Why sometimes the German AI declares war to Czecoslovakia soon after the UK proclaims

  • 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.
Long ago (vanilla 1.01!) ; the game was a sand box, where everything was possible, such as Germany attacking tcheko just before the first Vienna Award.

It was a outcry , people felt cheated, adn threatened to kidnapt Johann beloved daughter if the game was not made forcefully (stress out on forcefully) historical accurate.

So yes, lua was patched heavily, mostly by one person that spend almost one year to do so (dedication it is not?).

But , if a game is historicaly accurate, then Germany will never ever attack Tcheko or Italy, and Japan will always loose .... so what the point of it if we go to extreme. Problem being a player want Germany to be historical but with some randomness in something (never the same thing per player); so it is hard to please everyone.

Last hing on the AI, it was first made by P'Dox team, then rewrote or amended by a team of moder, then modified par ICE team, then Lothos came to rewrite everything; with weekly release with lot of idea taken from the 2 teams of moders but adapted to the final vision of P'Dox ; so yes it may seems 'erratic' but believe me Lothos here did somehow a master job (and God knows I hate the guy because of some previous work, but here he really showed his qualities and dedication). For exemple the base diplomacy may be by passed in some conditions, so there is still the need for the special modifiers; in some other case, those modifiers will brake the code if they are not present (even if their values are irrelevant) ...
 
Long ago (vanilla 1.01!) ; the game was a sand box, where everything was possible, such as Germany attacking tcheko just before the first Vienna Award.

It was a outcry , people felt cheated, adn threatened to kidnapt Johann beloved daughter if the game was not made forcefully (stress out on forcefully) historical accurate.

So yes, lua was patched heavily, mostly by one person that spend almost one year to do so (dedication it is not?).

But , if a game is historicaly accurate, then Germany will never ever attack Tcheko or Italy, and Japan will always loose .... so what the point of it if we go to extreme. Problem being a player want Germany to be historical but with some randomness in something (never the same thing per player); so it is hard to please everyone.

Last hing on the AI, it was first made by P'Dox team, then rewrote or amended by a team of moder, then modified par ICE team, then Lothos came to rewrite everything; with weekly release with lot of idea taken from the 2 teams of moders but adapted to the final vision of P'Dox ; so yes it may seems 'erratic' but believe me Lothos here did somehow a master job (and God knows I hate the guy because of some previous work, but here he really showed his qualities and dedication). For exemple the base diplomacy may be by passed in some conditions, so there is still the need for the special modifiers; in some other case, those modifiers will brake the code if they are not present (even if their values are irrelevant) ...

I think your point is totally correct, the game is based in historical facts but some variation in the events also based in historical facts is always fun, otherwise you always know what is going to happen and that's not fun in a game like this. The czechs could have gone to war to oppose anexation, so why not if that happens from time to time in the game? However some people pretend to play this game like if they were watching the history channel.
 
Why sometimes the German AI declares war to Czechoslovakia...?

I have made a first version of the German AI by simply modifying the production file and the same with the japanese and italian files, which needed less work, and the result is amazing. Germany now attacks

You must have changed some variables so now Germany is allowed to do so. But according to what you wrote here, AI GER should now be much more aggressive, so I can't see why you see such behavior as wrong...