Quote:

Original post by RPGeezus Can you describe the nature of the simulation you're planning on writing? In a lot of games the concept of flanking doesn't mean a whole lot. It's a term that gets used a lot often without consideration of any tactical merit to the problem at hand. Many of the RTS games I've played do not lend themselfs well to flanking-- with few exceptions. Shogun made flanking possible, but Warcraft didn't. In Risk, for example, there is no flank advantage. Units in one country oppose units directly in another. Front-lines are usually any location where conflict is possible within a given time frame. A concept like supply lines is very much dependant on how you implement them in your simulation. What is a supply line in a game like Risk, or a game like Axis and Allies? Supplies are not really a factor in these games, so you would have to look at supply lines as paths units can take to support each other. You had mentioned that you only found tactical information on the Diplomacy pages. What is the difference between tactics and strategy, to a computer? In the chess world we might talk about strategy being long term objectives, and tactics are the short-term goals we use to get there. In reality, for computer chess, the whole notion often goes out the window and we rely almost exclusively on tactics. The exceptions are cases like opening moves-- I gaurantee though, if a chess programmer were able to tactically analyze their opening sufficintly then they would not rely on an opening book. The further ahead one can look tactically, the closer the decision comes to becoming strategic. I guess what I'm saying is that if you can evaluate a position well enough, strategy can emerge on it's own without you having to worry about explicitly defining it. Sorry I don't have anything more technical to add. :) Will

There was a wargame called Perfect General 2 that came out in the mid 90's. (I'm pretty sure that's the one I'm thinking of, anyway).

There was another game that came out in the late 90's called The Operational Art of War.

One of the features that I remember in both games is that you needed a traceable line of supply back to the main base or you would lose a lot of attack/defensive power.

The concept that you can't simply charge the enemy in a group, without regard for the supply line, is something that's lacking in a lot of modern war games.

The problem is that the AI in both those wargames is a bit ridiculous. They don't understand that they can be encircled, or that they can encircle.

My mis-definition of strategy and tactics in this case is that

Tactics are the decisions that are made based upon one turn and the units that are avaiable to attack. I can see programming an AI that can decide which units are within range and should pile on that particular unit in order to do the most damage. It should actually be able to be fairly skillful at that.

I'm having a more difficult time picturing the strategy, that is, how to defend the supply line, and how to know when to attempt to encircle the opponent, how to allocate troops to best prosecute the attack. All of these things take the knowledge of more than just one turn. Some can be solved by intelligent scenario design, but some things like committing reserves and defending the supply line, are more difficult to do intelligently.

Ralph

As an alternative or addition to checking distances from the enemy to supply lines you can use hierarchical multiresolution approch, using premade map. Divide map into regions(possibly hierarchical), such as that if enemy in the region he is threatening to supply line, to which this region referencing. If all regions have precalculated values for moving from one to another you are basically reduce you map to graph, wich make decisions a lot easier. Someting like pathfinding algorithms, which reduce maps to graphs.

Hi,

1. you could make a map of influence. Armed friendly units increase the index around them, and enemy units (that you know of!) decrease the area around them. And if in you supplyline the value of the influence map is to low, you should expect an attack there.

2. Also you could have your units / Squads have Three areas around them, one in the front, one on each side and on to the back. Think of them as field of view. Then the unit / Squad could check if a friendly unit is in and of the Sectors. If no unit is in a given area, that area is unsave and you might get flankt.


Hope this helps.

I'll start my post by stating that I'm in the US Army. I have sufficient time in service (and enough rank) to have a fairly good idea of why we (in the Army) do things the way we do (strategically and tactically).

The difference between Strategy and Tactics is the separation between Plan and Action. Simply, Strategy is how you intend, overall, to achieve your goals. Tactics are what you use when acting in order to achieve the goals you setup through strategic planning. In real life, strategic and tactical units are separated by echelons of responsibility in wartime.

What you need to first do, before you sink any more time in design or planning, is figure out what exactly are all the elements of simulation you are trying to achieve. Worrying about supply lines means nothing if your units do not need to resupply fuel, munitions, or spare parts. Worrying about units being flanked means nothing if your units can retaliate any aggressor, regardless of position and orientation.

If you want to avoid getting gameplay bogged down in individual unit micro-management (i.e. tactical gameplay), then the model of Risk is what you're looking at, though with possibly added elements of positioning units and setting unit orders (such as defend this position, or move to this position, or attack this enemy unit, etc.) If you want players to have more of a hand in individual unit performance, you're getting into the realm of tactical gameplay, where games such as Command and Conquer model (in their own non-realistic way).

- Pfhoenix

Quote:

Original post by serg3d As an alternative or addition to checking distances from the enemy to supply lines you can use hierarchical multiresolution approch, using premade map. Divide map into regions(possibly hierarchical), such as that if enemy in the region he is threatening to supply line, to which this region referencing. If all regions have precalculated values for moving from one to another you are basically reduce you map to graph, wich make decisions a lot easier. Someting like pathfinding algorithms, which reduce maps to graphs.

Thanks, I'll have to think about that one for a bit.

Quote:

Original post by dragongame Hi, 1. you could make a map of influence. Armed friendly units increase the index around them, and enemy units (that you know of!) decrease the area around them. And if in you supplyline the value of the influence map is to low, you should expect an attack there. 2. Also you could have your units / Squads have Three areas around them, one in the front, one on each side and on to the back. Think of them as field of view. Then the unit / Squad could check if a friendly unit is in and of the Sectors. If no unit is in a given area, that area is unsave and you might get flankt. Hope this helps.

Thanks, I was planning to use influence maps. The problem with that idea is that it's not, strictly speaking, a line, but a set of lines. I'm not sure it's that easy to do in that case, or exactly how to analyze the resulting map to make decisions.

The second suggestion is a tactical one, and while it's probably worth doing, it doesn't help prevent 25% of my forces from being encapsulated.

Quote:

Original post by Pfhoenix I'll start my post by stating that I'm in the US Army. I have sufficient time in service (and enough rank) to have a fairly good idea of why we (in the Army) do things the way we do (strategically and tactically). The difference between Strategy and Tactics is the separation between Plan and Action. Simply, Strategy is how you intend, overall, to achieve your goals. Tactics are what you use when acting in order to achieve the goals you setup through strategic planning. In real life, strategic and tactical units are separated by echelons of responsibility in wartime. What you need to first do, before you sink any more time in design or planning, is figure out what exactly are all the elements of simulation you are trying to achieve. Worrying about supply lines means nothing if your units do not need to resupply fuel, munitions, or spare parts. Worrying about units being flanked means nothing if your units can retaliate any aggressor, regardless of position and orientation. If you want to avoid getting gameplay bogged down in individual unit micro-management (i.e. tactical gameplay), then the model of Risk is what you're looking at, though with possibly added elements of positioning units and setting unit orders (such as defend this position, or move to this position, or attack this enemy unit, etc.) If you want players to have more of a hand in individual unit performance, you're getting into the realm of tactical gameplay, where games such as Command and Conquer model (in their own non-realistic way). - Pfhoenix

First, thanks for serving the country. I did 6 years in the Navy myself.

In the type of game that I'm talking about, logistics has a relatively large impact on units, units out of supply will attack at roughly half-force, won't re-supply ammunition or gas, etc.

When I mentioned flanking, I didn't mean the simple flanking penalties that individual units suffer, that can be largely handled by a competent tactical layer. Once you add logistics into the equation, you have to worry about your overall force in some ways that are obvious to humans. The biggest, to me, is that if the enemy gets between you and your supply base, you are probably in trouble. Which victory hexes to advance on can be determined by scripting, but logistics happens during the battle, and the first casualty of any battle is the plan.

A common gambit used by humans against the AI for games like Civilization is to attack the weakest city. The same thing was true of the RTS games that I've played. That's probably not going to happen with 20th century and later armies because of logistics. Even if you could take the town easily, you would not be able to hold it without being able to resupply the units there.

I think that I am misusing the terms strategy and tactics, but I'm not sure what words can be used to make the distiction between actions at a formation level (prevent being flanked, etc.) and actions at the larger level (send formation X to chase after the enemy that's threatening the supply line.)

From what I've seen, computers do an excellent job of doing things at the unit level, but are horrible at the Strategic/Planning level.

The easiest way to model supply lines and their effects, then, is to make the resupply process physical, in game. By this, I mean, supply trucks. Supply routes. Trucks follow the established (by player) routes. Trucks get to units; units resupply, trucks head back to base to fill up to resupply again.

Modeling that alone will place immense importance on protecting and attacking the supply lines, as there will be something physically there to attack. In the real world, an attack on a supply line doesn't consist of a guy in a jeep, in the middle of a road that's traveled by supply vehicles, going "hey, look, I'm threatening your supply line". It's conducted by attacking the supply vehicles themselves.

Force players to expend resources establishing and then protecting supply routes and the vehicles on them, and you'll have twice the realism of any RTS available right now.

- Pfhoenix

Quote:

Original post by Pfhoenix The easiest way to model supply lines and their effects, then, is to make the resupply process physical, in game. By this, I mean, supply trucks. Supply routes. Trucks follow the established (by player) routes. Trucks get to units; units resupply, trucks head back to base to fill up to resupply again.

It never works in real games. Supply units are too slow. If it take several turns to reach target from depot it usually don't fit into tactical time scale - several turn is a lot in the turn-based game, usually days. If it one or two turns they are no diffrent form supply routes. Also they have pathfinding problems, get lost, get stuck, have difficalties to catch moving targets and falling under the enemy fire. Rerouting making tremendous time lag. They may be realistic sometimes, but no fun for player - too much micromanagement.

Quote:

Original post by serg3d Quote: Original post by Pfhoenix The easiest way to model supply lines and their effects, then, is to make the resupply process physical, in game. By this, I mean, supply trucks. Supply routes. Trucks follow the established (by player) routes. Trucks get to units; units resupply, trucks head back to base to fill up to resupply again. It never works in real games. Supply units are too slow. If it take several turns to reach target from depot it usually don't fit into tactical time scale - several turn is a lot in the turn-based game, usually days. If it one or two turns they are no diffrent form supply routes. Also they have pathfinding problems, get lost, get stuck, have difficalties to catch moving targets and falling under the enemy fire. Rerouting making tremendous time lag. They may be realistic sometimes, but no fun for player - too much micromanagement.

Thanks Serg3D. I wasn't aware of any turn-based games that modeled supply that way. I have played some tactical RTS games that tried to do that, and I always found it ended up allowing totally cheesy tactics. I suppose I should research supplies in wartime and see how they're actually distributed to see if there are some interesting decisions that can be made from this.