There are some that believe that with the emphasis on graphics and interface, something has been lost in the wargames and simulations that are being published these days. Some think that the depth of these games are not comparable to older designs that didn't have to struggle with today's high expectations for great graphics and an easy to use interface.

In the case of simulations (and many wargames), one of the goals is to be accurate to the situation being modeled. For many simulation gamers, their assumption is that a "true physics" model is always going to be more accurate than one based on formulas and die rolls, that complexity will always be better than simplicity. My experience is that often just the opposite is true.

When setting out to do a true physics model, the only way to make it accurate is to fully understand just about every factor that can affect the outcome. Our Great Naval Battles series of games tried to simulate naval gunnery and damage to a ship with incredible complexity. The shells were fired and moved to their targets based on "true physics".

When hits were recorded, the exact location of the hit on the ship determined the damage, and players had to worry about items such as counterflooding to offset too much water in any particular part of the ship. Sounds great, except for the fact that it was practically impossible to get all of the inputs to this model correct, leading to inaccuracies in the results. In the end designers and testers end up tweaking the model to make it "feel right", but had to do so while dealing with a very complex system.

Fighting Steel. Click for larger.

In Fighting Steel, our new WWII naval game in development, we have decided to go back to the traditional wargaming approach but with the advantage of having a computer do the complex calculations. We use a spreadsheet for gunnery that accounts for 20 different variables that can significantly impact a ship's chance to hit the enemy.

Although no physics are involved, the ability of this model to simulate history is much better than the physics model used in the GNB series. While we want to show off our 3D graphics and provide an interface that is easy to use and understand, we don't believe realism is being sacrificed.

We do however reserve the right to make design decisions regarding the choices we will allow the players to make during a game. Eliminating player placement of damage control parties in Fighting Steel is being done because we want to put the player into the role of a division commander instead of a captain of a ship or a damage control officer.

Once again, the routines that generate and repair damage will yield realistic results, but we don't believe that the how-to's is something the player should need to worry about. On the other hand, the ability to make decisions regarding when to use searchlights, starshells and recognition lights will yield a level of realism in night battles that has not yet been seen on the PC.

The real programming and design complexity we are dealing with has to do with getting the 3D auto-camera to provide players with a fun and exciting view of the action while providing players the ability to control what they are seeing. It's solving this issue that will determine to a large degree just how much fun the game is to play.

As for the complexity and realism that is "under the hood", gamers need not worry. Just because we want to make things easier to play doesn't mean that the realism isn't there. Fighting Steel aims to blend complexity with ease of user access, while maintaining a sense of involvement and intensity of action with the best out there.

Look for Fighting Steel in early 1999. For our Fighting Steel preview see E3 Naval Preview.