Now that (unfortunately) we are bombarded with military analysts commenting on wha t possible strategies are, and news reports of how the fighting is proceeding, how many people are now more interested in a more realistic portrayal of how war works? For a long time I've been advocating doing things like supply lines, communication requirements, fatigue, morale and things of that sort that accurately simulate the factors a battlefield commander must make. Maybe because I'm a bit older (31), I vividly remember and kept track of the first Gulf War. I also remember english involvement in the Falklands War, our invasion of Panama as well as Grenada. So I wonder if people are now more curious about the many levels and layers of war....instead of thinking that Warcraft accurately simulates warfare conditions. Just as an aside, I have to admit that for close to 2 months now, I've barely touched my game design. The impending conflict made me kind of sick to my stomach that I'd be making a game about war. But at the same time, one of my main goals behind my game was to illustrate how horrible war is. Anyone who's been tracking my game design comments probably knows that I've wanted to get away from a cannon fodder mentality of strategy gaming, and also to try to create an attachment to your forces. After thinking about it, I decided to get back on track with my game design....mainly because I realized that perhaps now more than ever, the story I wanted to tell is more important. Sounds weird, but my strategy game in single player mode will be story based (and much more detailed than the story in say Starcraft). I recommend watching "Gods and Generals" just to get an idea that you can make a game about strategy, while telling a story as well. Also, try to refrain from political commentary....just talk about how you feel about making a game about war as realistic as possible and whether it's a good idea or not. [edited by - dauntless on March 26, 2003 9:20:00 PM]

I think the first question is, which kind of realism?

There have certainly been wargames and ultra-detailed simulations done before that cover the technical aspects you outline since they started being converted into boardgames decades ago; but they don''t directly engage a player in a storytelling experience either, staying fairly distant with pedestrian graphics(or abstract cardboard counters and hexes) and fairly high learning curves and playing times, especially for the uninitiated. Correspondingly, they are little known to many/most gamers, and experience success mostly within a niche market.

But the focus of your game, if I read you correctly, is really about telling the story. The problem, then, is how to present the details in an engaging fashion. I think it''s certainly possible...I''m reminded of games like Oregon Trail and spinoffs thereof that make a historical journey into a fun adventure. One big difference between a war-based game and those games would be that you''ll probably need to either add more underlying simulation, or stick rigidly to a given plotline. But I think the same model would work, at least for a single unit, though not a commander of forces over multiple locations - instead of hunting for food and trading with others, you''d fight battles and try to convince locals to help you, and make the tough decisions when there''s an emergency.

I have little idea how you could make the horror of war more real than has already been done as the level of operations gets higher up and more abstracted, however. In my mind it always either comes out to either a caricature of the reality(like most RTS games) or a sterile wargame. I suspect that a good way to do it, though, would be similar to an RTS, but add gameplay elements that make the loss of life a more pivotal event for the player.

The first one would be that all your units would be identifiable by their commanders and gain reputations and histories.

Second, losing men, particularly through simple mistakes that "shouldn''t happen" and suicide missions, would impose significant costs(bad PR, lower morale, loss of game, scripted help doesn''t appear or gets delayed, diminished end-of-game scores).

Third, the graphic nature of war could be displayed in a dramatic way that is meant to inspire alarm in the player. If you significantly deformed and discolored the landscape over time through bombing, artillery fire and the blood of the dead, and were really able to make it look just like a scene out of World War One, but one the player creates, you could probably get a lot of mileage out of it.

RTF-
All very good suggestions. You''re right in that it''s hard to convey the horror of a war at a higher and more abstract level required in a strategy game. One of the things I''m going to do though is similar to your first suggestion....that Commanders and units will be more than just abstract concepts. Units will have histories and even "war heroes" (units with lots of "presidential citations" or medal winners). Commanders especially will be more akin to NPC''s, in that they will have characteristic traits and in the case of the single player game, will be crucial players in the storyline. In other words, you''ll get to know them as people, and you''ll see your units not just as little counters or icons, but as people you have to take care of.

As for how realistic it will be? Well, you can''t get to the first person level of detail (though I''ve played around with the possibility of it), but basically I want it to be realistic in the wargame sense of realism (boardgame and miniature gaming) as compared to RTS computer games (with a few notable exceptions like games from Battlefront.com). I think RTS games simply don''t have the level of detail required to simulate battlefied considerations, nor do they really address the human component of war (in RTS games, basically units are just cannon fodder).

While it was at the tactical scale of combat, I always admired the Close Combat series. I liked how the sound effects gave you a sympathetic feeling for the troops you tried to order. Hearing troops scream in pain, their death throes, their panic under fire all lent to a very real concern for your troops. I hope to give a flavor of that in my game as well.

The problem with simulating war is this...

AI

We have thousands of troops, army of one? nah!

As long as there is AI, there is st00pidity. Even with a small crack troop of 30 or so guys, AI is going to be the biggest downfall. How is the AI going to act? What should the player do? What keeps the AI from shooting the player in the back, getting lost in corners, not completing the objectives they should have completed, standing around while an enemy is shooting at them, or dying because they could get out of a doorway(DaiKatana ).

Rainbow six solves this problem by letting the player tell the AI what to do before the player does his own thing. Maybe if the war simulation would take the Rainbow Six route, of a planner and a participant, but that isn''t realistic.

To tell a truly compelling war story, you need a much tighter focus than a traditional (or even innovative) RTS can give you. You need to focus on a single company of men. The movies have given us several examples of all the difference this makes, the most recent being HBO''s Band of Brothers.

Barring that possibility, the next thing up is to have the player embodied as the strategic commander in the game - say a general - and directly interact with specific subordinates whom you work hard to make unique and memorable. These subordinates will convey information with emotional overtones that can add color to a story ("We lost the 125th, sir. The whole damn platoon. What am I going to tell Davidson''s mom, sir? He was to have been discharged tomorrow..."), and will also impact the player if they die. An example situation would be the player making a tour of forces on "the northern front" and giving direct commands to a captain. The captain''s unit successfully carries out the orders, but sustains heavy losses including the captain himself. The player is then saddled with the responsibility of directly or indirectly appointing a replacement chief officer. The downside to this approach is that it removes the player from "the thick of things," which can downplay the visceral horror of war presentation.

A third option is to have promotion increasingly remove the player from the frontlines. The player could start out as a fresh graduate from officer training school, right on the battlefield. As the player recorded successes, promotions would put him/her in charge of larger and larger bodies of men and operations, increasingly distanced from the frontline. However, with the removal from conflict comes increased bureacratic responsibility - like burials. Traditional cinematic techniques like flashbacks and voiceovers could constantly "remind" the player of what war was like, driving home the point that war is not some glorious exercise but rather an unpleasant necessity.

quote: Original post by dede
The problem with simulating war is this...

AI

It''s actually not that hard to make realistic soldier AI. The key is to focus on what the user sees and perceives, and to adopt a holistic rather than component-wise approach. Gestaltian theory doesn''t generally apply to current video game AI. Implement as much logic as possible at the company level rather than in the individual soldier, so that the troops function as a cohesive whole (in part because that''s their true internal representation).

quote: As long as there is AI, there is st00pidity. Even with a small crack troop of 30 or so guys, AI is going to be the biggest downfall. How is the AI going to act? What should the player do? What keeps the AI from shooting the player in the back, getting lost in corners, not completing the objectives they should have completed, standing around while an enemy is shooting at them, or dying because they could get out of a doorway(DaiKatana ).

Quite simply put, we cheat. We let the AI actually know a little bit more than it should know - in some cases we explicitly help it solve nasty little problems - so that it always behaves intelligently. And we do this by embedding useful information in the game world - navigational paths on random terrain that cause the soldiers to mimick the real-world movement of troops in such a situation; grid tokens that explicitly inform the AI soldiers whether opening fire presents a significant risk to fellow soldiers, information generated by checking the occupancy of a grid cell and communicating to the "next" cell directionwise. Objectives and responsibilities will be required by the game design to have been clearly outlined and assigned before engagement commences, eliminating ambiguity (every soldier is operating on an exactly plan, with fallback parameters to transfer his responsibility seamlessly if he dies).

One thing I've thought about lately is the difference between a game that purely focusses on the battle and the strategy and little else, and one that truly introduces the human aspect.

We've seen evidence of people taking many different sides for many different reasons.

When a pilot talked about seeing the oil fields on fire, it made me wonder exactly what it took for whoever it was to light that oil field ablaze. Then I learn about the Baath party, the Shiites, etc. and I realize there are a lot of different groups with different agendae - some of them loyal, some deceived, some afraid.

I think a game about war, if it really wants to capture the essence of what war is about, should be honest about the different ways that everyone deals with it, and keep that feeling in the mind of the player. Many games want the player to feel powerful and capable. A true war game should make the player feel like nothing more than a person, in a conflict where people no better or worse than him are dying, are afraid for their lives, their families, and the consequences of their actions, no matter what side they are on or what agendae they have. It should also inspire the player to have feelings about his own country, and the country he is fighting, and the people in both. I think that would make the game very educational.

The horror of war doesn't need to be portrayed in the immediate action, but can be portrayed in the chatter among the troops, the briefings, news clips, etc. or even occasional interactions in the game (ground operations, taking hold of a city, seeing children and women running, maybe even towards you for help or to curse at you, whatever the reason may be)

What does the enemy do when they are confused and desperate? What if you are losing? It isn't just about completing the mission. You're doing a job. You're doing your duty. And people will die.

StarCraft portrayed reasons for conflict just to give an excuse to battle all the different races. It was pure storytelling merged with a game that wanted the player to experience all of the possible dimensions of the design. A true game about war wouldn't present new conflicts and attitudes as an opportunity for game situations or a strategic obstacle, or a new problem to face and master. It is simply a new element to the whole affair.

Just my two cents.

It might be a good intro to have the player training in boot camp, then being promoted to field operations, then having to go to war (where they finally see the real sides to it).


[edited by - Waverider on March 27, 2003 9:59:01 AM]

I didnt read all the posts, so this may of been mentioned or may be a bit off topic, but I think that the most ''realistic'' war game I have played is Medal of Honour, which was freaking amazing!!

The first mission on the Xbox and about the 6th on the PC is like saving private Ryan. You are slowly making your way in one of those little troop carrier boats to the beach, while the driver shouts "30 SECONDS!!". The guy in front of starts praying and then throws up. "5 SECONDS" and you can see the beach. The front doors of the boat swing down (you are still about 20 meters out since thats as close as they can get and the boat beside you just got blows to peices). Just as you are about to step off, the boat is hit and you are flung into the water. Struggling to get to the surface because all your gear is so heavy, you crawl to the shore and regroup with a team that came in on another boat. There are people on the ground screaming while medics desperatly run around trying to fix them up. If you play it in surrond sound with the volume cranking, you have people yelling, screaming, 10000''s of bullets flying all over the place while grenades and morters explose with a deafning noise that shakes the screen. The sense of realism is awesome.

Come to think of it, Im going to go play it now

quote: Original post by Waverider
A true war game should make the player feel like nothing more than a person, in a conflict where people no better or worse than him are dying, are afraid for their lives, their families, and the consequences of their actions, no matter what side they are on or what agendae they have. It should also inspire the player to have feelings about his own country, and the country he is fighting, and the people in both. I think that would make the game very educational.

Here we begin to struggle with the conflicting objectives of education and entertainment. We still want the game to be entertaining, and the truth is that few media products can achieve that balance of being both entertaining and educational for mature/adult audiences (so so-called "edutainment" products don''t count; sorry Borderbund).

In film, the traditional tactic is to interleave highly visceral, taut and suspenseful action sequences with introspective/emotional/informative scenes, but hardly are these delivered from a single perspective. In a game, the player usually wants to feel attached to his/her avatar and experience/perceive things as the avatar logically would, which limits the amount of jumping around we can do with the camera for storytelling purposes. In essence, we want a Band of Brothers meets Cast Away meets Medal of Honor, framed in a strategic context so we can present a cohesive story without detracting too much from the game.

quote: The horror of war doesn''t need to be portrayed in the immediate action, but can be portrayed in the chatter among the troops, the briefings, news clips, etc. or even occasional interactions in the game (ground operations, taking hold of a city, seeing children and women running, maybe even towards you for help or to curse at you, whatever the reason may be).

Absolutely. What I like most about this approach is that it doesn''t eliminate the possibility of single-perspective delivery (a thematic, narrative wargame could allow for multiple POVs - the equivalents of "Allied" and "Axis" hackneyed in so many RTSes, but expanded to more roles - and present the same conflict in different ways and with different motivations).

quote: It might be a good intro to have the player training in boot camp, then being promoted to field operations, then having to go to war (where they finally see the real sides to it).

The downside to this is that it might take too long. Intro/tutorials should be brief (tutorial is a logical role for training camp, allowing the player to learn how to navigate and interact with game entities). The player could be plunged into the war experience, but carefully biased to supply all the shock value without making the beginning unnecessarily difficult.

quote: As long as there is AI, there is st00pidity. Even with a small crack troop of 30 or so guys, AI is going to be the biggest downfall. How is the AI going to act? What should the player do? What keeps the AI from shooting the player in the back, getting lost in corners, not completing the objectives they should have completed, standing around while an enemy is shooting at them, or dying because they could get out of a doorway(DaiKatana ).

quote:
Quite simply put, we cheat. We let the AI actually know a little bit more than it should know - in some cases we explicitly help it solve nasty little problems - so that it always behaves intelligently.

You do realize that all of my examples are from REAL games...

What should the player do, (Final Fantasy Tatics, Vandel Hearts, you have to follow the AI, not vice vers)
AI shooting the player in the back (Rainbow 6, Mechwarrior 4, almost endless)
getting lost in corners, (super soldior in Return to castle wolfenstien.)
not completing their objectives (final fantasy tatics, Shining force III)
standing around while an enemy is shooting at them, (oh god! The list is so long, I would be here all day!)
dying because they get cought in doors, (DaiKatana)

Even tho we can cheat, the only acceptible way to cheat, is to make the game so linear, that neither the player nor the AI have a choice, they have to do things in order.

As soon as an AI has to compansate for a player, then everything is moot.