Discussion: Tactical Battle Simulator

[Samurai Waki]

Actually controlling logistics and supplying an Army whilst maneuvering my main force, sending out scout parties, skirmishes... etc, I would love that. I am a patient man, and at least half the fun for me would be the process leading up to that battle, rather than... perhaps the battle itself. In RTW, I get rather burned out quickly, because I like building up my empire, and not having to fight ten battles in one turn. It gets rather anti-climatic after awhile.

[Duke John]

Funny that all of you start talking about a strategic simulator. Does this mean that when it comes to simulating pre-modern warfare you are far more interested in the operational level than the battle?

I can foresee a few issues if the simulation would be truly historical (i.e. historical events can but do not need to happen ingame). For example, you control the French army during the Hundred Years Wars game leading to the battle of Poitiers. The French King pursued the English who could not join up with the Duke of Lancaster while the French are able to achieve that. The English are starved as their supply lines are cut. As a player you would then assume that victory is in your pocket but in real life it proved to be a disaster for the French army.

During the Wars of the Roses, forces were largely identical in composition. A game covering that period wouldn't provide you with any stats (I wouldn't like to see that anyway) so what would indicate the odds of winning? Numerical superiority didn't say a lot as plenty of battles were won by the smaller army.

So when fortune needs to be fickle if you want to simulate history, would it annoy you that you cannot control the outcome, just nudge it a bit?

[CBR]

And really short: Could a battle simulator with a 1:1 scale and historically correct rates of movement and casualities still provide the player with interesting and challenging gameplay from deployment to routing?

If Les Grognards will be fun then doing it 1:1 scale will be fun too. That game is supposed to use time compression of 10:1 at least in the beginning when armies are just maneuvering and then go down to 1:1 time. And AFAIK will be the closest thing to a Napoleonic simulator we have seen so far. A similar approach could be used in your 1:1 scale battle simulator.

Of course the simple linear warfare of ancient/medieval warfare is very different than Napoleonic warfare and would give the player less ability to command and control the army.

You mention: "I have made renderings of 10,000 soldiers and it just becomes a big mass." What do you mean by that? You have a simple but working battle engine?

From a gamers point of view the main difference between the Total War engine and a more realistic engine would be command and control, and how units actually could and did maneuver. Also there would no longer be this convenient birds eye view that enables the player to know the numbers and depth of enemy units and/or reserves.

Even using the not so realistic combat mechanics of the Total War engine you could get a long way by just implementing the lack of information and more limited manuvering and control.


CBR

[Duke John]

You mention: "I have made renderings of 10,000 soldiers and it just becomes a big mass." What do you mean by that? You have a simple but working battle engine?

I have so many ideas and considering that the mods that I have worked on are almost all featured in gaming magazines, I figured that I must be doing something right. So I am now indeed learning how to program and take the leap into developing a game (as a hobby). I do not have a game engine yet (read, not for the coming months), although I am now learning C# after fooling around with Director and Python.

The renderings are mock-ups made with 3ds max and Paint Shop. When zoomed out enough to be able to see enough troops (1,000 to 10,000) the individual soldiers become just a few pixels and all the details are lost. Combine that with my fascination with coloured blocks as used in the Osprey books I have experimented how that would look.

Now I came up with a concept to combine a terrain like from flight simulator and coloured blocks moving over them for either a Napoleonic or Pike and Shot game. Imagine 100 batallions composed of 6-10 company blocks and you will learn to appreciate the true of scale of Napoleonic warfare. I am not fully convinced wether a 1:1 translation of command and control would benefit gameplay. I think game mechanics can be introduced to still allow the player to control all units but also keep the chain-of-command.

Another concept is a strategic Sengoku Jidai game where the player needs to manage loyalty, supply lines, resources (not RTS like, but like how Shingen expanded towards the sea for its salt), fight for territority by controlling castles, etc. Or a strategic Wars of the Roses game.

[Crandaeolon]

Could a battle simulator with a 1:1 scale and historically correct rates of movement and casualities still provide the player with interesting and challenging gameplay from deployment to routing?

Interesting question, probably one that game devs around the world have pondered throughout the history of gaming. In a strictly traditional TW-style real-time format I'd say no - the scale is simply too large, battles too long, with comparatively few events affecting the outcome. But, with an engine properly tailored to such gameplay, I think it could be done.

Such a game would probably be best done in a simultaneously-executed phase-based format, similar to the Combat Mission series. The large scale would demand some kind of time-compression method, and a time-compression method similar to the Total War series is likely not viable because of the scope of calculations required for an accurate simulation; you'd have to take statistical shortcuts instead of modeling every soldier individually.

Combining statistical and individual-level combat resolution is one possibility. You could have the battle engine render a part of the battle with a more accurate resolution system, while calculating more remote parts of the battle statistically. I believe such an engine, even one with 3D presentation with all the bells and whistles, is doable now or at least in the near future - though it alone is probably not enough to help with the time-compression problem.

Individual-level combat resolution on a realistic scale in real time is probably not possible with current hardware. The battle would have to be broken into separately rendered segments. For a historically accurate battle this should probably be some kind of event-based system instead of fixed-length "turns." An event-based system could solve both the time-compression and command & control issues as well - the engine could keep rendering the battle until something "interesting" happens, or until player input is required for new orders, changes of the battle plan etc.

This kind of battle engine could fit without problem into a TW-style strategic system. I don't know how much mainstream appeal such a game would have, or even hardcore-wargamer appeal... but a lot of this would depend on the strategic portion and presentation of the game.

I think game mechanics can be introduced to still allow the player to control all units but also keep the chain-of-command.

Steel Panthers: World at War has a C&C system you might be interested in checking out. The Combat Mission games are useful study material too, pretty much textbook examples of the simultaneously-executed phase-based combat resolution system.

Long post, I'd better stop this rambling now...

[Crazed Rabbit]

Now I came up with a concept to combine a terrain like from flight simulator and coloured blocks moving over them for either a Napoleonic or Pike and Shot game. Imagine 100 batallions composed of 6-10 company blocks and you will learn to appreciate the true of scale of Napoleonic warfare. I am not fully convinced wether a 1:1 translation of command and control would benefit gameplay. I think game mechanics can be introduced to still allow the player to control all units but also keep the chain-of-command.

You could have individual soldiers rendered, and then the whole unit would turn into a block when the camera was far enough away. Able to represent unit formation and cohesiveness well closeup, while keeping it simple from far away.

To me, performance comes before graphics, and since I have a old computer (runs MTW well, but RTW not-very-well), the soldiers, if individuals, would have to be blocky. Simple 3 sided red or blue pyramids (or sprites) representing soldiers would be better to me than unappealing soldiers (ie not as good as RTW units).

To me, a lot of the fun from a large scale simulation such as this would come before any fighting; if the game had limited line of sight and no flying camera - being able to see only what your troops could and from their perspective - and the initial position of enemy forces was unknown, you'd have to send out scouts over hills and through forests to find the enemy, and keep your troops organized so that you wouldn't be flanked or spread out.

Crazed Rabbit

[Duke John]

if the game had limited line of sight and no flying camera

I would need to have a flying camera as I don't have the false hope that I will be able to make an engine that can render thousands of men. Plus a close up view means that all the art must be detailed and that is a real drain of resources timewise. However, that does not mean that you would see everything, there can still be fog of war or line of sight.

For a historically accurate battle this should probably be some kind of event-based system instead of fixed-length "turns." An event-based system could solve both the time-compression and command & control issues as well - the engine could keep rendering the battle until something "interesting" happens

I agree, that is also what I was thinking about as an alternative. And I do like it. Although I would fast forward instantly to the next interesting event, while displaying all the action and movement that happens between events in Osprey style; coloured arrows and comment boxes. Small events would automatically be handled by a doctrine or settings on how to deal with certain situations.