| DID’s Perspective On Real Time Dynamic Campaigns
by Roger Godfrey
Digital Image Design 29th April 1998, Warrington, England.
As we approach the turn of the decade (and the millennium) it is interesting to consider how combat flight simulation has evolved in the course of the last ten or so years. A little under ten years ago the most advanced flight simulation available outside of the military was Falcon on PC\AT, Atari ST and Amiga. A very advanced product in its time, it boasted ten hand written missions and a relatively simple polygon graphics engine.
Ten years later we are spoilt with products that sport accelerated 3D graphics and accurate flight models, created by specialized aeronautical and software engineers. The aircraft simulations themselves are now so realistic that they are being used by the military to train personnel.
However, I personally suspect that the biggest change over the last ten years is that now, when we hand over our money at the local games store for the latest flight simulator opus, we are expecting to receive an accurate representation of the whole theatre of war. Certainly the emphasis over the last few years in combat simulation seems to be shifting to war simulators where the player interacts with the environment using his (accurately modeled) aircraft. At the heart of this change of emphasis is the Dynamic Campaign.
Often when picking up combat simulator products in your favorite software emporium the list of features will include "Dynamic Campaign". But what does that actually mean? One company recently promoted a product with "Dynamic campaign" listed on the box which turned out to be nothing more than a rather poor random mission generator. Should we consider this a "Dynamic Campaign"?
So What is a Dynamic Campaign Anyway
The definition of what Dynamic Campaign means certainly varies from simulation company to simulation company. For some companies it has become nothing more than a buzz word to stick on the side of the box to describe a linear or tree structure of missions; a shame, as a well crafted Dynamic Campaign can greatly enhance a simulation product.
For DID the term Dynamic Campaign means something else. Early in the development of Total Air War’s campaign system the designers chose to extend the concept of the Dynamic Campaign far further than the work performed on EF2000. Before work began on the development process the terms to describe the Campaign were defined. TAW is a Real Time Dynamic non-Linear Campaign. That’s quite a mouth full, lets define this a little more strictly:
Time constantly runs at one second per second while the campaign system is active regardless of whether the player is flying a mission or not. The campaign is not turn based and does not run in eight hour periods. There is no rerunning of missions or rewinding of the campaign system.
The system reacts to the simulated environment. For example a damaged runway will get repaired over time. However things that are dead stay dead. Office blocks will not magically be repaired nor shall craters be erased from the landscape.
The campaign is stochastic in that it is the opposite of deterministic. What this means in plain English is that the events are not predictable. The practical upshot is that even if the initial campaign settings of two games are the same then the events within the war will always be vastly different (you will never fly the same mission twice). A practical effect of chaos theory.
Why do Campaigns ‘Work’
Lets consider TAW a little further. The participants in the war will fight regardless of whether the player flies a mission or not. However when the player flies a mission he will encounter allies, neutrals and enemies that have their own missions within the simulated environment. Using SMART VIEWS the player can, if he wishes, view the progress of these missions. The events in the world happen around the player but do not revolve around him. When working on the EF2000 product we found this was essential to maintain ‘the willing suspension of disbelief’.
To make the simulation immersive the player has to believe, for the duration of the play session, that he is a fighter pilot. With computers as powerful as they are today producing convincing graphics is not the great problem it once was. However if the simulation of the aircraft and the combat environment is unrealistic it shatters the illusion and frustrates the player. Does this mean that the more realistic we make our games the more immersive they are? To a certain extent this is true but in accepting this we face a fundamental problem when designing campaign based games:
Realism does not always make for an enjoyable combat flight simulator.
For example who would want to fly a refueler around in circles for eight hours on patrol? Or who wants to fly CAP for six hours over an airbase without a single air to air engagement while all hell is breaking loose two hundred miles away? These are just two examples but the dynamic campaign designer faces more problems than could be listed here.
Of course you could argue that this is not important as a real fighter pilot experiences boredom on a regular basis and thus it is realistic. However we find it interesting that when we have fighter pilots visit the F-22 office at DID Towers they want to jump into the heart of the action; after all on the computer they don’t have to experience the boredom!
Another problem faced by the campaign designer is that in a true campaign environment a single pilot has little effect on the outcome of the war. In the real world a pilot could spend the entire conflict flying his missions perfectly but his side could still lose. In a video game environment losing a game through no fault of your own is an immensely frustrating experience and the campaign designer wants to prevent this from happening.
So how do we solve these problems? The Campaign and Simulation designers work very carefully to ensure that the product is realistic to maintain the suspension of disbelief and yet fun enough to keep the player interested so he comes back for more. Walking this tightrope is always a challenge.
Creating Flight Simulators and Developing Campaigns
As a games developer combat flight simulation is just about the most complex genre of video game that you could ever choose. From a programming perspective these simulations are extremely complex, require specialist personnel to create them and take an inordinately long time to develop. Specialized programmers are required for flight modeling, dog fight AI, avionics, world logic and graphics engine. This is not a comprehensive list as each project has it’s own unique requirements. Developing combat flight simulators is certainly not for the faint of heart.
Lets consider some of the issues that have to be addressed when creating a dynamic campaign system. On top of the development overhead for creating the flight simulation itself you have to consider the campaign system as well. In many products today this is the core component of the product; it is also the most complex, most difficult and usually one of the most costly modules to develop. To create a well crafted campaign system you require highly skilled programming staff who have good Artificial Intelligence skills and have an excellent understanding of the way wars are fought.
After they have been developed campaign systems take a lot of time to test. This is because of the nature of developing complex computer programs. As computer programs grow in complexity so does the likelihood of failure of the components within the program. To prevent this happening on the players home PC all bugs have to be tracked, logged and corrected during the development process. As you can appreciate this is not easy with a true dynamic campaign as their non-deterministic nature makes finding and repeating bugs very difficult.
Eventually after much debugging and eighteen hour days the product arrives in the shops. The consumers then buy and enjoy the product while the staff members responsible for creating it take a well earned vacation.
The Shameless Total Air War Plug
One of the most advanced (if not the most advanced) campaign systems currently under development is F-22 Total Air War from the F-22 Team here at DID. As previously mentioned the system is a Real Time Dynamic non-Linear Campaign. The system acts and behaves like a real conflict; the war is not put on hold while you are not sat in the cockpit of your F-22. It lives and reacts much like the real thing.
The product features 10 different campaign scenarios. Each one poses a unique challenge to the TAW player. The strategies used within the TAW campaign scenarios are based around real military doctrine on how the real F-22 will be deployed. However the enemy is not predictable and he is following his own strategies to achieve his own objectives. Only the most tenacious commanders and pilots will prevail through all the ten scenarios. F-22 TAW allows you to interact with the environment and control the outcome of the campaign using features never before seen in a campaign based flight simulator:
The Future of Campaign Systems
In the immediate future it is likely that more companies will produce combat simulation products that have Real Time Dynamic non-Linear Campaigns. Over time these systems will probably improve so that they encompass more facets of warfare on the contemporary battlefield such as accurately modeled tank battles or accurately modeled resource distribution in the Theatre of war.
Casting our crystal balls further in to the future things become far more interesting. Several companies over the last ten years have talked about having multi-player campaign systems you can dial in to; unfortunately few have come close to making good on their promises.
However Anyone who has played multi-player campaign with 8 pilots in EF2000 knows how much potential this holds. In the far future it is possible that Real Time Dynamic non-Linear Campaigns will run 24 hours a day on specialized servers. Thousands of participants can dial in and involve themselves in the war for as long as they want.
Of course we are a long way from achieving this ideal; there are many technical and game design problems to overcome before these kind of systems become a reality. In the mean time there is a lot of fun to be had from the current generation of dynamic campaign systems that exist in products available at your local games store.
The industry and the consumers now regard Dynamic Campaign systems as one of the most important components of a combat flight simulation product. Over the course of the last ten years flight simulations have been evolving in to war simulation products with the player taking the role of a single pilot within the battlefield environment.
Dynamic Campaign systems are very difficult and costly to develop which could mean that smaller simulator developers might not be able to compete with the larger more established brands. This coupled with the limited market for flight simulators and their development costs might lead to less combat flight simulator products on the shelf of local computer stores in the near future.
The technology of dynamic campaign systems is now well established and we should see them evolve over the course of the next few years. Hopefully future technology will allow us to create massive on-line multi-player environments for our campaigns; and hopefully there will be a market out there to allow us to do it.
In conclusions Dynamic Campaigns are very cool and they are here to stay. Many thanks to the Total Air War Team (especially Steve Hunt and Tim Preece) for their input on this article.
Response 2: Mark Doran: Who Needs Dynamics?
© 1997 - 2000 COMBATSIM.COM, Inc. All Rights Reserved. .
Last Updated May 2nd, 1998