List of possible AI improvments

[peskpesk]

1) England should always have on Fleet(with just one ship) and a depot on permanent supply duty in England.
2) England will always set their interception and Gibraltar always have minimum of 5I and a depot in city.
3) England should blockade all French non Mediterranean ports + Holland when at war with France.
4) Russia should wait to declare war on Sweden until winter has passed and their fleet is not iced in. Else the war might laps. Enter mainland Sweden before Finland is concurred. A cosak might be handy here.
5) Prussia should never declare war on England (unless they have a big fleet)
6) France should garrison their fleets better to keep England from invading on them and forcing them out to be beaten by the English fleets. This applies to all AI nations when England has a fleet and a corp(s) within seven sea areas from own fleet.
7) English Fleet should use interception when placed in Gibraltar.
8) AI should keep a reserve corp(s) in their mainland to be able to react to sudden declarations of war/invasions and be able to defend the capital.
9) Before the AI declare war it should check it has either

• land access to nation
• or has unblocked fleets and safe passage trough the connecting sea areas to access the nation

10) AI should calculate how hard pressed a nation is before declare war on it.

• A simple formula can be somthing like:
  
  - Number of enemy corps in their mainland
  - Number of lost free states/concurred nation from start of the war
  - Number of enemy garrison in their cities
  - Number of enemy captured province capitals
  - Number of minor nations at war with
  - Number of major nations at war with
  - Number of lost battles, including naval, from start of the war
  - Number of blockaded home ports
  +Approx size of gold reserve
  +Total number of corps/ships
  +Total number of garrisons
  + Number of enemy reinforcements within three month
  + Number of corps within 6 (8 if checking France) areas from the border
  
  + Number of corps in enemy mainland
  + Number of captured free states/concurred nation from start of the war
  + Number of garrison in enemy cities
  + Number of free nations
  + Number of won battles ,including naval, from start of the war
  + Number of blockaded enemy ports
  + Number of captured province capitals
  - Gold reserve
  - Total number of corps/ships
  - Total number of garrisons
  - Number of own reinforcements within three month
  - Number of corps within 6 (8 if checking France) areas from the border
  
  + Winter within three months
  + least Number of areas/or sea areas to enemy capital
  
  The above can be tweaked by adding weight to each number,
  example gold reserve might be weight at 0.5 * Number of gold.
  
  A negative number is nation that is hard pressed and should be easy to win a war against.

11) The AI should no declare war so easy.

• Do the AI have troops over fore the war
• Are they troops close to the border
• How many current enemies does the AI have
• How big is the gold reserve
• Are winter close
• Number of own wars

12) The AI should not surrender so easy.Thing to check before

• How big is the enemy force
• How many own troops can be redeployed soon to stop the enemy
• Are the winter close.
• Number of own reinforcements with in three months.
• Can the AI break the enemy chain of supply.
• How hard pressed is the enemy
• How many province capitals is lost

13) The AI should try only have fleet in ports with big costal defense and garrisons
14) The AI should try to break supply chains
15) The AI should guard own supply chains
16) The AI should expect war if nations builds up troops close to the border

Plz add more!

[Adraeth]

A simple rule might be that the AI should try to make an average of 2 armies (escept Uk): one to protect her country in defense and one to ravage enemy territory seeking THE DECISIVE BATTLE as in Napoleonic war concepts.
 
Example: Russia is at war with Turkey, (both are AI); Russia keeps Benningsen with 5 corps near Moskov-Kiev but launches Alexander with 5 corps in Moldavia to reach Costantinopole; in the meantime Turkey keeps Kushanz Ali in Costantinopole and makes offensive in Podolia or Caucasus (depends on initial deploy).
 
This, i think, might be achieved with different initial deployement and a check on AI trajectories like the pathfinding in other games; so if Turkey is deployed with strong numbers in Caucasus the pathfinding of the AI will check the nearest way to Moscov via Sevastopol and up.... and so on; the initial deploy should be changed by the players thanks to the editor.
 
Good Idea or Bad Idea? [&:]

[prielo]

If the AI's fleet is trapped (blockaded) in a port THAT HAS NO GARISSON, it should raise a garrision of at least one factor AT THE VERY NEXT POSSIBILITY!
 
 
Greets
 
Patrick

[alexvand]

Interesting that you say that the AI should not surrender so easily. Do you mean Conidtional?

I think that the AI needs to learn to surrender faster, especially when you are seeking an unconditional. I sat there sieging Berlin for 5 turns with the French and the AI threw the Prussian army at me every turn. The first turn it was a respectable battle where if they had pulled the right chit they might have won, but after that it was simply a slaughter. In the end the entire Prussian army was wiped out before I finally took Berlin. The AI has still not surrendered. (By the way, Austria, and England are also at war with Prussia.)

A properly timed usrrender can mean the difference between wiinning and losing this game.

[peskpesk]

Correct! Should be 12) The AI should not conidtional surrender so easy.

17 ) AI should surrender faster when faced with opponents seeking an unconditional peace and having limited chance(s) of recapturing the capital.

[Erik Rutins]

We absolutely welcome constructive ideas like these, keep it up! [8D]

[Suvorov928]

Well, I disagree about Russia not declaring on Sweden/Finland at start.  I do this in may games when I am Russia, simply to prevent any other major from doing so before I can invade Sweden.  However, I move one corps into finland, not the red city, to prevent a lapse in war.  I will then invade from my port in April, so I do not have ships at see during an Economic phase.
So maybe the A.I can be programed to move a corps into Finland, to prevent a lapse of war, untill it can reach Sweden.
Of course the A.I. should be somewhat random as well, so that maybe Russia will declare first turn, maybe wait till after winter, etc, so that it is not so predictable.

As far as Prussia, I have seen them decalre war on GB way too often in games I have played.  I think that Prussia and Austria, along with GB, should be inclined to work together against France, way more than fight amongst themselves, especially for the first couple of years, or untill a war has concluded between France and one of these nations.

[Murat]

ORIGINAL: Suvorov928

Well, I disagree about Russia not declaring on Sweden/Finland at start.  I do this in may games when I am Russia, simply to prevent any other major from doing so before I can invade Sweden.  However, I move one corps into finland, not the red city, to prevent a lapse in war.  I will then invade from my port in April, so I do not have ships at see during an Economic phase.
So maybe the A.I can be programed to move a corps into Finland, to prevent a lapse of war, untill it can reach Sweden.
Of course the A.I. should be somewhat random as well, so that maybe Russia will declare first turn, maybe wait till after winter, etc, so that it is not so predictable.

The defense against this is to leave Abo empty so that your winter campaign succeeds and severs Finland from Sweden, lapsing the Swedish war before you can sail and gifting Sweden to whomever won control. If it is France that wins, and the Swedish fleets turn the tide in the naval war expect a very unhappy Britain for the next several years. Russia should hold on this until spring since you need to land in Finland and Sweden at the same time. Just come to Sweden's defense if anyone declares war on it. Britain and France are the only ones that can do it in winter and Britain may win the early battle but they will lose the war. France can win both (if the British fleet does not interfere) but the commitment of forces required against a determined Russian army means they have to give ground in central Europe.

[Suvorov928]

No, becasue if Russia does not move into Abo, then Finland does not surrender.  the only way to conquer a minor, is to occupy the capital city.
Russia mearly has to sit one corps in Sveaborg and wait to invade Sweden.

[Murat]

Missed the 'not the red one' part. Yes if you just sit in Sveborg then you are OK but depending on controlling power you could be facing other problems come spring. Britain can make sure your invasion force never leaves for Sweden, France can make sure that when you get there your force gets pushed back. Both situations commit you to a land route which takes longer. Increased costs for winter supply/foraging added to this leans more towards a spring campaign in my mind.

[carnifex]

The Guard corps is not to be used by the AI as a lone raider.  The Guard corps is not to be used by the AI as a lone raider.  The Guard corps is not to be used by the AI as a lone raider.

[Murat]

ORIGINAL: peskpesk

1) England should always have on Fleet(with just one ship) and a depot on permanent supply duty in England.
2) England will always set their interception and Gibraltar always have minimum of 5I and a depot in city.
3) England should blockade all French non Mediterranean ports + Holland when at war with France.
4) Russia should wait to declare war on Sweden until winter has passed and their fleet is not iced in. Else the war might laps. Enter mainland Sweden before Finland is concurred. A cosak might be handy here.
5) Prussia should never declare war on England (unless they have a big fleet)
6) France should garrison their fleets better to keep England from invading on them and forcing them out to be beaten by the English fleets. This applies to all AI nations when England has a fleet and a corp(s) within seven sea areas from own fleet.
7) English Fleet should use interception when placed in Gibraltar.
8) AI should keep a reserve corp(s) in their mainland to be able to react to sudden declarations of war/invasions and be able to defend the capital.
9) Before the AI declare war it should check it has either

• land access to nation
• or has unblocked fleets and safe passage trough the connecting sea areas to access the nation

10) AI should calculate how hard pressed a nation is before declare war on it.

• A simple formula can be somthing like:
  
  - Number of enemy corps in their mainland
  - Number of lost free states/concurred nation from start of the war
  - Number of enemy garrison in their cities
  - Number of enemy captured province capitals
  - Number of minor nations at war with
  - Number of major nations at war with
  - Number of lost battles, including naval, from start of the war
  - Number of blockaded home ports
  +Approx size of gold reserve
  +Total number of corps/ships
  +Total number of garrisons
  + Number of enemy reinforcements within three month
  + Number of corps within 6 (8 if checking France) areas from the border
  
  + Number of corps in enemy mainland
  + Number of captured free states/concurred nation from start of the war
  + Number of garrison in enemy cities
  + Number of free nations
  + Number of won battles ,including naval, from start of the war
  + Number of blockaded enemy ports
  + Number of captured province capitals
  - Gold reserve
  - Total number of corps/ships
  - Total number of garrisons
  - Number of own reinforcements within three month
  - Number of corps within 6 (8 if checking France) areas from the border
  
  + Winter within three months
  + least Number of areas/or sea areas to enemy capital
  
  The above can be tweaked by adding weight to each number,
  example gold reserve might be weight at 0.5 * Number of gold.
  
  A negative number is nation that is hard pressed and should be easy to win a war against.

11) The AI should no declare war so easy.

• Do the AI have troops over fore the war
• Are they troops close to the border
• How many current enemies does the AI have
• How big is the gold reserve
• Are winter close
• Number of own wars

12) The AI should not surrender so easy.Thing to check before

• How big is the enemy force
• How many own troops can be redeployed soon to stop the enemy
• Are the winter close.
• Number of own reinforcements with in three months.
• Can the AI break the enemy chain of supply.
• How hard pressed is the enemy
• How many province capitals is lost

13) The AI should try only have fleet in ports with big costal defense and garrisons
14) The AI should try to break supply chains
15) The AI should guard own supply chains
16) The AI should expect war if nations builds up troops close to the border

Plz add more!

[1] agree, in London.
[2] in the computer game Gibraltar is less imporatant than in the table top. 1I is as good for holding it as 5I. Unlike corps, garrison are visible and whomever is attacking should always bring enough to cost the defender PPs.
[3] England should set up in the Channel and only blockade fleets that are a threat (have corps in the port). The goal is destruction of enemy fleets and keeping the channel open to you and denied to your enemies.
[4] agreed
[5] well if London is getting occupied and you can get a conditional dogpile on.
[6] yes to the garrison but keep in mind the transports only move 3 so their best factor movers are their slowest.
[7] the British fleet should not be in Gibraltar unless it is preparing for Med invasions which means that no navies remain to threaten the British Isles.
[8] Unless there are no fleets that can threaten the British Isles.
[9] I think the AI already does this.
[10] Certain starting wars require no calculation. AU/PR/BR -v- FR should be the default starting war. BR should have a good chance of going to war with SP if the FR fleet is ever decimated. SP should have a chance of going to war with BR immediately. Good idea on some sort of formula though for other wars.
[11] A formula would seem to help with this.
[12] + [17] A surrender formula would be needed too.
[13] this depends on the nation. France and Spain have their best chances of weaking Britain out in the open ocean and then retreating their fleets into a harbor with guns so that the British losses in running the guns should be higher than the # of ships in the port. But I agree the AI needs to retreat to the port with a garrison that has the most guns.
[14] agreed.
[15] agreed.
[16] agreed.

Each nation needs some individual examination. Maybe take some volunteers (matrix community seems pretty good about those) to come up with ideas for each nation and have another team for overall strategy (default nation AI). BUT allow these teams to be open so they can add people who are interested. Often I see people lock in the team they are using to assist in their product and not replacing people who no longer want to participate.

[Marshall Ellis]

This is good stuff guys! Thanks for the ideas! PLEASE keep them coming!

BTW: I am looking at ways to trim the AI's wreckless DOWs. I am currently placing a lot of weight on allies and who they are at war BUT I am not putting enough weight on who they DOW and how strategically important they are.

For example: Prussia being allied with Russia is good BUT if Russia ends up at war with GBR then Prussia is too easily DOWing GBR in support of her ally. Loyalty is a good thing. Stupidity is not. I should have Prussia understand that GBR is more important to her success than Russia is (Most of the time). I'm not saying Russia is not important to Prussia just that the AI should NOT go blindly into a war just because Russia is at war as well.

[Donegal]

Not an IA improvement but interface improvement. I would like some zoom out button, i really miss it. Other possible screen resolutions could be good too

[Thresh]

One idea was to have each country be given a list of "National Aspirations" with political points bonus's for achieving certain goals.

Todd

[Odysseus]

Could I add that the AI should also act on its wars. In my games I've seen the AI attacking other AI factions allright - GB has had corps alone in Paris (!), PR and FR, FR and AU have had tussles (though AU and PR have not co-ordinated their efforts). But against me, the result has been pretty mediocre. Not a single faction I've been at war with have taken a single step against me, really (including GB, PR, AU and TU). So, when a faction DoWs the player faction, the AI should take steps to get its army together in a nice stack and go try to whoop some ***. AU, for instance, DoWed with the entire army on the other side of the nation. I can't presently suggest any good list of variables to consider here - but something to the effect of moving a sufficient stack (which should be a function of expected opposition) to the theatre of war should do it....

[dude]

One of the things I'd like to be able to do in regards to diplomacy it to cut deals on territory... especially one's you don't have yet.  In playing the boardgame years ago we use to cut deal like...  I'll give you territory A if I conquer it if you get B...  or I'll give you A if you'll attack so and so... Right now I don't know how much if any benifit you get from ceding a territory to an ally.  As GB in my latest game I ceded some territory in the middle of Europe to my ally Austria, partly because I didn't want to defend it and partly hoping that would get me some good play in the Austrian Press [:D]... which I hoped would help me in getting them to lend me a corps or two, or attack or defend someplace later.
 
Is there some hidden scale that rates your relationship with others...?  So far my attempts at using the Diplomacy screen to request corps and other actions has pretty much failed... and I'm tired of them asking for more money while my army is deployed at the front while theirs are just sitting at home doing nothing!  Let's be a bit more reasonable on the requests... every time I ask an ally for help I shouldn't be rejected or asked for more money.  Most of the time with human player we did things that were in our best intrest if an ally asked (and occasionaly if not but would help the war effort at least)... but so far it seems that the AI thinks nothing is in its best interest.
 
Dude

[Jimmer]

ORIGINAL: Suvorov928

Well, I disagree about Russia not declaring on Sweden/Finland at start.  I do this in may games when I am Russia, simply to prevent any other major from doing so before I can invade Sweden.  However, I move one corps into finland, not the red city, to prevent a lapse in war.  I will then invade from my port in April, so I do not have ships at see during an Economic phase.
So maybe the A.I can be programed to move a corps into Finland, to prevent a lapse of war, untill it can reach Sweden.
Of course the A.I. should be somewhat random as well, so that maybe Russia will declare first turn, maybe wait till after winter, etc, so that it is not so predictable.

As far as Prussia, I have seen them decalre war on GB way too often in games I have played.  I think that Prussia and Austria, along with GB, should be inclined to work together against France, way more than fight amongst themselves, especially for the first couple of years, or untill a war has concluded between France and one of these nations.

I agree. In another thread, I lost my mind and forgot about running a corps over to Sweden proper. But, yes, this is a good strategy. Alternately, true diplomacy can sometimes mean this isn't necessary. If GB is honest and says he will allow Russia to take Sweden, then wait until the time is right. It's even more important if GB is asking for help against France as the price, or if Russia is intending to do so anyhow. The reason is that it can be costly to take Sweden, and that simply means GB has to funnel more money over to the Russians. If they're allies, they should attempt to spend as little as possible.

On the other hand, if GB is belligerent, waiting is also better then, too. If Russia is fairly confident that GB will declare war on Sweden (due to failed diplomacy, for instance), then waiting is like gold to Russia. All Russia really has to do is put factors into Sweden himself to help defend it. He'll get the Swedish army and navy to bolster his own forces. Russia will have to declare the war, but that's pretty cheap when accepting Sweden comes with it.

[Jimmer]

ORIGINAL: Murat
The defense against this is to leave Abo empty so that your winter campaign succeeds and severs Finland from Sweden, lapsing the Swedish war before you can sail and gifting Sweden to whomever won control. If it is France that wins, and the Swedish fleets turn the tide in the naval war expect a very unhappy Britain for the next several years. Russia should hold on this until spring since you need to land in Finland and Sweden at the same time. Just come to Sweden's defense if anyone declares war on it. Britain and France are the only ones that can do it in winter and Britain may win the early battle but they will lose the war. France can win both (if the British fleet does not interfere) but the commitment of forces required against a determined Russian army means they have to give ground in central Europe.

But, Russia will almost always fail to capture Finland by March, anyhow. If he tries to take it, he'll be at extreme supply range, and have to spend all his cash. The best course of action for Russia, assuming a January DoW on Sweden is to move one (big) corps 1 space into Finland, and then wait until Spring. If Russia is doing NOTHING else on the map, he can press a little further, but still avoid taking the capital too early.

Now, the defence against THIS is also simple, but by no means guaranteed: Put the Swedish corps, with all factors save 1, on the far eastern end of Sweden. The hope is that it will win the battle or draw, and cause a lapse of war. The diplomatic picture gets REAL interesting at this point. Especially if France got Sweden. 'Cause now GB has something serious to concern herself with.

[Jimmer]

ORIGINAL: carnifex

The Guard corps is not to be used by the AI as a lone raider.  The Guard corps is not to be used by the AI as a lone raider.  The Guard corps is not to be used by the AI as a lone raider.

Why not? That's how Napoleon lost at Waterloo: He put Ney in charge of the Guard with orders to wait until he needed them. Ney sent them into combat 1000-2000 at a time. When Napoleon wanted to "commit the guard", there were few left available to him.

I think this is why Ney's third stat is a 1. At Waterloo, he was in charge of two corps, and didn't do so well.