Stellaris - Influence cost issues when expanding in clusters while having high AI numbers [3.1.1 b2ed]

[FortunaDraken]

Description
Influence cost issues when expanding in clusters while having high AI numbers [3.1.1 b2ed]

Game Version
Lem 3.1.1

What version do you use?
Steam

What expansions do you have installed?
All

Do you have mods enabled?
Yes

Please explain your issue is in as much detail as possible.
If you play a game where you start in an area where you have no hyperlane attachment to the main galaxy, and the number of AIs is at least 22, the influence cost to take a starbase suddenly skyrockets to maximum even when you own the starbase right next door. Strangely though, the influence costs stay normal is the number of AI is below 20 - I've done quite a bit of testing on it, and AI number was the ONLY thing that affected the influence costs.

This mostly affects modded games with cluster starts, but I have confirmed this does happen in vanilla by using a mod that let me close the L-gates if you own Terminal Egress. To my understanding this is part of the files that modders don't have access to, so it's kind of relying on Paradox help.

The attached save game is using L-Cluster Access in order to display the bug (the L-gates are closed, despite being white)

Steps to reproduce the issue.
1 - Have a large or huge map with at least 22 AI.
2 - Make the L-cluster spawn but do not have open L-gates (I used a mod for this, I assume the devs can probably do it with some dev magic)
3 - Own Terminal Egress and move all population onto a planet in Terminal Egress.
4 - Delete all starbases still in the main galaxy.
5 - Survey a system next to Terminal Egress, then attempt to build a Starbase on it using a Construction ship.
6 - If AI is 22+, influence cost will be maxed despite being right next to an owned starbase

Upload Attachment
File(s) attached

 

[Draconas]

To add some further context to this : it appears that the bug relates to the system not being properly added to the internal system cache. Shown here using an extra-galactic constellation, it can be seen that the effected system with a huge influence cost returns -1 on map cache distance, which i assume is analogous with "null" or some unable to calculate value.

This -1 value persists through saving the game and reloading the save, indicating it is not simply an issue of the thread that manages the cache getting behind. This behavior is new in 3.1.1 where it was possible for the map cache to run slowly on huge galaxy's but it would always catch up. This indicates to me that it is probably caused by the enhanced cluster spawn code (that does substantially reduce the lag)

It also does not always effect all stars: in the screenshots below, the highlighted star Qurus is affected, however the left hand star- biham, has a map cache distance an a regular influence cost. For reference these stars are spawned in the same way as the L-cluster, using the same batch technique.

 

[Pillowesque]

Having this issue as well with any of the "isolated" mods that managed to limp their way to 3.1 -- I'd also like to throw in that even if you console command cheat your way to having control of these systems they will never form a sector.

 

[FortunaDraken]

To add some further context to this : it appears that the bug relates to the system not being properly added to the internal system cache. Shown here using an extra-galactic constellation, it can be seen that the effected system with a huge influence cost returns -1 on map cache distance, which i assume is analogous with "null" or some unable to calculate value.

This -1 value persists through saving the game and reloading the save, indicating it is not simply an issue of the thread that manages the cache getting behind. This behavior is new in 3.1.1 where it was possible for the map cache to run slowly on huge galaxy's but it would always catch up. This indicates to me that it is probably caused by the enhanced cluster spawn code (that does substantially reduce the lag)

It also does not always effect all stars: in the screenshots below, the highlighted star Qurus is affected, however the left hand star- biham, has a map cache distance an a regular influence cost. For reference these stars are spawned in the same way as the L-cluster, using the same batch technique.

Out of curiosity Draconas, since you know more about it than me, could this also be the reason why fleet reinforcement now takes years when it really shouldn't be? Because the distance cache is messed up?

 

[Draconas]

Out of curiosity Draconas, since you know more about it than me, could this also be the reason why fleet reinforcement now takes years when it really shouldn't be? Because the distance cache is messed up?

It could be, if the game thinks that a given system is much further away than it actually is, the time fleet reinforcements are in limbo is calculated by that distance.