Region Discount ideas

[Zarax]

Well, I'm by no means an expert modder, but I found this thread an excellent source of ideas.

What do you think about placing economical improvements as requisites for certain units?

[WesW]

Well, I'm by no means an expert modder, but I found this thread an excellent source of ideas.

What do you think about placing economical improvements as requisites for certain units?

Reading that old thread again got me to reconsidering the setup I had developed for my mod to encourage economic development in key provinces.
The system I have now uses the two types of Peasants as the goal units, with region bonuses in all the key regions. The Christian unit focuses on trade provinces, and requires a Castle and a Merchant's Guild. The Muslim unit focuses on fertile provinces, and requires a Castle and a 60% farmland improvement. Both units are available to all factions, but their priorities are set to zero, so the AI should never actually build them. I placed them at the end of the file, so that their bonuses never show up on the strat map.

I have given every province a unit bonus, as I feel it adds a nice strategic dimension to the game. Doing this can either help or hinder the AI, depending upon how competently you set it all up. Since the AI will strive for the bonus unit, it won't waste as much time and money on buildings it will rarely use.
The key is knowing the value of the provinces and selecting units whose requirements match the province. For example, pick your knights and other advanced units for your capital and richest provinces. What was so bad about the CA default settings was when they picked a low-tech unit for a rich province, such as Venice or Provence.
From what I gathered by reading the thread, the AI will drop back to standard priorities once it has achieved the ability to construct the bonus unit. If it is still trying to obtain units in other provinces, this can result in a stoppage of construction for some time, which I believe was mis-interpreted as a premanent end to construction.
It would be extremely difficult to tell whether or not the AI prioritizes the second bonus unit once it can build Peasants in my setup, but I feel that the certain economic benefits outweigh the military uncertainty.

[Zarax]

I agree with the principle WesW, in fact I'm using a lighter version of the system you described here...
What I would also suggest is to give to at least the smaller factions a small navy to begin with, as testing suggests that AI has problems in determine the army size, prioritizing unit construction to building one and too often resulting in bankrupt.

[Patron]

I don't think you should alter the game to make it easier for a certain faction, like giving the English kataphraktoi or the Spanish vikings. I think you should aim to make your mod more realistic, otherwise you might aswell play age of empires.

[EatYerGreens]

I don't think you should alter the game to make it easier for a certain faction, like giving the English kataphraktoi or the Spanish vikings. I think you should aim to make your mod more realistic, otherwise you might aswell play age of empires.

I don't think that's quite what bretwalda had in mind. If I read right, he wants to give every province a 'speciality' unit of some description, so that the motivations for conquest of these regions are not solely economic or strategic. For that to work, the special units would have to be accessable to all factions. As you say, this is probably not a good thing...

I presume that there's also a desire to make the game somewhat tougher, in that these regions will stay out of reach for some time and the player will find themself up against valour-bonused opposition.

However, if successful in spite of that, I see the 'snowball effect' being worse though - once the player has his hands on more and more of these bonused areas, the final stages of the game will become more and more of a walkover as the opposition forces lose their access to them and also face gradual economic strangulation.

The in-built inability to demobilise weak early units and replace them with those of higher quality is one of the biggest handicaps to the AI in the entire game, IMO. Other than through battle losses, there are few opportunities for it to do this. In fact, the retreated and/or ransomed bits of army they get back after losing a province is the most likely cause of them getting into negative cashflow situations, from which failure to demobilise means there is no escape.

Notice all those conditions in the unitprod files - situations like 'poverty-stricken' and so forth. These must alter their build priorities drastically.

Having lost a territory, a smallish faction might end up spamming peasants because that's all it can afford, when what it should do is dismiss all the dross and get as many of the toughest fighting units which it can produce out of the remaining treasury - like knights and so forth. The taking of a faction's last province ought to be the toughest you ever fight against them, no?

Regrettably, demobilisation isn't a factor which appears in the moddable files, so it's down to the hard-coding (or lack thereof) that this is the situation.

It would be interesting to know if the AI is programmed to stop mobilisation of army units at a particular level of yearly profit, so that it can still afford buildings, or to save up for expensive improvements, or if it just carries on mobilising until it calculates that the unit which its priorities says it needs has a maintainance cost higher than its remaining cashflow?

Some 'slack' is required so that the retreated army, as above, can be supported for as long as required for it to play its part in the retaking of what was lost.

What the AI really needed, in terms of unit build priorities is a simple formula such as
{number of men in unit} x {fighting effectiveness} / {capital cost/annual maintainance cost}

so that it prioritizes high power units over weaker ones (so long as it has the buildings required to train them) and gets best value for money by penalising units which are cheap to buy but (relatively) expensive to maintain.

If this is exactly the formula currently in use then I'm surprised at the results it seems to give.

The 'fighting effectiveness' term needs some flexibility about it, not something as simple, or rigid, as (attack + defence) scores. When a faction is on the ropes, it needs to prioritise units good at defence, so it doesn't lose any further territory. Once that has succeeded, it needs to switch to making troops good at attack, to take back what it lost. If it's doing well for itself, it should alternate between replenishing the pool of attackers, to take new ground and defenders to hold onto the gains and so on.

[bretwalda]

I don't think that's quite what bretwalda had in mind. If I read right, he wants to give every province a 'speciality' unit of some description, so that the motivations for conquest of these regions are not solely economic or strategic. For that to work, the special units would have to be accessable to all factions. As you say, this is probably not a good thing...
(...)

Not even have to be available to all factions... What I meant was an extension of the province valor bonus system of the vanilla MTW VI - as I noted that some of the provinces have no bonus and some have useless bonus unit.

I really liked XL mod but I was unhappy to see that some of the bonuses were removed... that is how the idea sprung out of my mind...

[EatYerGreens]

I really liked XL mod but I was unhappy to see that some of the bonuses were removed... that is how the idea sprung out of my mind...

I see no reason for not going ahead with your plans, I only interjected to point out that some bonuses can have bad consequences for the AI controlled factions, which may upset the gameplay somewhat. The other thread pointed out how removal of some resulted in better province development. For instance, taking the bonus off Saharan Cavalry in Algeria (or Tunisia?) had the Almos developing properly, setting up the Gold Mine at last and trading properly, giving it a stronger army overall than any amount of bonus SC would ever have done.


I forget where I saw this but there's a feature in the -ian command whereby you can set the whole campaign to run on 'autopilot' - basically all factions under AI control while you just sit back and watch. It will run through the years without user intervention and this vastly speeds up the testing process. I hope I can find this reference again as I can't remember the actual keypress required to set it off.

My idea of a well-balanced mod would be one whereby this kind of test will result in no single faction getting either crushed out of existence, nor running rampant all over the world map, or perhaps a certain amount of to-ing and fro-ing as each factions gets its hands on its latest hi-tech unit at different times.