AI is still walking through forts in 1.19

[durbal]

It occurs much less frequently, but the AI still is walking through forts if the fort is mothballed when they begin their movement. Just had the Papal State charge back from another war it was in in Austria all the way through Italy through my fort in Naples (mothballed at war start, activated shortly after) to try to get to the mothballed fort in Messina.

 

[User4035]

Mothballed forts never block movement. Never have.


Blocking movement is extremely valuable which is why forts are expensive to maintain.

 

[Valynor]

Yes, but condition should be checked when the actual transit occurs, not when the walk starts.

 

[Metroidkirby]

Yes, but condition should be checked when the actual transit occurs, not when the walk starts.

You're right, but if there are 0 soldiers in the fort, it's normal that it can't block movement

 

[zdlugasz]

Yes, but condition should be checked when the actual transit occurs, not when the walk starts.

That bug is old as ZoC (I think)

 

[durbal]

You're right, but if there are 0 soldiers in the fort, it's normal that it can't block movement

It had a garrison for two months or so by the time they arrived.

 

[durbal]

Just had a situation where Muscovy walked right past my fort in Ryazan (never mothballed) to get to my other provinces. So apparently there are more bugs that allow this.

 

[durbal]

Gonna leave this here. How can they just walk through Shirvan?

 

[User4035]

Just had a situation where Muscovy walked right past my fort in Ryazan (never mothballed) to get to my other provinces. So apparently there are more bugs that allow this.

oh the bugs are there for sure.
I don't think paradox knows how to fix them which is why they wanted to try that new area system.

I just found one where Novgorod and Muscovy were fighting at start of game, My lithuania decked muscovy and marched into Pskov. Then I couldn't' get out. I got access with everyone around me, thinking that maybe muscovy had access through Livonoian order and then cancelled it which blocked my troops. But then I had to take some long roundabout.

 

[durbal]

I don't understand why this stuff is still going on. It can't be that hard to make sure that when AI issues movement orders that it can actually get to where it wants to go. Whole wars are ruined by this stuff when the AI just decides that it doesn't want to actually siege a fort and would prefer just moving right through it.

I'm playing a Great Horde game right now and Muscovy routinely just walks through a fort in Ryazan whenever it wants. It's creating the illusion that it needs to siege the fort, but whenever I move near the fort it just walks right off into any neighboring province to reinforce ongoing battle.

 

[zdlugasz]

I don't understand why this stuff is still going on. It can't be that hard to make sure that when AI issues movement orders that it can actually get to where it wants to go. Whole wars are ruined by this stuff when the AI just decides that it doesn't want to actually siege a fort and would prefer just moving right through it.

I'm playing a Great Horde game right now and Muscovy routinely just walks through a fort in Ryazan whenever it wants. It's creating the illusion that it needs to siege the fort, but whenever I move near the fort it just walks right off into any neighboring province to reinforce ongoing battle.

This is of course only my personal opinion, although supported by many evidence (responses and explanations from devs)
- this is business for them: cost and return, it is better to do new DLC then correct old one, as fixing old DLCs does not bring new revenue
- in majority cases only critical and simple (in implementation) bugs are corrected
- Eu4 uses legacy=very old architecture and code thus their technological debt is tremendous, and clearing some seemingly simple things are not so simple

result:
- Pdox does not fix bugs which are not directly game breaking and require huge effort (confirmed by Pdox employees long time ago when I asked why some bugs were not fixed 1 or 1.5 year after I have (repetitively) reported it)

My guess partially explaining some of this behaviour: during effort to "optimize" game during Art of War Pdox changed frequency how often many of calculations are performed (e.g. only once per month, to limit load of processor). Issue of ZoC appeared half year later (IIRC) of course, but the modus operandi remains: limit number of operations. As it was mentioned somewhere above, it looks like AI for army checks selected land route only when marching order are issued. Maybe I am wrong, because it worked for straits and terra incognita (but only at the day of arrival), but those might have be cases taken into account during Art of War, as they were known, while ZoC did not exist at that time and were not added later.
And of course ZoC are bugged as hell, especially since introduction of "shortcuts" (which are stupid idea IMO anyway).

 

[DDRJake]

If there are still issues with AI marching through active forts in 1.19 then I would love a savefile where this happens so we can investigate and make further fixes.

Sadly, screenshots do not help in this regard, there is too much that doesn't get shown with just a screenshot.

 

[londoner247]

If there are still issues with AI marching through active forts in 1.19 then I would love a savefile where this happens so we can investigate and make further fixes.

Sadly, screenshots do not help in this regard, there is too much that doesn't get shown with just a screenshot. Feel free to tag me in s

Assuming the issue is forts being mothballed when the move order is given then I'm not sure that this is reproducible in a save file as things like that get re-checked on a reload, don't they?

 

[SolSys]

Assuming the issue is forts being mothballed when the move order is given then I'm not sure that this is reproducible in a save file as things like that get re-checked on a reload, don't they?

Yeah, but if that really is the issue then it should also be easy enough to reproduce [just need to make sure its not something else as well].

 

[Gratak]

Assuming the issue is forts being mothballed when the move order is given then I'm not sure that this is reproducible in a save file as things like that get re-checked on a reload, don't they?

Yeah, but if that really is the issue then it should also be easy enough to reproduce [just need to make sure its not something else as well].

I don't think they get rechecked. And this has absolutely nothing to do with AI, and it seems that the OP knows this. Why did you name the thread like this if your actually complaining about something else entirely?

 

[Kollatius]

Gonna leave this here. How can they just walk through Shirvan?

Looks like they're moving to allied territory so they can always get there...

 

[durbal]

If there are still issues with AI marching through active forts in 1.19 then I would love a savefile where this happens so we can investigate and make further fixes.

Sadly, screenshots do not help in this regard, there is too much that doesn't get shown with just a screenshot. Feel free to tag me in s

Will do.

Looks like they're moving to allied territory so they can always get there...

They were coming from Dagestan.

 

[Kollatius]

They were coming from Dagestan.

But they're you're enemy's ally correct?
If I recall correctly one of the rules for ZoC was:
"Can always enter friendly (read: owned or allied for this particular case) territory"

 

[Rikissa]

Maybe someone can help explain what happened in this case - I control the forts in Mewar and Rewakantha

Vij walked into my territory by way of Baroda - why weren't they caught in Rewakantha's Zone of Control? The fort must've had a garrison, this was a few months after I took it and waited for them to come resiege it

 

[Alerias]

I can at least say with confidence its happening a lot less than before. Its not perfect but its sure a hell of a lot better than either the old system or (IMHO) the one that was thankfully rolled back.