Bug Reports and Technical Issues

[SultanRedSnake]

general observer mode is easier to handle for me than civilization autoplay

Ah I hadn't taken care to distinguish between the two up to now.

you can do so with breakObserverMode()

Yas these are my favorite tricks to learn.

but it was a lot harder this way.

That is not a state of affairs I would countenance ever.

Were they resurrected into the Classical era?

Yes it was Classical.

 

[Dracosolon]

Baibars leads the Ptolemaic Kingdom in 200 BC.

Now it's the opposite, medieval Ramesses.

 

[LottevU]

Does anyone have an idea why the mod takes an eternity (just had a 14 minute wait) to load the title screen every time? For reference, I'm using the 'original_release_unsupported' steam version. Having a great time with it, so it adds up.

 

[Leoreth]

That's extremely long, starting the mod takes a while for me but it's always under 30 seconds. Maybe there is something off with either your hard drive or memory?

 

[AOS9001]

The first time I load DoC after restarting the machine or a few days of not playing it, it takes about 5 minutes to load. After there's a cache established, it takes me about 30 seconds too.

I remember it used to take 15-20 minutes to start loading the mod, back in the old old days of 1.15 or so, when I was running the game on a 2011 laptop.

 

[SOVNARKOM]

There seems to be a bug with the dynamic name code of Rus as the vassal of Mongolia, which is currently directly displayed as TXT_KEY_CIV_MONGOL_RUSSIA.
It seems that this code in Python does not exist in XML (the code for Golden Horde is TXT_KEY_CIV_MONGOL_RUS)

 

[trevor]

Don't know if it's a my computer issue but pressing the "exit to main menu" button results in a CTD.

 

[SultanRedSnake]

prevented AI from declaring war to open a path to an independent city target

So was this the key to the Spanish DoW puzzle afterall?

 

[Leoreth]

It was. It is not exactly scripting, but when the AI has decided to target a city but no way to reach it, it might declare war on the player in the way. Spain is susceptible to this due to its geographic location and the propensity to target Rome.

 

[SultanRedSnake]

 

[SultanRedSnake]

So now the question is what would encourage Spain to DoW the HRE like this? Situation: 1230AD, Isabella Pleased with Frederick, both at peace, no HRE cities on Spanish war map. War plan seems to be limited since only a few Lancer squadrons were sent and they went about pillaging. Don' think war lasted more than 10 turns.

 

[Leoreth]

I can look into it, but I realised from investigating your earlier save that DoW saves often aren't sufficient because the AI usually has a war plan going for a few turns before that.

 

[Kott]

Why don't they just build a boat to transport troops to Rome?

 

[Leoreth]

They can, but it's a different unit AI. And the naval AI of the game is bad and my suspicion would be that a lot of the code assumes that rerouting units via ships is not an option.

 

[banzay13]

They can, but it's a different unit AI. And the naval AI of the game is bad and my suspicion would be that a lot of the code assumes that rerouting units via ships is not an option.

Only US (mostly) can do successful naval invasion - so AI can handle with it. On very late part of game - Germany or other superpower can do it too. So the question here - AI can't concentrate military power with available economy/transport situation. In Cross modmod AI doin naval thing much better - more transport capacity for naval units (all have +1 to standart), irc more available turns (+1 to standart?) and, transport (not military) units can cross all naval borders (like Caravels)

Also, why Japan so weak with water military. Japan's AI not willing build a lot of ships - so invasion of Korea or Asia is very rare.

 

[Leoreth]

Like I said, the AI can do it but many decisions are happening on the unit level and the AI is not very good at coordinating things into doing a naval invasion instead.

 

[SultanRedSnake]

In line with similar experiments, I manually loaded a Spanish cog (ASSAULT_SEA) in Barcelona with two Lancers (ATTACK_CITY) and it buzzed straight for Southern Italy. I continue to suspect that work need only be done on AI loading logic. Can't figure out how to take a screen with SHIFT and CTRL held so I'll relay that the target plot is 67,51. Or is this not nearly so easily done? The rub, as it were.

 

[Leoreth]

Again. Like I said, the unit level AI supports this. The issue is coordination between units and cities.

 

[SultanRedSnake]

Ahhhhhh I got em: "unit level AI" I breezed past that and it didnt quite click first so I tried "AI at the Unit level" and I see the significance; its counterpart "AI at the City Level" . Working at the junction of *distinct*, already complex and wide-reaching systems would make this a higher order job then, innit? We'll move it to the DoC Dream Doc it'll be safe there. Also at least one other major mystery of mine has been solved.

 

[Leoreth]

Yeah and there isn't really a higher order. Just units and cities acting based on some very broad AI modes and goals. I don't really want to do anything about this at this point of development.