Military Model V

[Mark_Everson]

I think a combination of disease and logistical problems will give us the needed limits on TF size eventually. This would end up being something similar in character to what's I think in EU.

Richard, I can't find a good description of military supply limitations anywhere. The basics are:

1. Forces need to be supplied with food and other goods depending upon the army size, composition and technology.

2. These supplies are obtained by Merchants functioning as military quartermasters. They buy supplies using funds allocated by player or military AI. Supplies are bought at local markets or shipped, and the merchant needs to transport them to the troops from where they are purchased. No player intervention will be required, other than allocating overall military funds.

3. At player/AI option troops can also try to pillage/requisition supplies from where they are.

4. Because of 2 there will be supply lines that can be attacked, and if troops are surrounded, only local supplies can be utilized.

[Gary Thomas]

Hans Delbrucke has a fair bit to say about the logistic problems of large armies in his four volume history of warfare. There is also a book on Alexander the Great's logistics (I am far from home, so I can't give the exact title).

Essentially, if not fed, the troops went home or died, and the army rapidly became the size that local supplies could support. This is what happened to Alexander's army on the way back from India. In that case they died rather than going home.

On an earlier subject of the defensive effect of terrain, offhand I cannot think of a single battle (aside from sieges) in which terrain type had any effect on the battle. Sometimes one side was able to make better use of the terrain (as the Greeks at Marathon, or the Carthaginians at Cannae), but there seems to be no particular advantage to the defender.

Essentially, ancient battles were almost always by agreement - unless both sides wanted a battle it did not take place. They tended to just look for a nice flat piece of open land, and went at it.

There are a very few atypical examples such as Carrhae. But even here the land was flat and open. It was just that the Parthians were able to force a battle.

Moving slightly out of the period, in the 100 Years War, when the French realized that they couldn't beat the English in the field, they proceeded to win the war by attacking supply lines, but avoiding combat. It took a while though.

Cheers

[LDiCesare]

The defender bonus that id coded stems from a bonus to the scouting phase (15%) which induces a bonus in the manoeuvre phase (around 4%) for a total bonus which exists, but tips fights only if the armies are quite matched.

[Mark_Everson]

Originally posted by Gary Thomas
On an earlier subject of the defensive effect of terrain, offhand I cannot think of a single battle (aside from sieges) in which terrain type had any effect on the battle. Sometimes one side was able to make better use of the terrain (as the Greeks at Marathon, or the Carthaginians at Cannae), but there seems to be no particular advantage to the defender.

For pre-modern 'pure' field battles you may well be right. However I bet some of those sieges would have been field battles if one ignored terrain bonuses to defence. And I think a lot of battles were Not fought given a strong defensive position by one side. One example is Alexander's campaigns in Afghanistan. A less resourceful commander would have lost many of those battles by trying to 'charge the hill' rather than sneaking around it or using sieges. In Clash I think this will result in things like the commander looking at his orders and seeing that he shouldn't attack at less than 1.5:1 (effective) odds. Because the enemy are in hills he will elect to go elsewhere rather than attacking the enemy on terrain of their choosing.

The presence of hill positions with field fortifications has, I expect, changed the character of many ancient battles. I expect to find some good stuff in Roman vs. 'barbarian' battles in Gaul and elsewhere. I don't really want to hit the books now for supporting examples, but will at some point.

And terrain effects certainly need to be used IMO when modern weapons become available. I hope you're not implying that the Italian campaign in WWII would have turned out the same way if Italy were a big flat plain.

Although it may not be in there yet, I think a terrain bonus for the tactical defender is Really something we should have. A combinations of playtesting and matching reality can determine the size.

[LDiCesare]

I think that field defenses can be modeled in 2 ways:
1)There are engineers who can build defence during the manoeuvre phase if they have manoeuvred correctly.
2)The unit has been entrenched for some time, which can be modeled in the same way as the fortify order in civ.
Both ways are cumulative, and can model war in the open.

Now sieges:
Siege warfare should be handled differently, as most sieges ended with one of the armies leaving. Btw, Cesar bet Vercingetorix at Alesia in Gaul because Vercingetorix was besieged in a foritified camp in hills, and he surrendered (while reinforcements were near, but he didn't know it).
I suggest that sieges be handled totally differently from open warfare. The whole point in sieges is to cut supply lines, not to fight. Even in WWI, Verdun was besieged and, although the various forts changed hands several times, the siege failed mainly because a supply road remained between the city and the rest of the country.
Sieges would deal hardly any damage per se, but if we have merchants and loss of units through food loss etc., the defenders (and attackers) would lose their units thru attrition.

[Mark_Everson]

Hey Laurent:

Field defences: I am with 1 and 2 above. But there are also last-minute field defences. As little as a few hours time can make a big difference, especially in gunpowder-weapon fights. These are Much more effective on broken and hilly terrain, and that is one of the origins of the "terrain defence bonus" and why I want to keep it. Fights in the American Civil War were frequently decided by who was on the tactical defensive. Against the firearms of the day, being able to shelter behind things like low stone walls if available gave the defender a critical advantage.

Sieges: I agree sieges are different. And we need to discuss how to handle them. But that wasn't my point...

My point about sieges was that terrain defensive bonuses are one of the things that Lead to sieges a la Vercingetorix. That is sieges happen when the enemy position is too strong to attack without unacceptable losses. If Clash has no terrain defensive bonus my guess is that in the game the Romans would just attack him in his fort rather than besiege. IIRC it was the combination of the fort, on a hill, that led to the siege. But I'm of course guessing

There's a rule of thumb I've read that field battles tend to happen when real powers of the forces almost always are in the range 1:1 to 2:1 odds. Only when things are close are generals willing to risk their precious troops in a battle. If odds are worse, the general with smaller force will attempt to find solid defences and will be besieged rather than fight. Of course this rule of thumb doesn't work after about the Franco-Prussian war because with modern mobilization there is frequently noplace big enough to hide.

[Gary Thomas]

The defender bonus that id coded stems from a bonus to the scouting phase (15%) which induces a bonus in the manoeuvre phase (around 4%) for a total bonus which exists, but tips fights only if the armies are quite matched.

I like that.

In Clash I think this will result in things like the commander looking at his orders and seeing that he shouldn't attack at less than 1.5:1 (effective) odds. Because the enemy are in hills he will elect to go elsewhere rather than attacking the enemy on terrain of their choosing.

He doesn't have a choice in Clash - if opposing armies are in the same square, there is a battle.

And terrain effects certainly need to be used IMO when modern weapons become available. I hope you're not implying that the Italian campaign in WWII would have turned out the same way if Italy were a big flat plain.

I did say ancient battles.

Although it may not be in there yet, I think a terrain bonus for the tactical defender is Really something we should have. A combinations of playtesting and matching reality can determine the size.

Why, if it isn't realistic?

I think that field defenses can be modeled in 2 ways:

Field defenses were very important in ancient times. The most famous example is the Roman fortified camp. Other nations also fortified their camps, but not so religiously.

But there are also last-minute field defences. As little as a few hours time can make a big difference, especially in gunpowder-weapon fights. These are Much more effective on broken and hilly terrain, and that is one of the origins of the "terrain defence bonus" and why I want to keep it.

I am in the throes of getting the hard-coded terrain data into an xml file (with great originality, I thought I would call it terrain.xml). This is what made me start thinking of defensive bonuses. Essentially, my reasoning went down the track that movement effects are troop type specific, so perhaps defensive bonuses should be. I can see a rifleman getting a defensive bonus from broken ground, but I cannot see cavalry doing so.

Anyway, my proposal is that defensive bonuses be treated the same way as movement, on a per element basis, by troop type. This can then directly affect their defense value. It also fits in with existing code.

It may be necessary to take dispersion (or whatever it is called) and mobility into account.

On a related matter, I strongly feel that there should be a means of giving the player this information, in some detail, so they know why their "fortified" horse archers got cleaned up by inferior skirmishers in a swamp.

Also, I think it should be possible to give a command orders to avoid battle in a square. If both sides do that, there is no battle, they just co-exist, uneasily, in the square.

Cheers

[LDiCesare]

Anyway, my proposal is that defensive bonuses be treated the same way as movement, on a per element basis, by troop type. This can then directly affect their defense value. It also fits in with existing code.

Actually, this is more or less in the code . I gave a defense bonus to terrains, and this bonus is actually a penalty to attack strength. The penalty is applied twice for non-foot units, to simulate infantry superiority in harsh terrain. Having a per unit type bonus/malus is of course even better, but may be overkill considering the variety of units we have.

last-minute field defences

This applies only if the army is indeed fortifying from my point of view. Just giving fortify order would give a defense bonus, which could be more important depending on terrain type.

He doesn't have a choice in Clash - if opposing armies are in the same square, there is a battle.

There are actually choices, one is morale, which will make the army attempt to flee if outmatched, and the other one, more important, is the AI, which may decide not to attack square X next turn because they are not strong enough.

[Mark_Everson]

Originally posted by Gary Thomas
>
In Clash I think this will result in things like the commander looking at his orders and seeing that he shouldn't attack at less than 1.5:1 (effective) odds.
>
He doesn't have a choice in Clash - if opposing armies are in the same square, there is a battle.

No, he doesn't have much of a choice as the code currently works. And there is Laurent's point. But more broadly, a demo 4 feature that hasn't gotten back in that I think we need is the ability to give overall orders on what odds attacks should be aborted. I think this is critical to avoid a lot of military micromanagement. Orders for an attack could include "attack regardless of odds" but I think the default behavior should be that attacks can be aborted if the odds are poor, with poor defined by the player or AI. This would simulate the field commanders orders and best judgement, and I think its very important.

I can see a rifleman getting a defensive bonus from broken ground, but I cannot see cavalry doing so.

Cavalry have this Really Tricky manuver they can do when they need to hold a position, its called Dismounting . I agree bonuses should be different by unit (or element) class, just don't necessarily agree they should be required to be zero. But this may be a distinction best left out of the game.

On a related matter, I strongly feel that there should be a means of giving the player this information, in some detail, so they know why their "fortified" horse archers got cleaned up by inferior skirmishers in a swamp.

Agree. If its a really important consequence of terrain or whatever, it should be in the "Battle Description" the player can read in the newspaper, or whatever DisplayEvents evolves into. We can also provide detailed numerical analysis at some point as a debugging tool, and feature for those who Really want to get into the numbers. Detailed descriptions are key IMO to the player getting into the military system.

Also, I think it should be possible to give a command orders to avoid battle in a square. If both sides do that, there is no battle, they just co-exist, uneasily, in the square.

Yeah, by explicit order, or by the "don't attack unless at good odds" type I discussed above.

[Gary Thomas]

I have been refactoring (sorry, Mark) the movement effect code. In doing so, I have managed to move a lot of the calculation stuff to the setup stage, so each BaseTerrain will point to a TerrainEffect object which will immediately give movement effects, defense bonuses and road, railway and canal building costs for that type of terrain.

The terrain effect data will be read from a terrain.xml file, instead of hard coded.

This is nearly finished. Reading a map from a file will follow shortly after.

Cheers

[colorrr]

I think that a gradual defensive terrain utilization could be achieved. Let's say that all units can always use defensive terrain. Then different units would have a distinct 'dig-in' value.
The early military units ability to dig in would be so crude as to make battle effectively take place on an open plain.
As units advance, so can also their ability to dig-in, making them stronger in defence.
With the modern area rifleman, they would be far outmatching their stone age warriors in utilizing the defensive terrain (and constructing defensive positions, overlapping fields of fire, bunkers etc.).

[Mark_Everson]

Hi colorrr, welcome to Clash!

That should work, although even for ancient units I don't think its necessarily minimal. Gary's already mentioned the Roman fortified camps, which are a good example of non-negligible digging in.

Gary: Sounds worthwhile. I figured that was part of the xml refactoring anyway

[colorrr]

Thanks for the welcome mark :-)
I look forward to contribute something to the project.

About the Roman legionares. Well, they would be much more advanced than stone age warriors that would be pratically hopeless in taking advantage of the defensive terrian (No brain, no pain).
So the legionares should be better in taking advantage of the terrain. Did you ever see the 2 part movie about Attila the Hun?
In the end the romans make a stand on a hill, slaughtering the huns as they storm the hill, stopping the hun advance for the winter. On the open the huns would have beaten them.

The fortified camps of the legionares I think should be pictured in another way. These can be built everywhere, on open plains if you like, and if we want to include them in CoC, I think they would be better pictured as the military units being able to build the camp.
Or if they stay in place 1 turn they construct one automatically.
Another simpler way would be to just raise their defense value.

[Mark_Everson]

Hey colorrr:

I look forward to your contributions too! Every extra worker accelerates our progress, and we'll have a real game that much faster.

I think just giving the Roman legions a defense bonus for their fortified camp approach is the way to go. This will tie into their orders. If their orders are "quickest advance" then they wouldn't build the camps, and not get the bonus. Never saw the movie, sounds like a good one!

[LDiCesare]

The fortified camps of the legionares I think should be pictured in another way.
These can be built everywhere, on open plains if you like, and if we want to include them in CoC, I think they would be better pictured as the military units being able to build the camp.
Or if they stay in place 1 turn they construct one automatically.
Another simpler way would be to just raise their defense value.

I think the fortify order as in civ represents this quite well.
Additionnally, in the current code, legion units have engineer elements which are used to put up fortifications at the beginning of a fight.