• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

unmerged(255)

ho Mixobarbaros
Aug 27, 2000
1.730
2
Excuse my temper, but after playing several recent GCs it's quite obvious that England has to be tweaked further. I basically cheat just to keep'em float, yet no matter how much cash you send them (well over 5000)it almost seems like the English AI has been programmed to loose -at the very least- the Marches. What's even more frustrating is watching the wars between England and Scotland--while a 50 thousand strong English army is having tea in Bristol, a small Scottish detachment takes over the Marches...and the territory immediately changes hands.

Another problem with the English AI is that it doesn't prioritize when it comes to colonizing North America. Tobacco, cotton and fur producing lands are left largely untouched whereas hundreds of British colonists flock to Micmac and other 'crappy' regions to fish and mine salt...

Maybe it's just me but England was doing alot better before
IGC 2.3 was released...

tuna
 
Another problem with the English AI is that it doesn't prioritize when it comes to colonizing North America. Tobacco, cotton and fur producing lands are left largely untouched whereas hundreds of British colonists flock to Micmac and other 'crappy' regions to fish and mine salt...

The problem here really has nothing to do with the IGC. if you want to see crappy colonization watch a GC game :) This can be helped a little bit by changes to the AI file. Some people are already looking into this. There are also game engine economic issues which help eng do a crappy job of colonizing.

ErrantOne
 
I've encountered the "having tea" problem with lots of countries, though England is one of them. The AI will simply hang out with a superior army while the enemy ransacks several of the border provinces. I think part of it might just be historical fortune. When you think, for example, of how close the English came to collapse from the Spanish Armada, you realize how vulnerable the country was in the 16th century. English survival and victory was probably not guaranteed and seeing the English chopped up is reasonable, though I hate to see it happen.
 
England is just annoying-that AI could probably lose provinces if you gave them level 6 fortresses from the start!

The AI should really have to cycle through its idle armies more often. It seems it only does this at the moment when one of its army's captures a province or when another war is declared involving it.
 
You see all countries with idle armies. It seems the ai decides to take some provinces and then it will sit. It won't do anything else until one of its provinces has been taken (it will take it back), or a country takes back a province that the ai had previously taken. It will then go after that province again.
 
It's quite a flaw though, these idle armies. Reloading does make them move, but it often also causes some other army or AI country to break off a siege, which they were actually sensibly undertaking, or cause an army that's attrited away across a dozen provinces to get to a target to turn round in the direction of somewhere else half a continent away!
 
It's all part of the crude military ai. Amongst other things, the ai is programmed to avoid combat and prefer sieges. If you move your own army/navy several provinces, note that it will route around enemy forces, unless you force the route the way you want it. AI countries would execute the move avoiding combat.
 
True-this is why I usually subdivide long moves as the odds are good that the AI army will have moved out of the way by the time your army would have got to that province. Although this habit can be quite useful when the AI plots your sea routes early in the game....hmmm, why is it dodging that sea zone......Ah, enemy ships-kill!:)
 
Ive always thought that huge problems like developing good AI could be solved if somehow the AI code was seperate and independant from the gameplay, and could be tweaked and improved by the users. Instead of AI hard wired into the game, AI becomes a player whose decisions lie based on rules in AI rule file(s)/script(s). This way over time serious players will add continous improvements, which the central company could compile over time and release updates.
 
Sounds like a good idea, but I'm no programmer. Anyone out there know if this would really work?

Not trying to insult you Pythagoras, but a second opinion is always useful with such a radical suggestion!:)

By the way-on another thread I'm starting to hear stories of decent England performances-I'm stilll in a state of shock!
 
Originally posted by Pythagoras
I set up a thread in the wish list forum "Improving AI"


Must go take a look.

My latest effort to improve England involves beefing up their army on the border with Scotland-they've managed to extract some money from the Scots for once, but that may be due to the way the war began. Oh well, we'll see.:)
 
I often get England receiving indemnities from Scotland. Usually in a quick war where they take the Scottish capital and gett hem out of the war. In my current game so far the English have taken Strathclyde!