AI is still walking through forts in 1.19

[Gratak]

Do a quick and easy test.

Enter a fort ZoC. Try to move to an adjacent province in that fort's ZoC. You are forced to move back to the original province first and then back into the ZoC. The AI doesn't have to do this.

Ok. Took me 5 minutes playing without trying to do something special to find an example for you. Hohhot is only captial fort, so Ereen is their only ZoC fort. As you can see if you compare the dates, my army had not chance of walking to khara and back.

edit: btw although you did not ask for this, the original case was actually whether I would be able to walk from Hohhot to Ordos. Here's the screenshot for that (a little later since I only just remembered, but I actually walked that way to occupy Ordos):

edit 2:
And another one. Walking through the fort. This guy came to dornogovi from the west!

 

[Viperswhip]

I just RP that I was betrayed by a local guide or unhappy peasant who lead the army on secret roads. Normally I am killing the AI, so they rarely get into my lands for me to test this bug these days, but I know it happens once in a while. It has happened in real history that enemies went around defenses in unexpected ways, just roll with the punches until they fix it, if they fix it. It's still slightly better than before when we could just march everywhere (pre forts).

 

[durbal]

Yes. That's the point. But you will see that you can move almost freely within the ZoC because of shortcuts.

Here we have an army in Pest that originated in Somogy.

Once in Pest, it needs to leave the ZoC through Bekes to get to Szeres.

But to get to Pozsony we can go directly (presumably because it is adjacent to my friendly province).

 

[Gratak]

Here we have an army in Pest that originated in Somogy.

Once in Pest, it needs to leave the ZoC through Bekes to get to Szeres.

But to get to Pozsony we can go directly (presumably because it is adjacent to my friendly province).

Good that you ignore my screenshots. And you contradict yourself even with your own screenshots:

Patan is in the ZoC of the fort in Mewar, so armies entering Patan can only go back to their original province or to the fort in Mewar

Coming from Somogy you entered the ZoC and can still leave it via Bekes, which is far beyond any friendly province.

And I never said that everything is possible. Just a lot more than enter ZoC and leave it the same way or siege the fort.

 

[durbal]

Good that you ignore my screenshots. And you contradict yourself even with your own screenshots:

Coming from Somogy you entered the ZoC and can still leave it via Bekes, which is far beyond any friendly province.

And I never said that everything is possible. Just a lot more than enter ZoC and leave it the same way or siege the fort.

Yep, I didn't realize that enemy provinces adjacent to friendly provinces don't apply a ZoC block. This is consistent in all of your screenshots. It's not a shortcut, it's just normal rules being applied.

Also just a friendly reminder to please tone down the attitude. This isn't debate club -- there are no points to win -- it's just a discussion forum.

 

[Gratak]

Yep, I didn't realize that enemy provinces adjacent to friendly provinces don't apply a ZoC block. This is consistent in all of your screenshots. It's not a shortcut, it's just normal rules being applied.

Also just a friendly reminder to please tone down the attitude. This isn't debate club -- there are no points to win -- it's just a discussion forum.

And how do you explain being able to walk to Bekes?

And yeah, I just hate that people again and again bring the same incorrect arguments instead of complaining about the stuff that is actually a problem. And the main problem about this fort system is that it's utterly unpredictable. Coupled with a few cases that are utterly "unrealstic", like forts at narrow points sometimes not blocking (and when 1.18 was newly released I posted an example where I was able to do something like that myself, too lazy to search that now). And I wouldn't be surprised if the only two reason everyone says this is an AI problem are the following:
1. Bias. You wouldn't even try some strange stuff and even if you do, you will remember the one time the AI killed you because of an unexpected behaviour, not the one time you had an advantage because of it.
2. The AI can probably test which military access could help them and take whichever they need at any time.

And btw: I did play 1.14 which was before they stealth introduced the shortcut rule (or at least making it apply to more cases). Try playing that and I'm pretty sure you will feel the difference in where you can walk (not sure about AI).

 

[durbal]

And how do you explain being able to walk to Bekes?

Bekes isn't in a zone of control.

And yeah, I just hate that people again and again bring the same incorrect arguments instead of complaining about the stuff that is actually a problem. And the main problem about this fort system is that it's utterly unpredictable. Coupled with a few cases that are utterly "unrealstic", like forts at narrow points sometimes not blocking (and when 1.18 was newly released I posted an example where I was able to do something like that myself, too lazy to search that now). And I wouldn't be surprised if the only two reason everyone says this is an AI problem are the following:
1. Bias. You wouldn't even try some strange stuff and even if you do, you will remember the one time the AI killed you because of an unexpected behaviour, not the one time you had an advantage because of it.
2. The AI can probably test which military access could help them and take whichever they need at any time.

Yeah, some of the weirdness explains quite a bit. Whenever I get a weird situation so far I realize it's some odd obscure rule at work doing strange stuff. Like for example if you declare war on Muscovy as Ryazan -- you can enter Movska from Vladimir and then walk right through the fort in Movska into Kaluga because the fort doesn't apply ZoC due to adjacent friendly provinces.

 

[Gratak]

Bekes isn't in a zone of control.

But Pest is. By default ZoC rules you should only be able to leave the ZoC via the way you came. Shortcut rules allow you to go the direct way.

 

[durbal]

But Pest is. By default ZoC rules you should only be able to leave the ZoC via the way you came. Shortcut rules allow you to go the direct way.

That's not what I'm talking about with shortcut rules. The AI can sometimes go places that player cannot. I'll try to get a save where attaching my army to an allied AI will allow them to pass through forts while I can't when unattached. This used to happen in 1.18 but I'll see if I can get it to occur in 1.19 as well now that one of the AI bugs was fixed.

 

[durbal]

A few more things to report:

Occupied forts don't apply ZoC

This is untrue.

Here I am trying to get to Golestan:

I should be able to walk right through the occupied for in Mazandaran. But I can't. It's making me go all the way around.

Also, the previous point about 'it's always possible to get to allied lands' is also untrue as shown by this same screenshot.

Then how does this occur from a previous war?

Here are the Timurids walking from Golestan through Mazandaran (an enemy they're at war with) to Qavzin.

Here is the save game file (one while it is occuring, the other from two months or so prior).

What is going on? The AI is obviously taking weird shortcuts to places it shouldn't be able to go as shown above and in my previous screenshot where they went from Dagestan through a fort in Shirvan.

 

[durbal]

Here's another savegame where Scotland is walking right through the fort in Shirvan.

 

[Gratak]

A few more things to report:



This is untrue.

Here I am trying to get to Golestan:

I should be able to walk right through the occupied for in Mazandaran. But I can't. It's making me go all the way around.

Also, the previous point about 'it's always possible to get to allied lands' is also untrue as shown by this same screenshot.

Then how does this occur from a previous war?

Here are the Timurids walking from Golestan through Mazandaran (an enemy they're at war with) to Qavzin.

Here is the save game file (one while it is occuring, the other from two months or so prior).

What is going on? The AI is obviously taking weird shortcuts to places it shouldn't be able to go as shown above and in my previous screenshot where they went from Dagestan through a fort in Shirvan.

1. Occupied forts apply ZoC. Yes. Thanks for the proof.
2. Always going into friendly land has never been possible since 1.14. It's the other way round: You can get anywhere adjacent coming from friendly land.
3. Certainly in the unpredictable category. But I've also shown a screenshot above where I was able to walk right through an enemy active fort.

Here's another savegame where Scotland is walking right through the fort in Shirvan.

Not clear where allied, neutral and enemy land is but you also have mothballed forts around. You sure that fort was not mothballed when Scotland gave the move order? Considering that Scotland has quite a way to walk to you, that may have been many month ago.

 

[durbal]

Not clear where allied, neutral and enemy land is but you also have mothballed forts around. You sure that fort was not mothballed when Scotland gave the move order? Considering that Scotland has quite a way to walk to you, that may have been many month ago.

It would've had to have been issued 2+ months prior. It's possible.

That doesn't explain the Mazandaran or the previous Shirvan screenshot from the QQ war though.

 

[Gratak]

That doesn't explain the Mazandaran or the previous Shirvan screenshot from the QQ war though.

Correct. I commented on this one seperately:

3. Certainly in the unpredictable category. But I've also shown a screenshot above where I was able to walk right through an enemy active fort.

 

[Zelius]

Here we have an army in Pest that originated in Somogy.

Once in Pest, it needs to leave the ZoC through Bekes to get to Szeres.

But to get to Pozsony we can go directly (presumably because it is adjacent to my friendly province).

No, no. You can walk directly from Pest -> Pozsony because Pozsony borders Sopron, and Sopron is adjacent to Somogy, your Return Province. Nothing to do with adjacency to friendly provinces.
AFAIK a similar rule allows you to exit ZOC at Bekes but not the province north of it (without any shortcut rules, you should only be able to exit at the province you came in from).

 

[durbal]

Here's another savegame. In this one, Saxony is going to try to siege the garrisoned fort in Braunshweig. When they are about to be attacked, they will walk through the fort to avoid the battle.

Can a dev please acknowledge that this is going on still? We're not crazy. The AI can walk through forts (especially in narrow places) using the shortcut rule.

 

[FB5813]

Here's another savegame. In this one, Saxony is going to try to siege the garrisoned fort in Braunshweig. When they are about to be attacked, they will walk through the fort to avoid the battle.

Can a dev please acknowledge that this is going on still? We're not crazy. The AI can walk through forts (especially in narrow places) using the shortcut rule.

I'm not going to claim I understand the rules of forts at all, but just based on playing lots of games and trial and error I don't know why you'd expect that fort to block anything. From your territory there are quite a few fort combinations I'd expect to do something such as:

1) Hoya and Braunshweig, Cassel or Thüringen should stop movement to Luneburg from west to east
2) Braunshweig and Hessen or Nassau should stop movement to Thüringen from west to east
3) Thüringen and Hessen or Nassau should stop people moving to Braunshweig from south the north
etc. etc.

...and depending on the military access of the war participants I wouldn't guarantee any of these either. Best off deleting your forts in such an awkwardly shaped empire.

That said, this obviously doesn't help in clarifying the rules at all. And I could have picked any number of posts in this thread and posted a similar thing - nothing personal at all. People should just stop expecting single forts to do anything. [Fort - province - province - fort]^n along your border in a nation at least 2 provinces deep is the way to use forts without getting bajanxed by mystery rules. Expect this and only this to work and your fort headaches will disappear!

 

[Yxklyx]

Occupied forts don't apply ZoC

Unless they are controlled by the AI! I'm certain that my armies are affected by my captured forts.

 

[Gratak]

Unless they are controlled by the AI! I'm certain that my armies are affected by my captured forts.

They do. For player and AI. But as usual the rules are so complicated that you often do not realize why an army was not blocked... I guess in this case it is because a fort does not project his ZoC into foreign land. The Vija fort does not project into the players land, even if it is currently occupied by the player. You would certainly have the same situation if player and AI were interchanged. But the AI would not post on the forum complaining that the player cheated.

 

[TheMeInTeam]

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.

In principle, if the rules about forts are clear in the UI and applied consistently, it should be possible to identify if and when they are not obeyed.

The rules as presented being what they are, it's equally reasonable to treat exile walking into carpet sieging and being otherwise incapable of moving to a fort you occupied because you occupied it as both "WAD". How are people supposed to identify bugs and save consistently when it's hard to tell the difference? Getting "blocked by hostile fort" when there is no fort at all and stack wiping the enemy on the spot is almost certainly a bug. Beyond that, why is it not a bug to get trapped from occupying a fort? What ARE the rules now, after patches changed them but didn't explain them?

For the player base to reasonably make decent bug reports, it is important to explain how or why a mechanic is deviating from how it is expected to work. Usually, the difficulty is in explaining that deviation, but in this case there is ALSO serious issues with "how it is expected to work". It's not like you can make a reasonable estimation in advance using the game UI alone, absent trial and error. Since it's been around over a year should we just treat intentional exile to carpet as intentional design? Why not, when random military access allows nations to cross areas that would otherwise have ZoC because they could in theory have gone around?