• 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.
By frozen I mean (not sure about Kristof73) that the AI is forced to obey preferences set inside .ai files. The .ai files are switched switched by special ai-only events that have triggers written in such a way that even a simple ahistorical move on the map can prevent the AI from switching to new .ai file thus blocking it forever in a try to capture Poland for example.

Look:
Code:
NOT = { war = { country = POL country = GER } }
This condition in the event 20003 that is responsible for switching from switch/GER_Poland.ai to switch/GER_Poland_After.ai means that as long as we have POL at war with GER (and that happens when you evacuate to Cayenne), GER will not go for France. Sad but true...

I think G'Kar understands what I had in mind.

This is of course caused by the attempt to guide each nation historically. I think that AI without all those restrictions would offer a lot of fun, even in a very ahistorical game though the downside is of course that the situation will quickly degenerate from historical setup, just like in CK2.

If you have free time, please experiment with giving each nation a single, simple AI where each other nation would be enemy, each province important, each enemy hated the same. My guess is there would be never a situation in which the AI stops and sits idly, of course I know it can also result in germany declaring war on Tibet. But I think current .ai scripts and events are an overkill as one can never design them to work with every possible situation on the map.
 
Dominik said:
By frozen I mean (not sure about Kristof73) that the AI is forced to obey preferences set inside .ai files. The .ai files are switched switched by special ai-only events that have triggers written in such a way that even a simple ahistorical move on the map can prevent the AI from switching to new .ai file thus blocking it forever in a try to capture Poland for example.

Look:
Code:
NOT = { war = { country = POL country = GER } }
This condition in the event 20003 that is responsible for switching from switch/GER_Poland.ai to switch/GER_Poland_After.ai means that as long as we have POL at war with GER (and that happens when you evacuate to Cayenne), GER will not go for France. Sad but true...
Ah , I see . Japan for example is going to war against USA even China has evacuated to an island or there never was a sino-japanese war. But that false triggers can ruin the game is true

Spocky
 
Maybe this would help

Hey Dominik

Try change all the entries in german ai.event file from

NOT = { war = { country = POL country = GER } }

To :
NOT = { control = { province = 303 data = POL } }
NOT = { control = { province = 306 data = POL } }
NOT = { control = { province = 307 data = POL } }
NOT = { control = { province = 479 data = POL } }
NOT = { control = { province = 480 data = POL } }
NOT = { control = { province = 481 data = POL } }
NOT = { control = { province = 483 data = POL } }
NOT = { control = { province = 484 data = POL } }
NOT = { control = { province = 305 data = POL } }
NOT = { control = { province = 485 data = POL } }
NOT = { control = { province = 486 data = POL } }
NOT = { control = { province = 487 data = POL } }
NOT = { control = { province = 489 data = POL } }
NOT = { control = { province = 488 data = POL } }
NOT = { control = { province = 500 data = POL } }
NOT = { control = { province = 501 data = POL } }
NOT = { control = { province = 502 data = POL } }
NOT = { control = { province = 503 data = POL } }
NOT = { control = { province = 504 data = POL } }
NOT = { control = { province = 505 data = POL } }
NOT = { control = { province = 506 data = POL } }
NOT = { control = { province = 507 data = POL } }
NOT = { control = { province = 508 data = POL } }
NOT = { control = { province = 509 data = POL } }
NOT = { control = { province = 209 data = POL } }
NOT = { control = { province = 208 data = POL } }
NOT = { control = { province = 213 data = POL } }
NOT = { control = { province = 211 data = POL } }
NOT = { control = { province = 210 data = POL } }
NOT = { control = { province = 222 data = POL } }
NOT = { control = { province = 229 data = POL } }
NOT = { control = { province = 562 data = POL } }
NOT = { control = { province = 214 data = POL } }
NOT = { control = { province = 212 data = POL } }
NOT = { control = { province = 212 data = POL } }
NOT = { control = { province = 221 data = POL } }
NOT = { control = { province = 225 data = POL } }
NOT = { control = { province = 226 data = POL } }
NOT = { control = { province = 227 data = POL } }
NOT = { control = { province = 228 data = POL } }
NOT = { control = { province = 231 data = POL } }
NOT = { control = { province = 218 data = POL } }

Basically i have just checked for if POL are controlling any of the Polish Nat. provinces from the POL ai file. Instead of checking if GER and POL are at war.
Iam sure someone can find a way to abuse this by doing something out of the box thinking :rolleyes:

Have fun Gormadoc
 
Thanks! It was a lot work!

Possible abuses: getting claims from SOV or annexing Lithuania :) Then you can hide and wait till the Germans move to the west.

The best would be a trigger like this (short and safe):

NOT = { border = { country = POL country = GER } }

It should check if those two countries have bordering provinces. I hope Johan will add something like this in the future...

Without it we can only use:

lost_national = { country = POL value = 100 }

Without any national province POL isn't a threat anymore too.
 
Last edited:
Dominik said:
Thanks! It was a lot work!

Possible abuses: getting claims from SOV or annexing Lithuania :) Then you can hide and wait till the Germans move to the west.

The best would be a trigger like this (short and safe):

NOT = { border = { country = POL country = GER } }

It should check if those two countries have bordering provinces. I hope Johan will add something like this in the future...

Without it we can only use:

lost_national = { country = POL value = 100 }

Without any national province POL isn't a threat anymore too.

Thats true, more triggers to help us make the ai better would be very appreciated. But i fear it is a long shot.
Since i have included all Polish Nat. provinces getting claims from Sovjet will not help them. Anneksing Lithuania or other neighbors would maybe work. But i think the german AI will continue to try to take those provinces even if the garison ai has been switched to build up forces for an offensive in the westh. Iam more afraid that the ai cant handle both fronts though.

Lost national provinces will have the same effect so that easier to implement.

Gormadoc
 
Well, the AI can fight on many fronts - like west, east for GER or west, egypt for ENG. And continuing the offensive if POL hides on LIT territory is acceptable. I think I'll try a new game with this trigger changed, luckily I have the current save on 2nd Sept 1939 :)

Yesterday I found a bit of free time and run a game with default .ai (my own, not the original one) for every country. Certainly there was much more fun and action :) I guess no AI freezes too :)

http://www.isengard.pl/hoi/ScreenSave1.jpg
http://www.isengard.pl/hoi/ScreenSave2.jpg
http://www.isengard.pl/hoi/ScreenSave3.jpg
http://www.isengard.pl/hoi/ScreenSave4.jpg
http://www.isengard.pl/hoi/ScreenSave5.jpg
http://www.isengard.pl/hoi/ScreenSave6.jpg

I had no time to populate combat = { } lists so without my initial kicks there would be no wars but that can be overcome by spending a bit more time on the ai. So I provoked Allied vs Axis war and Poland vs SOV while they were distracted killing Nat. China :) But that's all. I saw Rep. Spain landing in North Africa, Portugal in France, Canada in Portugal. Czechoslovakia almost destroyed Hungary but Germany intervened :) Sweden and Denmark also were overrun.

The only thing I don't like is that SOV AI didn't return more troops after war with POL, they were all sent to China and SOV-POL border was protected by 1 inf per province (that's why I intervened and declared war) :)
 
palmtree said:
Thanks for the great work. This mod was a much needed boost that helped keep my interest in single-player Doomsday.

yep. Sadly, it seems to be dead :(
 
eRadicate said:
yep. Sadly, it seems to be dead :(

If you are referring to DAIM as dead mod I can assure you that DAIM is still alive ;) .
There is a neww DAIM version in progress d but we are far beyond schedule because of such things like RL .
The new DAIM version for example will have an overhauled Japanese AI with surprising results

Spocky
 
Hmm, my praise seem to have been a bit premature. I played as the US and up until 1941 I was very pleased with the AI in general, I noticed that France put up a bit of a fight, and Japan did very well in China and in Asia in general.

However, two months into Barbarossa the Germans were at the gates of Moscow. I found that odd, thinking the Russians should put up a better fight, so I loaded up the last save with me as a Russian and what I saw was that the Russian AI had completely neglected to reinforce his forces. The IC was split between a build queue of tanks and upgrades, but nothing was going towards reinforcements. There wasn't a full strength Russian division on the front, most of them were half-strength or less.

This is obviously a crippling bug for the Russian AI, that gets worse and worse over time.

Also, one buglet I saw, which isn't game breaking. The British AI has built up 30 divisions to defend El Alamein, but refuses to go on the offensive even though the Italian forces facing them number less than half of their strength.
 
Hmm, weird thing that. I just checked to be sure and yes, SOV is told to use up to 10% of its ICs for reinforcements. Reinforcements should have a higher priority than the build queue (hardcoded), so I wonder why the AI is not using the ICs this way. Note that reloading as an AI nation and taking a look at the build queue can be misleading at times, though. Sometimes there seems to be a problem when there isn't one and vice versa. Ask Spocky about Japanese supply production to get an idea. :D

And yes, DAIM isn't dead. We're a bit in lurking mode only, always ready to jump at those writing good or bad things about it. :)
(Being behind our schedule is my fault for the most part, actually. My conscience is eating me up already. :()
 
I'm wondering if I should reload an earlier game and attempt to salvage things. I played the Russians for a month in September but with every division so severely under strength, and with their few crucial tank divisions needing two months at least before they're fit for battle again I'm not sure if I could even hold Moscow-Leningrad to the winter.
 
DAIM Germany vs. DAIM Soviet Union is unbalanced at the moment, for reasons I don't really understand, to be honest. I'll try to make that better again for the next version.
 
palmtree said:
Also, one buglet I saw, which isn't game breaking. The British AI has built up 30 divisions to defend El Alamein, but refuses to go on the offensive even though the Italian forces facing them number less than half of their strength.

This maybe connected with a supplyBUG I recently found so this should be solved with the next version
;)

Spocky
 
Last edited:
Spocky said:
If you are referring to DAIM as dead mod I can assure you that DAIM is still alive ;) .
There is a neww DAIM version in progress d but we are far beyond schedule because of such things like RL .
The new DAIM version for example will have an overhauled Japanese AI with surprising results

Spocky

Great! HMMM, I think that the Japanese ai is quite good right now. Dont you rather build some normal ai for N.China. They always lose and the handle the Japanese invasions very very poorly. The Japanese are acctually doing a good job againts allies and they are actually conquering things[In my game... they conquered everythig historically and even conquered India and reached to PERSIA! They even landed in Africa... :eek: ]. I love the British ai. Its seems like it is finaly attacking from the shores... for a change... Something is weird with the German ai. It never conqueres Norway from the shores!!! Why??? The Italians are still failing in my game =[. The USA still seems to be a passive nation and they arent doing much. No D day =[. Except the German-Norwegian thing I think everything is now better. Im very glad that you are working on a new version :)
 
National China is too weak ingame to handle the Japanese AI vs. AI. Sometimes Japan doesn't conquer them though.

Germany versus Norway and more reliable D-Day landings are on my list. (Although Norway isn't worth it in the game.)

The performances of DAIM Italy vary from game to game in my experience. I've seen them succeed and fail in Africa.
 
G'Kar said:
National China is too weak ingame to handle the Japanese AI vs. AI. Sometimes Japan doesn't conquer them though.

Germany versus Norway and more reliable D-Day landings are on my list. (Although Norway isn't worth it in the game.)

The performances of DAIM Italy vary from game to game in my experience. I've seen them succeed and fail in Africa.

Hmmm, Im sorry for the stupid question but cant you at least give them orders to use the Chinese factions troops as shore defenders? It worked for me... and I am not a HOI pro trust me...

Norway realy doesnt worth much :( BUT, if the British have it... northern Germany is going to be in big problem... isnt that true?
 
Well, you can't tell the AI to use specific troops for specific purposes. But I'll check the settings for beach garrisons in general.
I'd like DAIM Germany to conquer Norway reliably too, I'll see what can be done. ;)
 
I am just wondering: IF AI Germany succeeds against SU AND UK, will it try to conquer the US? That'd be so awesome to see. And, will the AI react different regarding the aggressiveness? i always play on furious, and i heard the vanilla AI handles Furious and Normal quite differently!