppers and no new supply, you
were cannabilizing units to keep others in the air.
Unlike role-playing games, though - there were no sheets of
statistics to tell you what each man was capable of or how they were
feeling. You had to infer that from their personal files, how they
responded to you, and how well they performed in battle. Balancing all
those needs was tricky but offered a unique perspective in combat
simulations.
Team Apache also offered a dynamic and persistent environment.
Destroyed vehicles from
previous missions littered the battlefield as the unit progressed
forward. Losing a pilot brought up a screen with the player/commander
writing home to his family. Certain incidents were "remembered" by
Operations personnel and came up to bite the player in the butt at a
later date. The suspension of disbelief was achieved by appealing to
the player from a personal side, as well as immersing him in a very
dynamic environment of cause and effect.
Apache units themselves were viewed and employed as a "big
stick", with a company of 6
aircraft possessing more than enough firepower to dispatch an armored
battalion. As a result, they were typically used in 5 or 6-ship
elements. However, many missions required the player to divide up the
company in smaller elements to "fight fires" that crop up unexpectedly.
The graphics engine was state of the art at the time, and was
impressive to say the least. Here are some notes from Neil Mouneimne's
review:
Some aspects of that detail merit special attention. For example,
when you hover just above the ground, rings of "dust" are blown across
the ground. From an external view, when the cockpit isn't reflecting
light back at you, you can see the Apache crew inside looking around
for targets. The rotors, chain gun, and optics suite are all nicely
animated.
There is a scattering of individual trees that you need to avoid,
although they are more for show than for tactics. Entire downtown
cities are modeled where you can play chicken between the skyscrapers,
stalk down city streets, or land on the roof of a building and take a
breather. (Fortunately, the collision detection is accurate enough that
you can fly right in between buildings, lampposts, houses, and trees -
as long as you don't actually hit anything)
The rocket motor glow in the back of a FFAR, Hellfire, or Stinger is a
nice touch. Enemy tracer fire glows brightly, streams well, and looks
positively dangerous. The infrared sights are also very impressive -
not only do you get the "image halo" as in Longbow 2, but you get a
very believable "washout" when an explosion blinds your sights
temporarily. The effect is very similar to looking a little too close
to a camera flash.
Does this mean the graphics are superior to Longbow 2? Yes and no. It's
really apples and oranges. Team Apache boasts a much smoother
framerate, has some nicer special effects, very impressive urban areas,
and has that "workstation" look to it…
With that history in mind, we contacted the Team Alligator crew
to discover where Team Alligator will build on the early product, and
where it will seek to diverge or move beyond the standard previously
set.
Q: Thanks for taking the time to connect with us. Team Apache was a
great game and appealed to a broad cross section of sim fans. The
greatest appeal was perhaps the graphics engine itself. Where will Team
Alligator move beyond the earlier engine? What will we see that will
tell us this is a new standard?
A: Thanks for the complements, though like any group of
creatives we are probably our own harshest critics when it comes to
previous products like Apache. In Team Alligator we've moved from our
old Icarus graphics engine to our new Daedalus engine, which has opened
up a whole new realm of graphics capability for us and allows us to do
things no helicopter sim has done before.
The problem we had with the old engine was one common to most
helicopter sims: that of adapting an engine originally designed for jet
sims to a helicopter game. One of the biggest flaws was that the
terrain resolution was simply too coarse. What works well for a jet
sim, flying at ten kilometres a minute does not work for helicopter
flying one kilometre in the same time.
The terrain resolution, the environment, is crucial in a helicopter
game. Helo players look for subtle wrinkles in the terrain to hide
behind; they look for trees and buildings to use as cover. They are
flying at less than treetop height, so you need to build a whole
convincing world at ground level for them to fight in. In a helicopter
sim you need a world about as detailed as that of a tank sim. But you
also need it to cover a much larger area to allow for the helicopter's
greater speed and range.
That's a demanding set of parameters. You want a large world,
but you want it detailed. And that's what our new Daedalus engine
delivers. It allows us to put a lot of polygons on the screen at any
one time. In turn this has permitted us to create a detailed terrain
system with very fine ground resolution, and let us populate the world
with convincing trees, buildings, telegraph poles, you name it. Special
effects too. We now have the tools to create an authentic working
environment for helicopters to fly and fight in.
Q: Tell us about environmental bump mapping? How is it employed and what
does it accomplish?
A: Environmental bump mapping is currently only supported by the Matrox
G400, and is an incredibly clever way of achieving greater detail in
game graphics without using extra polygons. There are a variety of
different bump mapping solutions Team Alligator will be using, and
environmental bump mapping is only one of them.
Bump-mapping works by moving parts of a texture (a texel, which is one
pixel in a texture) around in relation to lights. This simulates bumps
with respect to lighting.
Emboss bump-mapping draws the polygon twice, once with the texture as
normal and a second time it 'subtracts' the bump-map from the original,
which darkens the texture in 'bumpy' areas- to simulate shadow. In
addition, on this second 'pass' the bump-map texture is shifted a few
texels in relation to the direction of a light, which means the shadowy
areas can move as the light moves.
Environmental bump-mapping is performed by applying a texture on which
the light is drawn (e.g. a white specular highlight) to the normal
texture. This means the shiny highlights only brighten the bumpy areas
of texels facing the light source. This is used to simulate ripples on
water, heat hazes, etc.
Q: What APIs will be supported and what will be the maximum resolution?
A:We will be supporting only Direct 3D, and at the moment we are
undecided as to the maximum resolution but we have run up to 1280x1024
without major problems. However, Ka52 looks gorgeous at 640x480, so
there's no problem for players with lower end machines.
Q: How configurable will this engine be in terms of detail level and features?
A:
The largest challenge for the team making Ka52 Team Alligator is one of
technology scaling. We are making a product here which has to excel on
both a P200 with a 4Mb 3D card as well as on a Pentium 3 600 or so,
with a G400 or similar top end card.
As such we are designing the game with a whole raft of options and
features which can be turned on or off depending on the power of the
machine running the sim. Additionally, the graphics engine dynamically
calculates the number of polygons to draw each frame based on the
current performance of the PC Alligator is running on. This means that
the poly-count will rise and fall to keep the frame rate at an
acceptable level.
Q: The Ka52 helicopter that I saw at E3 was graphically stunning,
yet I was told it wasn't finished. What difference will I see next time?
A: You're too kind. The list of additions to the game between the E3
demo and the final product is almost endless. We are still pre-Alpha.
The version you saw was the bare 'vanilla-flavoured' graphics engine
and terrain system with very few special effects (no, honest!); we've
hardly begun to add in all the neat lighting effects, dust 'brown out'
effects, battlefield smoke and other goodies. You really ain't seen
nothin' yet!
Q: How much detail will go into object animation? Will we see
turrets moving, guns elevating, rocket flares, tracer fire, men popping
out of hatches?
A: The answer to all of these is 'yes'. To which you can add animated
infantrymen, moving cockpit crew and sensor turrets on the helicopters,
aircraft undercarriage action and some pretty cool explosion
animations. Our 3D cockpit alone has dozens of animations: levers;
switches; control pedals, cyclic and collective; and a fully-animated
weapons officer in the right-hand seat!
Q: The "team" in Team Apache was the heart of the sim. How does Team Alligator compare?
A: We have quite a similar mechanism for Team Alligator, though we have
rationalised many aspects of the team management. We've also expanded
it; whereas in Team Apache the player only managed a team in the
campaign, in Alligator you can create a 'crew file' and fly that team
in the stand-alone combat missions too. The aircrew's improvement and
advancement is much more visible than before. If you recruit a guy for
your unit and fly him for long enough without dying, you will see him
advance all the way up the ranks from a Mladshiy Leytenant (2nd
Lieutenant) up to Polkovnik (Colonel).
Q: The "team" aspect, while a unique component of Team Apache,
was also confusing to many players. Is there a way to compensate for
this in gameplay for less interested weekend sim pilots?
A: Yup. As in Team Apache, there will be the facility to simply ignore
the team management and fly missions or campaigns with a default
aircrew.
Q: For those who took the time to learn the ropes, the "team"
component provided a new level of immersion. I remember the first time
I noticed that a particular team mate was NOT into formation flying
while another had a habit of shooting at anything that moved. I paid
more attention to my team the next time and had much better results. I
felt like an actual commander! How has this aspect of the game been
received by most of Team Apaches' fans?
A: A lot of people liked this and really got into the whole Company
Commander deal. (Though there were a few for whom the crew voices and
dialogues grated.) Some players were even picking up some of the more
subtle features we had built into the mechanism, such as the
compatibility between the various aircrew. That feature survives, by
the way.
There will be some pilot/weapons officer combinations where the guys
just click, and some where they don't work well together. It'll be up
to players to find these out, though some time in the future we might
publish a 'cheat sheet' giving away the combinations.
Oh yeah, and we're getting the voices right this time. We have a number
of Russian speakers and a Russian intelligence expert helping us get
the dialogue authentic. You will even be able to fly the missions in
Russian!
Q: Let's talk about weapon systems and targeting. I remember my first
flights in Team Apache. I was frustrated because it was much more
difficult to retain a target lock than in Longbow II. Later I spoke to
a military chopper pilot who told me that the modelling was more
realistic in Team Apache. Will we see similar modelling in Team
Alligator?
A: One of our eternal frustrations with other helicopter sims has been
the poor or perfunctory battlefield modelling we keep on encountering.
They have targets that are always easy to see and track; weapons with
phenomenal accuracy and kill rates; almost no attempt to model passive
or active target defences at all. They create worlds in which
everything works perfectly and in which enemies line themselves up for
death like ducks in a shooting gallery.
We have infantrymen in Team Alligator. Ever seen an infantryman in the
field? With very little effort he can make himself invisible at 5
metres. So how can anyone, in a noisy rattling helicopter, expect to
see him at 5 kilometres, or 500 metres, or even 50? It doesn't happen.
In real life, finding targets, even something as large as a tank, is
*hard*. Tracking them for a shot without mechanical aids is *hard*.
It's why combat aviators get paid the big bucks.
Now, some of that difficulty we simply can't translate into
Team Alligator; PC screen resolutions are just not high enough to let
the player search for targets by eye. But even with the mechanical aids
we provide, such as the 'silicon weapons officer', the radar and the
Shkval tracker, you will find it tricky to find these guys and retain a
firing solution on them when you do. And we apologise if that causes
some petty frustrations, but that's life! If you want realism you are
going to have to deal with it.
By the way, our bad guys won't take things lying down. Not only
will they shoot back at you, but they will lay smokescreens that will
make your laser missiles worthless, or they'll have radar-directed
devices like Arena for shooting down missiles. They'll have reactive
armour on the glacis plate that stops your HEAT warhead dead. And they
won't just rollover and die because you say so.
Q: Even spotting targets was very different in Team Apache. If I
flew over the jungle at 100 knots I missed everything, but when I
slowed down I was more vulnerable but picked up many targets. These
constant tradeoffs are challenging, but very realistic. Is this still
the philosophy in Team Alligator?
A: Everything's a tradeoff isn't it? Speed for life; the vulnerability
of a hover for the ability to search and track. A good combat aviator
knows there's a time and a place for each of these. There's a lot of
this sort of thinking behind Team Alligator and we hope players will
pick up on it.
Our job as modellers and designers is not to make the player's job
easy, it's to make it *hard*. We want the player to use correct tactics
and consider the tradeoffs they need to make for each tactic; to make
them think *ahead* of time as to what's going to happen at that next
ridge or the next tree line.
As in Team Apache, we've put the emphasis in the game on tactical
difficulty. We want the players to *fight*, and fight smart; we didn't
want Alligator to be a dry exercise in systems management, which so
many sims seem to be moving toward these days.
Q: Team Apache introduced weather effects in a big way,
including some incredible storms with lightning, fog and rain. How will
Team Alligator follow on these advances?
A: With better effects, more convincing-looking weather, and modelling
that impacts the battlefield, such as sensors. I can't really say much
more that that. Wait until you've seen the demo, our weather effects
speak for themselves.
Q: Since we are flying a Russian helicopter we are going to see some new weapons. What are they?
A:
The Alligator carries weapons that are fairly familiar to helo sim
fans: the cannon; the laser-guided anti-tank missile; the rocket pod;
the lightweight infrared air-to-air missile. But one of the nice things
about this bird is that, unlike the Apache, it also carries some 'big
boys' toys'. These include the Kh-25ML laser-guided missile, capable of
taking on bridges and bunkers; the 122 mm demolition rocket; and the
500 kilogram bomb.
Q: Russian doctrine for helicopter combat must vary somewhat from
NATO doctrine. Do the differences impact gameplay in Team Alligator?
A: They do, but in subtle ways. The Russians tend to mix their
formations more than NATO forces do. They lack dedicated scout
helicopters, so tend to use gunships in this role. They are also more
likely to mass forces; they still regard the helicopter very much as a
'flying tank'.
However, the Ka-52 represents a new direction for the
Russians, and they don't appear to have completely formulated a new
doctrine to accommodate it. It is referred to as an 'intellectual
support helicopter', and it's clear the aircraft is designed to operate
in support of other aircraft, providing command as well as firepower
support.
Q: Sound modelling in Team Apache was one area that could be improved.
What sound APIs will be supported for Team Alligator and how will this
impact game play?
A: Our biggest problem with Team Apache was that we simply had *too
much* sound in it! Only the fastest machines could cope with the sheer
quantity available, and that in turn affected quality. We will be using
all the features of the latest Direct Sound, including the
environmental filters where appropriate (eg echoes in canyons, muffled
sounds in snow and rain, etc).
Q: We have recently seen some excellent flight modelling in
Apache Havoc. Flight modelling was a sore point for the more serious
sim pilots with Team Apache. Can you compare the characteristics
modelled in Team Apache to those that are modelled in Team Alligator?
A: We were frankly surprised at the hostility we received from the
hardcore fraternity for the flight model on Team Apache. We admit it:
we built the FM for accessibility. With the agreement of our Apache
advisors (including Desert Storm vets) a few compromises were made, but
it was because we were trying to get Joe Public to pick up this game
(something which we did with some success) and overcome the mystique
that helicopters were hard to master.
But we admit, we were stung by the criticism, however
unwarranted we felt it was. So earlier this year, we began a debate in
the COMBATSIM.COM™ forums, talking to the hardcore about what they
wanted. The subject of flight modelling came up repeatedly, and we
listened. We made some contacts through that discussion that has
resulted in some of your readers now actively helping us out with the
development and testing of the flight model. We hope the results will
show.
We will continue to retain the Team Apache 'simple' flight
model in Alligator as a novice option. But we will also have a
'realistic' option which in turn will have selectable features so that
players can choose effects such as phase lag, retreating blade stall
and vortex ring state on or off.
Players will be able to decide precisely what level of realism
they want. For example, if vortex ring state keeps on slamming them
into the ground all the time, they can switch it off. We are quietly
confident that our flight modelling will lead the field by the time the
sim is released.
Q: Low level flight means that terrain is critical. What is the
terrain resolution in Team Alligator and what kinds of terrain will we
see?
A: As stated earlier, our new Daedalus engine allows us to throw a lot
of polygons, objects and texture around. However, we've had to be
careful, trying to find a good balance of elements to ensure the
terrain not only looks good, but runs fast.
Our terrain runs on an adaptive mesh that will alter itself to match
the speed of your machine. On high spec machines, terrain polygon sizes
are measured in centimetres! As you've already seen in the demo, our
art team has done a marvellous job creating terrain textures with
contrast and depth of field that provides a real feeling of low-level
flight. (And they tell us they want to go back and make it even
*better*!) The rivers and bodies of water in the sim have to be seen to
be believed.
Q: Low level flight means that terrain is critical. What is the
terrain resolution in Team Alligator and what kinds of terrain will we
see?
A: As stated earlier, our new Daedalus engine allows us to throw a lot
of polygons, objects and texture around. However, we've had to be
careful, trying to find a good balance of elements to ensure the
terrain not only looks good, but runs fast.
Our terrain runs on an adaptive mesh that will alter itself to
match the speed of your machine. On high spec machines, terrain polygon
sizes are measured in centimetres! As you've already seen in the demo,
our art team has done a marvellous job creating terrain textures with
contrast and depth of field that provides a real feeling of low-level
flight. (And they tell us they want to go back and make it even
*better*!) The rivers and bodies of water in the sim have to be seen to
be believed.
We are dressing the campaign areas with thousands upon thousands of
objects, from buildings to trees to telegraph poles. Which reminds us:
we have a bone to pick. Your correspondent at E3 who commented that we
didn't have "tactically useful trees" in our demo obviously didn't
stick around long enough to check out our huge forest objects, which
are about as tactically useful as you can get!
We officially apologize and repent. I remember seeing the
trees and they did look great, but obviously not all of us saw them- Ed.
One thing we are not doing, however, is creating any cities. We thought
long and hard about this. The problem with cities is simply size; there
are many good technical reasons why we can't create a full-scale city
at high resolution with the technology currently available. Look at
Apache Havoc or even our own Team Apache. These are the best anyone has
yet achieved, but they are too small; a little piece of downtown with
no suburb. They look phoney.
So we are boxing clever. We are not aiming to produce convincing
cities. Instead we are facing up to the surprisingly more difficult
challenge of building a convincing town. And believe us, that's a
really tough task!
Q: The campaign in Team Apache was essentially a semi-dynamic
one, with a persistent environment but branching mission structure. Is
this also the case with Ka52?
A: It certainly is. There are tradeoffs involved with going for a
scripted mission environment or a fully-dynamic one. Both have their
adherents, but we are not convinced anyone can yet produce a genuinely
accurate and satisfying dynamic campaign. We are convinced that there's
nothing to beat a really well-crafted mission; for variety, breadth and
balance it is the best. Until the shortcomings of the dynamic campaign
can be overcome, scripting remains the way forward.
However, we have moved forward to a halfway house with our
semi-dynamic campaign mechanism. We are able to employ features such as
persistence of death and sophisticated flagging techniques to create a
replayable campaign around a branching mission structure. We are
creating literally hundreds of missions for the campaigns. Players
should be able to get at least two or three playings out of each
campaign, which by our reckoning is pretty good value for anyone's
money!
Q: How many campaigns will be modelled and what is the setting?
A:
We have two campaigns: Belarus and Tajikistan. Belarus is set against a
background of the collapse of the Russo-Belarus Pact in the face of
nationalist sentiment. Tajikistan deals with an uprising by the
Islamist opposition against the regime in Dushanbe.
The Belarus campaign is a high-intensity war set amid rolling
European terrain against a competent Russian-equipped Belarusian army.
The Tajikistan game takes place in the harsh hills and lush river
valleys of southern Tajikistan. This will be a low-intensity campaign
against hill-fighters every bit as tough as the old Afghan Mujahedin.
We will have a third area, Siberia, for the training missions and as a setting for many of the networked missions.
Q: Tell us about damage and physics modelling?
A: The
full-force physics modelling will see the player get battered around
the sky, as in Team Apache. We have a number of physical effects we
hope will impress people. As for the damage, we are taking the
sophisticated damage modelling of Team Apache a few steps further in
resolution and detail.
One thing to note is that we have expended quite an amount of effort
on modelling the warhead/ armour interaction in this sim. In the
absence of top attack weapons, attacking main battle tanks from the
front may not be a tactically sound option; you'll need to scoot around
to the sides or rear.
We distinguish between various grades of steel and composite
armour, as well as first- and second-generation Explosive Reactive
Armour. We also distinguish between types of warhead, from kinetic
energy rounds such as HVAPFSDS and APDU to chemical energy ammunition
such as HEAT and Tandem warheads.
Q: The AI in Team Apache sometimes appeared quirky. Tell us about AI modelling in Ka52, both for friendly pilots and enemies.
A: The helicopter AI has been one of the big areas of improvement and
redesign in Alligator. We use a similar mechanism for both good guys
and bad guys. When planning a mission you will have waypoints with a
number of parameters, such as orders and actions on contact with the
enemy. You will also have to specify Rally Points: fallback areas for
the team.
In flight, the player will be able to override these standing
orders with commands of their own. We learned a lot from Team Apache
and now have quite a powerful set of orders which allows players to
order their team easily into and out of the battle. We believe we've
even cracked the difficult goal in AI design: the ability to have
helicopters operate at true Nap-of-Earth heights.
Q: The command structure in Team Apache was seen as too limited by some pilots. Tell us about the goals for Team Alligator.
A:
You mean that the Team Apache player was limited to commanding their
own Apache team? Let me put your mind at rest. The new AI system
permits the player to command mixed formations and co-ordinate multiple
flights in the mission. Up to 16 helicopters under player control, plus
interaction with fixed wing assets and ground forces!
Q: Tell us about the mission builder/planner for Team Alligator.
A: At the moment these exist purely as tools for our games designers.
They are not convenient or stable enough to release with the game. We
might release them separately at some future date, but there's no plans
to do so at the moment.
Q: With the goal of more appeal to serious players, how will Ka52 remain accessible to the novice?
A:
We have the ability to set a simple flight model, and most of the basic
weapons functionality is automated enough for the novice to pick up the
game very quickly. We have an instant action mission for the player who
wants to get stuck in within moments of opening the box, but also a
comprehensive set of training missions for the studious newcomer who
wants to learn.
Q: Tell us about any changes to the view system compared to Team Apache?
A:
We have a similar suite of views to Team Apache, and are adding a
padlock view, since that seems to be desired by many people. One of the
principal differences is that we no longer have a 2D cockpit; only a 3D
option. That cockpit is hugely detailed, containing various
instruments, multi-function displays, and even an animated weapons
officer.
Q: The direction for sim design these days seems to be interoperability. Does SIMIS have any plans for connectable simulations?
A: Yes, but not with this product. It is something we are looking at seriously for future games.
Q: What are the goals for online play: number of players in which environment?
A: We are looking at six players for the co-operative missions, with up
to eight players in the deathmatch missions. Siberia will be the
setting for the death matches, with the other campaign areas available
for co-operative combat missions and the campaign.
Q: What will be the minimum and recommended system for Ka52?
A:
Minimum spec is a P200 with 32 Mb RAM and a 4 Mb accelerated graphics
card. Recommended spec will be a 300 - 400 MHz machine, with 64 Mb RAM
and an 8 Mb graphics card.
Q: Tell us about any changes to the view system compared to Team Apache?
A:
We have a similar suite of views to Team Apache, and are adding a
padlock view, since that seems to be desired by many people. One of the
principal differences is that we no longer have a 2D cockpit; only a 3D
option. That cockpit is hugely detailed, containing various
instruments, multi-function displays, and even an animated weapons
officer.
Q: The direction for sim design these days seems to be interoperability. Does SIMIS have any plans for connectable simulations?
A: Yes, but not with this product. It is something we are looking at seriously for future games.
Q: What are the goals for online play: number of players in which environment?
A:
We are looking at six players for the co-operative missions, with up to
eight players in the deathmatch missions. Siberia will be the setting
for the death matches, with the other campaign areas available for
co-operative combat missions and the campaign.
Q: What will be the minimum and recommended system for Ka52?
A:
Minimum spec is a P200 with 32 Mb RAM and a 4 Mb accelerated graphics
card. Recommended spec will be a 300 - 400 MHz machine, with 64 Mb RAM
and an 8 Mb graphics card.
Q: Thanks. Team Alligator is looking and sounding great and we're sure it will be a new landmark for helo sims in '99!
