[AI] AAAHogEx

[AAAHogEx]

version93 released!

・[Change] Revised the calculation of airport maintenance costs.
・[Fix] An incorrect route was created if a new industry started with the same ID after a previous industry was closed.
・[Fix] Excessive CPU usage in valuator function. viewtopic.php?p=1273556#p1273556

[AAAHogEx]

version 95 released!

・[Fix] Rare crash in pathfinder
・[Fix] Updating rails sometimes updates other routes or fails
・[Fix] Trains getting stuck

[xarick]

Found a crash on version 101

[AAAHogEx]

Found a crash on version 101

Thanks as always for the report, xarick.

I’ve uploaded the fixed version to Bananas as version 102.

[Fix] A possible crash when using aircraft
[Change] Improved long-term profitability when inflation and infrastructure maintenance are enabled

[xarick]

AAAHogEx blocks its own ships with depots.

[AAAHogEx]

AAAHogEx blocks its own ships with depots.

Thanks for the report. I’ve released a fixed version, v108.

[AAAHogEx]

version110 released!

[Add] Option to Prefer Reusing Existing Roads

Before building new roads, the AI checks whether existing roads can be used. This helps prevent the map from becoming cluttered with too many roads. However, enabling this option may slow down AI performance and increase the likelihood of traffic congestion.

[GelGiGue]

[Add] Option to Prefer Reusing Existing Roads

hooohoh man, that's something i've been wanting to see forever. the inevitable question this raises is of course, could something like this be done for railways? y'know, to finally have an actual network. if middle stations are off the table, that would certainly make up for it. though the catch to that is probably trying to get the ai to route the lines and signal directions tidy enough to pull that off.

on a completely unrelated note, i've been thinking about the idea to have some sort of option to discourage crazy multi header trains. in some cases it makes sense, but the ai always has a thing for having 1830s planet-etc. or other weak locomotives triple upto sextuple heading trains which, never looks right in early starts. i've also sometimes seen it calculate that the most economical option is to have 8 small tank engines hauling a coal train up a hill.

[AAAHogEx]

hooohoh man, that's something i've been wanting to see forever. the inevitable question this raises is of course, could something like this be done for railways? y'know, to finally have an actual network. if middle stations are off the table, that would certainly make up for it. though the catch to that is probably trying to get the ai to route the lines and signal directions tidy enough to pull that off.

I was thinking the same thing—that track sharing could probably be done with the same system I built for shared roads. If I can overlook issues like track upgrades, speed limits, and trains bunching up, the implementation cost shouldn’t be too high, though the profit performance might end up being poor. That said, it could still prove handy in more complex industry mods such as FIRS.

on a completely unrelated note, i've been thinking about the idea to have some sort of option to discourage crazy multi header trains. in some cases it makes sense, but the ai always has a thing for having 1830s planet-etc. or other weak locomotives triple upto sextuple heading trains which, never looks right in early starts. i've also sometimes seen it calculate that the most economical option is to have 8 small tank engines hauling a coal train up a hill.

It calculate the number of locomotives based on factors like gradients, engine horsepower, and cargo weight, but it might be erring on the high side. I’ll revisit the formula when I get some time.