Could you please introduce yourselves and describe your role on Conflict Zone?
I’m Serge Autard, production manager for MASA’s games.
I’m Olivier Denis, team manager for Conflict Zone.
I’m Sylvain Constantin, Conflict Zone’s game designer. My role was first to create the game design and scenario. Throughout the development phase, I worked with the production team and our in-house testing team to improve the gameplay and make Conflict Zone as good a game as possible.
Hello I’m Stéphane Maruejouls, lead programmer of Conflict Zone.
I’m Christophe Meyer. Julien Devade and I run the support and liaison team with MASA’s R&D department. We were in charge of supervising the implementation of DirectIA®.
Could you tell us more about MASA?
MASA’s a four-year-old European company with offices in Paris (FR), Brighton (UK) and Peking (CHN), and representatives in New York (USA) and Hong Kong (CHN). With a staff of 120 (including 45 PhDs) – set to grow to over 200 in 2001 – MASA targets consumer and industrial markets with a range of exclusive artificial life and optimization technologies.
MASA develops and sells business intelligence solutions designed for use in mission-critical, high-performance environments.
Our Artificial Life proprietary technology platform (DirectIA®) is a unique generic software engine that represents the most advanced commercial application of adaptive intelligence – a multi-disciplinary field that enables a software application to learn from, adapt to and anticipate changes in its usage and environment so that interaction with the application is personalized and relationship-driven.
DirectIA® is the sole generic tool of its kind and the only integrated SDK to let game developers quickly develop a complex AI that mimics what living beings would be doing in similar situations. Conflict Zone relies heavily upon a specific version of the DirectIA® engine in a version that hasn’t been commercialized yet. For instance, it introduces the SmartOpponent module which lets the PC adversary adapt to the player, whatever his level.
How did MASA move into video games?
MASA has been developing DirectIA® for industry for 4 years. AI in video games has been very disappointing until now, so it was a very natural process for us to develop ‘showcase’ video games based on our own advanced AI technology. We thought the RTS genre was especially appropriate to demonstrate the technology. So that’s how the development of Conflict Zone started. And now we’re in the process of developing other games in different genres. We felt that virtually the whole game industry was suffering from a lack of new and original game-plays. After the 3D revolution, we think it’s now time to develop a whole new set of game-plays, hopefully setting new standards in each genre, and this – in our view – can only be done by drastically enhancing the so-called AI technology that’s currently used in games.
You aren’t game developers by profession, although you now have a consumer division. Why this change in company ethos?
It’s true that when MASA was created, the aim wasn’t to become a video game developer. Nevertheless, the moment MASA decided to produce video games to promote the great potential and relevance of DirectIA® in this field, the development of the video game department began. We mainly recruited game professionals with several years’ experience – backed up, naturally, by the AI middleware department.
So it’s not at all a question of changing philosophy. It’s simply a matter of finding a new application for our research. What’s more, our video game business has been so successful that we’re now creating subsidiaries. The company that was set up to deal with everything relating to on-line games is called MASAGAMES (code name). It’s based on a specific on-line version of DirectIA®, plus a network platform for our games. An enormous number of people connect up to these games, and they also perform outstandingly well.
We reckon that – with DirectIA®, and capitalizing on our experience in this field – we’re going to be able to offer game-plays that simply haven’t been available in video games up to now. Video games are definitely not only technological ‘window dressing’ for the MASA group. They have a special place in their own right, and we intend to prove it pretty soon now!
What does MASA bring to the gaming table that no other developer does? What makes you special, unique?
Well in addition – of course – to DirectIA®, the video games department has naturally inherited the methodologies and the production and quality assurance processes set up by the industry department. For example, our on-line subsidiary will probably be the first in the video-game world to benefit from ISO 9001 standards.
This will help us control deadlines and code quality and push forward the production process more calmly than a traditional developer can do. It also gives us the chance to spend more time fine-tuning the game-play – which is still the most important thing! And once again, on this score, our R&D center, the researchers who go to make it up, and above all DirectIA®, should give us a considerable head start!
How many people work on the project?
The core team’s made up of about 15 full-time staff for Conflict Zone. To that team, we must add another 15 from MASA’s R&D department – people who’re dedicated for months on end to developing specific DirectIA® components for the game industry.
How would you describe Conflict Zone?
Conflict Zone is a 3D RTS game that’s set in the context of modern conflicts (like the Gulf War or Eastern Europe). It offers (much-appreciated!) basics of the following kinds: developing a base, managing resources, fighting battles for the control of resources and strategic positions, etc.
However, it stands out distinctly from its predecessors and competitors in several ways. First of all, the game-play is enhanced by having to take into account a lot of realistic factors, and this adds a lot to the sheer enjoyment of the game: there are civilian populations on the ground, for example, and the media are filming the action. These two ‘components’ are at the heart of the game, the civilian populations constituting a new and highly strategic ‘resource’ (no cynicism intended!) that has to be watched over, and the media being masters of your popularity and, consequently, your means. So Conflict Zone is an extremely tactical game in which brute force is not always the best policy.
What’s more, for the first time in a strategy game, the player can delegate whole tasks to commanders (building a base, defending it, attacking a strategic position, etc.). The commanders each have their own personality and they can control around fifty units, adapt to a given situation and learn as they go along.
Likewise, all units in Conflict Zone behave in realistic, autonomous ways, so they react intelligently. Gone are the days when idiotic units in RTS games either failed to react or behaved nonsensically – at least in Conflict Zone!
As for the rest, the best way to discover it is by playing the game!
The RTS is rather an over-stocked genre – why have you chosen to work in it? Do you find the paradigms are restrictive or can you find ways of working around them? Are you worried about people’s expectations and will the game match or exceed them?
To create our first video game, we were lucky enough to have a fantastic tool at our disposal – namely our artificial life SDK (Direct IA™). All that remained was to decide which kind of game to choose!
It’s true that that are a lot of RTS games on the market, but we reckon that enormous contributions can still be made to the genre, principally in terms of management and unit autonomy, but also with regard to delegating orders at ‘macro’ level. Lastly, our core team was made up of fans of this kind of game, so we were very excited and motivated by the idea of taking up the challenge to produce an innovative, top-quality RTS – which is what we hope we’ve done! We’d have been happy enough just to meet the expectations of RTS fans and to throw a few surprises at them. And we’ve still got a lot of ideas in store for our next RTS game!
What do you think are the strongest features of the game?
Without the slightest doubt, the truly innovative features in terms of game-play were the ones that Direct IA™ allowed us to develop, namely:
- the game intelligence and the computer opponent’s ability to adapt;
- the system whereby part of the player’s troops can be delegated to commanders, which is one of Conflict Zone’s strong points, in addition to being highly original;
- the autonomous behavior patterns of all units (armies, civilians, media, animals, etc.), providing a foretaste of what’s meant by ‘artificial life’ and considerably improving the game-play.
The inclusion of non-military units (media and civilian) – outside the player’s control and yet performing a dominant role in the game-play – is also highly innovative and one of Conflict Zone’s key factors.
What are the challenges that developers of 3D real-time strategy games face, as compared to designing a 2D game? What becomes easier and what becomes harder?
The first challenge is the interface and controlling one’s position and relative direction, because it’s more difficult to navigate easily when using a 3D camera with complete freedom of movement. This is a powerful tool that you have to be able to master perfectly. I believe we achieved this goal, mainly by offering several types of pre-defined cameras, including one which gives a 2D-type point of view for players who find the wealth of possibilities represented by 3D problematic.
Next comes the frame-rate, because it’s particularly difficult optimizing a 3D engine to animate an entirely 3D world, in high resolution, with hundreds of ‘intelligent’ units on the ground. I don’t think we did too badly on this score!
But 3D also brings lots of benefits. You get much more involved in the game. The player isn’t a spectator any more, but an actor in the events. The environment (relief of terrain, buildings, etc.) is represented on a realistic scale, coming fully into its own. You get totally immersed. Once you’ve got used to 3D RTS games, it’s much more difficult to go back to 2D games. They feel completely static by comparison.
Are there any inherent limitations associated with developing a 3D RTS game?
As far as graphics are concerned, our production specifications were very challenging. We needed a lot of units on the screen, many texture surfaces for the units and a particularly detailed landscape. To achieve this, we developed a system of pre-calculated multi-meshes for the units and a real-time optimization system for the landscape.
The optimization technology we developed freed us from the potential limitations inherent in 3D games, above all the CPU time required for 3D display, on top of the one required for the AI.
How much of an effect does a 3D engine have on the gameplay? Does it assist the creation of believable physics or AI?
Physics engines appeared just after 3D engines, and the two are closely linked.
The link with AI is far more indirect, but it exists all the same. Effectively, 3D makes it possible to create a realistic environment and adopt whatever point of view you want. As a result, the behavior of units on the ground – be it autonomous or group behavior – takes on a whole new dimension since it occurs in 3D surroundings, just like the real world, and can be observed at leisure thanks to the freedom of the cameras. It’s a real pleasure to watch one of your commander’s units taking the enemy from the rear by slipping down a narrow gorge while the bulk of your troops make a frontal attack! You can switch from circling helicopter-type aerial views to quasi-subjective points of view with perfect continuity, and this lends an imposing, cinematographic dimension to the action and – of course – to the behavior of your own and enemy units.
It’s only in the past year or so that 3D real-time strategy titles have evolved beyond polygonal units maneuvering and fighting in only two dimensions. In what ways do you believe the gameplay aspects of 3D strategy titles will change over time? What would you consider to be the ultimate evolution of this process?
As things stand, I think that neither the players nor the creators are really prepared to exploit the full potential of 3D – that is, a game with 3 axes. Homeworld was the first game to exploit 3D fully. It indicated what an RTS game played in 3 dimensions could be. It was an interesting initiative, but not totally conclusive. When all’s said and done, using the 3rd dimension is neither very intuitive nor particularly interesting. But I’ve no doubt that in years to come, games like Homeworld – such as the very recent Cataclysm – will push the conquest of the 3rd dimension much further, using a more intuitive interface and strokes of game-play inspiration that make the use of 3D absolutely essential. In any case, the playing field that’s most appropriate for this approach is clearly outer space (or sweeping backdrops).
Space was once an unexploited medium for all developers, not just those creating strategy titles. Nowadays, however, the market is virtually flooded with such games, with Homeworld showing that 3D space was a viable scenario for a strategy game. What sort of setting would you predict to be the ‘next big thing’ for all genres, including strategy? Is there another previously unexploited scenario that could be opened up by the move to a 3D engine?
As far as ‘classic’ RTS environments are concerned – those based on terrestrial conflicts, mainly between human beings – we’re going to see an expansion in the range of available ‘historical’ environments, with highly specialized games on the Roman Empire, the Napoleonic wars, the Far West, the 2nd World War, etc. We’re already seeing a number of teams embarking on this kind of project.
But to get back to the use of 3 dimensions, I’d say that – apart from outer space, of course – the other interesting and as yet virtually untapped environments are the oceans and the depths of the earth. As such, Microids’ ‘The Ants’, for example, is a highly original and enthralling project.
Did any games have an influence on the development of this title?
The Command & Conquer series, Starcraft and Total Annihilation, are games we love and which influenced us a bit for some specific parts of the game design. But our core game-play is completely original! It wasn’t so much this or that stroke of inspiration in the game-plays that we sought to reproduce, but rather their success in terms of balance, progressiveness, rhythm or simple enjoyment. Those are very tricky parameters to define!
What game does the development team play? What games have inspired the creation of Conflict Zone?
The team’s made up of RTS fans. As a result, games like C&C, TA or Starcraft (which is still popular at lunchtime) were much played and appreciated. Though some of these games weren’t directly inspiring from the game design viewpoint, they did provide benchmarks for important considerations like the balance of power, game rhythm or length of play.
But, above all, there was a general feeling that all these games had a frustrating lack of AI (to control units, groups of units and the computer opponent). That’s what made us keen to break new ground with this kind of game.
Otherwise, the few breaks that the team got during development were livened up by games like Quake 3, Counter-Strike, Starcraft, Re-volt and Midtown Madness 2!
How was the idea for the gameplay developed?
Our first intention was to create an RTS game based on contemporary warfare. We also really wanted to change the classic RTS system of resource gathering for something new and more realistic. Then came the ideas of command points, popularity points and media influence.
We also wanted to create a game that was more strategic than tactical, more based on high-level choices and less on micro-management. That’s why we decided to design and integrate a commander system to help the player.
What is the scenario?
Conflict Zone is entirely about modern warfare, in a fictitious but realistic environment. In the near future, several aggressive independent countries attack neighboring countries, for one reason or another. The ‘ICP’ (a United Nations-like organization) has to fight to protect the countries attacked through various campaigns of 4 to 5 missions each, taking place all around the world.
During the game, the player learns that the aggressive countries are manipulated by ‘GHOST’, a secret terrorist organization, which is in fact, the real threat.
There are two different campaigns and the player can either play for ICP or GHOST.
The media are present on the battlefield, and as ‘Popularity’ (linked to what the media relate) is a resource, the way the player handles it has a significant impact on the outcome of the battles: Popularity is one of the major keys to success in Conflict Zone!
Is the role / influence of the media different when playing ICP or GHOST?
Yes, of course! The ICP are the ‘good guys’. They have to behave well with civilians and they have to have as ‘clean’ a record as possible to score well with public opinion. On the other hand, the members of GHOST are the ‘bad guys’, and they hardly care at all about civilians. The only important thing for them is their military efficiency: the ‘kill’ ratio. These major differences make the two sides even more interesting to play, since the game-play alters dramatically depending on which side you play for!
How is the role of the media included? What people were involved in the process and which real experiences is the simulation based on?
We watched TV reports made during the Gulf War and the war in the former Yugoslavia and were helped out by some of the major networks. We grew to understand how media reports can influence public opinion and therefore the war itself.
But we didn’t want to create a realistic media system, nor did we wish to relate real events. Our goal was simply to create an original and interesting game-play feature that handles the media.
We believe our system is realistic enough to oblige players to think through the consequences of their actions at any given time.
What was your biggest challenge during the development of the game?
The biggest challenge, apart from the ones we’ve already mentioned, was probably time!!! We only had 18 months to create the entire game from scratch. It was very challenging to have to develop everything and fine-tune the AI and game-play accordingly. In the end, I think it’s safe to say that we’re all very proud of the game!
How was the intelligence created? What models are used and how is the behaviour of independent and general units simulated?
Artificial life (adaptive technology) is the MASA Group’s main focus.
After several years of R&D, we developed a generic tool called DirectIA® which constitutes a world first as middleware (and the only official middleware of this type on PSX2). It’s also a first in terms of scientific design, having nothing to do with classic AI technology of the finished-state automaton or neuron-network types.
Conflict Zone uses DirectIA®, and this is the first time that such a tool, initially designed for the professional world, has been applied to a video game. It enables us to change the relationship between the player and the game from several points of view.
First of all, the units are endowed with emotions and highly developed behavior patterns. If they’re individually unable to make sense of a predicament, instead of allowing themselves to be destroyed they can evolve behaviors to cope with the situation more appropriately: automatic coordination, mutual assistance between units, wise choice of targets, etc. So we had absolutely no recourse to any of the classic models that are usually encountered in video games.
Of course, it’s of fundamental importance in a game that units don’t act completely as they please. The player should keep constant control over his troops. That’s why we hit on a compromise between DirectIA’s possibilities and what the players themselves expect from a strategy game. We’re very satisfied with the result we got, because players don’t end up with the frustrating feeling that they can’t control their troops’ behavior, nor that the latter are puppets, devoid of all intelligence, as is the case in the majority of RTS games.
Secondly, the player’s commanders have different personalities and abilities. These evolve as the player goes along. He has to learn how to recognize the strengths and weaknesses of all of his commanders and make the most of their different qualities.
Thirdly, the player confronts an artificial opponent which adapts itself to his strategies just as a human player would do (using MASA proprietary Smart-Opponent technology). The opponent doesn’t follow a script to destroy the player. Instead, depending on the environment, it creates stressful and dangerous situations for the player. The opponent learns from the player and its real aim is to push the latter to surpass himself – innovating, developing new strategies. In doing so, the opponent can sustain a level of play that tends to correspond to the player’s own.
Who is Conflict Zone aimed at? Will it appeal to all? How easy will it be to get to grips with the game mechanics and how difficult a challenge will it set? How many hours of gameplay will it realistically provide?
Conflict Zone is intended for a broad public and not just for RTS fanatics. It isn’t a very difficult game to get used to. All strategy game fans will make sense of it almost immediately, and even players who are new to the genre will quickly get accustomed to it.
We decided to introduce different levels of difficulty so that everyone can get something out of it, whether they’re an expert or not at this kind of game. Of course there’s a tutorial, and the campaign structure itself is designed to explore and get used to the whole game-play progressively, in particular with regard to the use of commanders. So even if Conflict Zone is a very rich RTS game, which can be played in great depth, nobody should find it disconcerting.
The play time required in solo mode to finish the 2 main campaigns is between 50 and 60 hours approximately. To this you should add the possibility of re-playing all missions against an AI of increasing strength which analyzes its opponent’s way of playing and tries to counter it (Smart Opponent ™). For the first time in an RTS game, then, there’s a real point in re-playing missions!
Naturally, there’s also the multiplayer mode which enables you to play against AI and/or human opponents on cards that were specially designed for this game mode. More than 20 cards are supplied, allowing you to play in skirmish or multiplayer modes.
Lastly, on the game CD we’ve included the level editor with which all the cards were created, so players can now create as many new cards as they wish.
As far as I can tell, there are over 50 different types of units. How many are there exactly? Are they all playable?
Each of the 2 sides has just over 60 types of unit (30 buildings and 30 mobile units), making 130 in total, and they’re all playable.
In addition to these units, there are non-playable media, civilian and animal units! In all, then, we’ve modeled and animated more than 200 different types of unit!
How do NPCs (animals, civilians) affect the gameplay?
First and foremost, the presence of animals and civilians helps strengthen the impression of a credible, realistic universe that we wanted to create in Conflict Zone. The animals move around calmly and run off when soldiers pass by. The civilians try to avoid bombing raids, request assistance from NGOs to reach refugee camps, etc.
But the part played by animals and civilians isn’t only decorative. There’s also a strategic side to them. This is specially true of civilians, since they’re one of the players’ essential resources (popularity and financial means, in particular, depend on them). Controlling inhabited zones, then, represents one of the main stakes in the Conflict Zone games.
As for animals, whenever they run away or let out cries of panic, the player should take heed: they may serve to alert him to the presence of the enemy in the vicinity.
How many campaigns are there and how long are they?
There are 2 campaigns, each composed of 16 missions, plus one bonus mission per campaign, accessible under certain conditions.
There are also more than 20 skirmish and multiplayer maps, plus the map editor.
Compared to other RTS games such as Command & Conquer and Starcraft, how large will the levels be?
The maps vary a lot in size, growing as the campaign progresses. The biggest maps can cover more than 25km², which is a considerable size and a lot bigger than most other RTS games can offer. To give you some idea, it takes more than 3 minutes for the fastest vehicle to cross the largest maps from one end to the other.
What is ‘Smart Opponent’?
Smart Opponent is a technology that allows the computer AI to react and adapt to the player’s behavior and anticipate his strategy by analyzing and learning from his way of playing. The more someone plays with Conflict Zone, the more it’ll be difficult for him to surprise and beat the AI. Don’t expect to win the next round if you use exactly the same tactics! With this technology, the player confronts a PC opponent who never ceases to be challenging.
The political/media aspect of the game is quite an innovation. What’s the inspiration? How does it affect the overall gameplay (for both camps)? What new skills will players have to learn in order to succeed?
There’s no particular inspiration. We just noticed how the media and public opinion can influence modern warfare. We found it very interesting, and decided to incorporate that aspect into our game (which we wanted to be realistic).
The media aspect plays a key role in Conflict Zone, since the popularity of a side is directly linked with its ‘technology tree’: for example if the ICP kills civilians, it suffers a drop in the popularity stakes which could result in being unable to order new tanks.
The players have to keep an eye on their popularity level and behave the way their superiors expect them to behave: an ICP player has to rescue and protect civilians, and GHOST players have to forcibly enlist civilians in their army and act aggressively towards the ICP. The players must learn not only to try to destroy the enemy, but also to monitor their own behavior. In fact you have to deal with brand new parameters – and that’s great fun!
Multiplayer (and especially online gaming) is apparently the wave of the future. Does Conflict Zone offer a multiplayer option? Why has MASA chosen to concentrate on AI while other companies are spending less and less time on it? Is there a future for single-player gaming ?
Of course Conflict Zone has a LAN and Internet multiplayer mode, allowing up to 8 players to play together, and we must say that the Conflict Zone multiplayer option is really tremendous fun!
But, of course, there’s also a future for single-player gaming! We think that people still love, and will continue to love, single-player games: you can play without outside boundaries, at your own rhythm and face different challenges. And we’re absolutely certain that, after 3D, AI – even more than physics -will be the next frontier in video games. So much remains to be done and improved!
What are the Internet features in the PC version?
Conflict Zone can be played in multiplayer mode on the Internet (from 2 to 8 players with a high-speed connection, whereas 2 to 4 players would be more reasonable for traditional modem connections).
Is there an estimate of what the required specs for the PC version will be?
Yes, the minimum requirement is a Pentium II 266-type PC, with 64 megabytes of RAM and a Voodoo 2-type accelerator board or better. The ideal machine, to exploit the game’s full potential, would be a P3 500 (or higher) with 128 megabytes of RAM and a recent 3D card of the TnT 2 variety or better.
Do you think a video game such as Conflict Zone can help the army manage their relations with the media?
No! I’m convinced the army already knows its job! At least I hope so… (laughter.). Seriously, though, we’ve been particularly aware of that question ever since MASA’s industry department began working with different armies throughout the world on conflict simulations. So we’re sure they know what they’re doing!
Did you encounter particular problems in the development? Would you please name and describe them?
Conflict Zone has a really sophisticated AI. As we had only 18 months to create the game entirely from scratch, we had to develop and implement several parts of the game AI at the same time, and it was sometimes hard to synchronize it efficiently. But in the end, working this way probably saved us a year or more of development!
What are the future plans for the game? Are you already thinking about expansion packs and sequels? Could the AI engine be used in different genres?
We’ll probably offer new maps for skirmish and multiplayer modes, and maybe some new military units. We’re also thinking of creating new AI players to fight with, which is even more interesting. Check out the Conflict Zone web site!
We’re already contemplating the possibility of an add-on with new campaigns and a lot of extra stuff. And, of course, if Conflict Zone encounters the kind of success we’re hoping for, we’ll develop Conflict Zone 2! We already have plenty of ideas for new features, new game-plays and improvements.
DirectIA™ is an evolutionary, modular SDK which will, of course, be used in our forthcoming products, both RTS games and others. We’re learning how to get more and more out of it and this fantastic tool and, with each new version, it’s also making progress. So the future’s very promising!
Do you have an anecdote to tell?
Before we created a video game department in its own right, our parent company, MASA, developed artificial intelligence programs for major industrial accounts. About a year ago, military representatives (from a country whose name we can’t reveal) came to see us to assess our technology. They saw a demo of Conflict Zone, which was then in pre-alpha, and they were very enthusiastic about the 3D engine and, most of all, the artificial intelligence. So much so, in fact, that they ordered a specific version – a Conflict Zone by-product, which was more precise and realistic – on which to train their officers!
Do you have any additional remarks?
Yes. We developed very intelligent autonomous behaviours for the units in Conflict Zone. And, paradoxically, we came to realize that too much autonomy and intelligence can spoil the player’s enjoyment because it deprives him of total control over his units. So we had to reconcile ourselves to finding compromises in order that the different units appear much brighter than in other games while remaining soldiers who obey their general’s orders! The fine-tuning was no easy matter, but we’re proud of the result!