Bug Reports and Technical Issues

Yes, it's intentional.
 
I previously reported only Spain being hesitate to use their starting settler but have since seen symptoms of a broader force in effect. Spain, Swahili, and Java all are currently not settling at all.
 
Of course what helped the Americans disrupted a couple of other places in the world. If that is not the story of our time.

I am looking into it, perhaps I can also help the settler AI overall.
 
What's funny about this is that Java, Spain and Swahili each seem to have different causes. Haven't looked at Spain yet, though.
 
Leoreth said:
I am looking into it, perhaps I can also help the settler AI overall.
Click to expand...
Ok so have run more autoplays the scope of this issue goes well beyond my previous post; so yeah I think the overall approach is the place to start.

For now I can offer a clue that I hope helps crack the case: once 1800AD hit, Spain suddenly founded Santiago with the starting settler and for the first time began production on a new settler which went on to found Alicante. Also, as far as I could infer from plain turn eye witnessing and the Statistics screen (in particular, no Galleons lost despite centuries of war with naval rivals), they were not spawning Colonist unit groups. I less confidently suspect this was also the case for England and France. Portugal and Netherlands seemed unaffected and carried on as they have been.

EDIT: In one play, Swahili eventually used their first settler to found Mahajanga in 1595. So far no cases of settling their typical home area cities though.
 
Last edited:
SultanRedSnake said:
they were not spawning Colonist unit groups
Click to expand...
EDIT: Scratch that. It looks more evidently to be the case that the civs in question did receive their first Colonist group spawn but then just parked it at home which, if I understand the script's basics, would preclude further spawns to proc? Some CTRL+hovers have turned up loaded Galleons remaining in place with the MISSIONAI_PICKUP flag with current position as target plot.
 
Last edited:
Is El Escorial bugged? I have 9 gold and silver combined, but it only provides +10 moneys.
Spoiler :


Unbenannt.png

 
SultanRedSnake said:
I previously reported only Spain being hesitate to use their starting settler but have since seen symptoms of a broader force in effect. Spain, Swahili, and Java all are currently not settling at all.
Click to expand...
Leoreth said:
What's funny about this is that Java, Spain and Swahili each seem to have different causes. Haven't looked at Spain yet, though.
Click to expand...
I can confirm that the reported issues with Java, Spain and Swahili were not only accurate, but indeed had entirely independent causes. Not only that, none of them were caused by the recent America oriented changes. Arrogant European once again blames Americans for everything.

Java: there was a genuine bug that prevented the AI from considering valid city sites on other landmasses at all. This probably has negative effects across the board, but especially ensured that Java got stuck on the eponymous island. I also improved their city AI a bit to make them (or theoretically anyone in a similar sitution) prioritise building settlers once there is nothing else to settle on that city's landmass. Outcome: Borneo and Sulawesi are now reliably settled.

Spain: was affected by a much earlier change intended to help the AI expand. Some AI civs get location with high settler values revealed at the start, so they know they can send settlers there instead of preferring suboptimal locations. This also revealed the Caribbean to Spain from the start, and they decided that these city sites are so preferable that they did not want to settle in Spain anymore. Of course the New World is unreachable, so instead they did not do anything with their settlers. The fix is to simply not reveal other continents to the AI this way. Outcome: Spanish settler patterns are back to normal.

Swahili: the problem is that they start with settlers in ships, which the game did not like. The primary problem is that the naval settlers move before the starting settler, so they were unloaded to found the capital. After that the AI proved unable to load those settlers back into ships. Swahili kind of rely on ships to actually get settlers to their desired city sites, because of the jungles in the way. I prevented the capital founding foolishness to stop the main problem. There were some further issues where the Swahili settler sites fluctuated while they reveal more tiles, which can lead to settlers being unloaded somewhere and the destination immediately no longer being considered a desired site anymore. These cases were mostly addressed by adjusting their settler maps. Outcome: Swahili manages to found three cities with its starting settlers.

There were furthermore some general issues that I improved. Ships with settlers are generally directed to pick up a defender and a worker before embarking. This is useful logic in general, but assumed that ships had three cargo slots. Transports with only two cargo often got stuck waiting to have a worker, even though they already had their slots full with the settler and defender. I also relaxed the conditions for bringing these two units. If the destination landmass has zero cities, it is considered safe and no defender is needed (previously this was merely an exception for one tile islands). Likewise a worker is only needed if the destination landmass has fewer workers than cities.
 
Leoreth said:
I can confirm that the reported issues with Java, Spain and Swahili were not only accurate, but indeed had entirely independent causes.
Click to expand...
You're welcome. Put a lil extra in my paycheck, boss?

Thanks for taking the extra time to write this breakdown. Juicy bits like this doesn't just satisfy our curiosity but conduces to more effective play testers.
 
Spain and Swihili both looking back to normal. Java did not in the one game I've ran so far even with some WB moves (unrevealed tiles, unloaded transports, deleting settlers, etc).

England has only settled a core city and the Caribbean in going on five plus games now. This includes those in which I kept them at peace with other colonial powers to remove that variable and made sure they had Galleons with SETTLER_SEA role (and puled similar tricks as with Java). This is partly good in that Ireland no longer gets overcrowded, but even Cornwall is left empty.

Swahili founded Mombasa forthwith but has had a loaded Settler idle indefinitely.

Don't think Norse has founded SW Norway nor Iceland in as many games as I can recall. Don't think Sweden has founded their typical trans-Baltic cities either, and only founded Kalmar when I moved settler there.
 
Last edited:
I’ve registered to chime in on the settler discussion, but looke like the issue might have been already solved – I’ve notices across some of my recent games that both the Greeks and the Romans were reluctant to use their settlers, and most likely the transport fix will fix it. Thanks a lot! ^_^

The Assyrian leader music doesn’t seem to be playing in my games. I don’t think a savegame will be needed here, and I don’t find this issue with anyone but the Assyrians (when contacting them, when zooming in on one of the cities, or during the initial countdown, there is just silence). Not sure if it might be an issue with my installation or something.

And lastly, I’m sorry if this was already discussed and I missed it, but some South American tiles that start as mountains and flip to hills later don’t have any city name assigned. In one of my recent games as Spain, I had the Incans found a city on one of those (36, 22; 1w of San José de Chiquitos silver tile in Bolivian lowlands), and it was stuck named ‘^’. Not sure when and how you’d like this reported, but I hope it helps.

(And thanks again for working on this project, I’ve been playing the original RFC like 15 years ago, and I’ve recently discovered this, and it’s absolutely amazing. Totally love it. Been lurking on the forums, likewise, for more than a decade already, and finally decided to write something.)

Edit: no, in fact I’ve downloaded the latest git release and started a Phoenicia game, and the Greek boat settler has been remaining idle in the capital for my first 15 turns, and that having in mind that Greece starts a dozen of turns earlier than Phoenicia. Savegames on initial turn and 15 turns in attached.
 

Attachments

Last edited:
Leoreth said:
New update:
- improved AI for settler transports
- adjusted Swahili settler map
- adjusted Javanese starting units
- fixed AI determining overseas settler sites
- improved debug display of AI settler sites
- fixed a bug that revealed overseas territory to AI civilizations and disrupting their settler AI
Click to expand...

Looks like you didn't include the updated DLL in the latest git release, just the source files.
 
Oh that's possible, thanks for pointing it out.
 
You must log in or register to reply here.

Similar threads

M
The Rome thread
Replies
15
Views
2K
manio curio
M
D
Game won't start
Replies
1
Views
452
Taccola
Taccola
A
[submod] RFC Dawn of Civilization: Aeons - Version 1.3! (34 new Civs! New wonders, buildings, units, and an overhaul of various mechanics!)
3 4 5
Replies
92
Views
10K
noppler
N
Leoreth
City Names Feedback Thread
Replies
15
Views
915
SultanRedSnake
SultanRedSnake
A
RfC DoC configurable?
Replies
3
Views
604
CrossRFC
CrossRFC
Back
Top Bottom