Bug Reports and Technical Issues

[Leoreth]

I fixed a bug where the game speed was not properly taken into account. An equivalent city should have five turns of unrest at normal speed which is on the high end. I can look into your game if you want to see if everything works right.

But shouldn't cities in unrest also consume no food?

The AI should not build Mole Antonelliana without peaks, I'll look into it.

 

[Vectors]

I fixed a bug where the game speed was not properly taken into account. An equivalent city should have five turns of unrest at normal speed which is on the high end. I can look into your game if you want to see if everything works right.

But shouldn't cities in unrest also consume no food?

The AI should not build Mole Antonelliana without peaks, I'll look into it.

I think it is a combination of commit d6d6612789d30c368b57f725f69a7948bc132b4a which implements gradual population loss in combination with the commit that properly scales unrest to culture level and game speed. the cities lose one population per turn. Starving was an incorrect term, it was general population decrease

 

[Leoreth]

Are you sure population is being lost over the entire duration of the 15 turns? The way it works is that a set number of population is lost (does not scale with game speed, obviously), if the number of unrest turns exceeds that it will lose one population per turn until that number is lost.

 

[Vectors]

Are you sure population is being lost over the entire duration of the 15 turns? The way it works is that a set number of population is lost (does not scale with game speed, obviously), if the number of unrest turns exceeds that it will lose one population per turn until that number is lost.

They seem to consistently drop until they hit 1.

 

[Leoreth]

Okay, that's not intentional then. Will look into it.

 

[Vekkimaster]

As said I downloaded the (now almost) the latest patch of the mod...played the Romans (big suprise) and hit the year 1280AD...no Ottomans. And never did they spawn. Something I'm missing or are the Ottomans nowadays Bye-Bye

 

[1SDANi]

As said I downloaded the (now almost) the latest patch of the mod...played the Romans (big suprise) and hit the year 1280AD...no Ottomans. And never did they spawn. Something I'm missing or are the Ottomans nowadays Bye-Bye

IIRC the Ottomans don't spawn if the Turks didn't obtain any cities in Persia or the Middle East

 

[Vekkimaster]

This error started to happen after Krak des Chevalier obsoleted although can't say if that is the reason.. Basically I can start to construct a castle, but 1 turn before being complete that error comes up.

 

[Leoreth]

Please share the save where this is happening.

 

[Nyayr]

Doesn't the invention that obsolete the krak also obsolete the castle? I do know sometimes, you need to close/reload a save, so the rechecks (like if you become secular before save/reload it doesn't show the new religion wonders always for me).

Edit: N/M Castle is obsolete with urban planning, and the krak with combined arms. I guess save/reload doesn't fix it?

 

[TJDowling]

So I'm getting this bug which says every turn that my cities can't build castles anymore and despite being told that I can't build castles anymore they keep building uninterrupted and I keep getting the gold refund for lost production turn after turn (without losing the production). So this bug is incredibly profitable as I'm currently trying to build a half dozen or so castles but it seems to be a bug none the less.

 

[Nyayr]

I got a repetitive crash ending the first turn of Byzantium.

PYlayer 5 Civilization Gupta Empire City Tanjapuri was razed by Player 49
PY:City Razed Event: Tanjapuri
15:14:21 DEBUG: BugEventManager - event religionRemove: (5, 49, <CvPythonExtensions.CyCity object at 0x1A087BC8>)
15:14:21 DEBUG: BugEventManager - event improvementBuilt: (2, 91, 30)
15:14:21 DEBUG: BugEventManager - event cityLost: (<CvPythonExtensions.CyCity object at 0x1A087BC8>,)
15:14:21 DEBUG: BugEventManager - event cityCaptureGold: (<CvPythonExtensions.CyCity object at 0x1A087BC8>, 49, 3)

Removing the barbarian that capture Tanjapuri fixes the crash, so it is related to it. Seems they razed a city and it started calculating what was lost?

 

[Leoreth]

Yes, sounds like it's related to the conquest rules changes.

 

[Vekkimaster]

Please share the save where this is happening.

The autosave is few turns away from this bug...keep your eye on the Egyptian cities of mine.

I was wondering...when I build Krak des Chevaliers (gives castle to every city) and it's obsoletes with combined arms...ok...but the "normal" built castle obsoletes after urban planning...so...after the wonder Krak des Chevaliers has no longer an effect to my cities, I lost all but normal built castles...and I have to build them all over again...seems a bit unlogical...

Quoting myself with this one. I think that when the effect of Krak des Chevaliers obsoletes, it should not erase castles away from the cities that wonder gave you (castles to all your cities). Because now it does. Only the defeated defenders retreat to your capital-effect should go away.

 

[Vekkimaster]

The autosave is few turns away from this bug...keep your eye on the Egyptian cities of mine.

I still have the save before Krak des Chevaliers obsoletes from my city in case needed to pin point the bug.

 

[DarkLunaPhantom]

I think that when the effect of Krak des Chevaliers obsoletes, it should not erase castles away from the cities that wonder gave you (castles to all your cities). Because now it does. Only the defeated defenders retreat to your capital-effect should go away.

This is inherited from BTS. If I recall correctly, in BTS the Stonehenge, which gives free Monuments, obsoletes with Astronomy which then erases all your free Monuments. Monuments themselves are also obsoleted by Astronomy in BTS, but the ones you built yourself still give you culture after that (as that is the only effect that does not obsolete and is their primary purpose). So in effect, getting Stonehenge would rob you of late game culture you could normally get from Monuments. This is the way the game handles free buildings by default as they differ from "real" building in the code.

 

[Jarlaxe Baenre]

This is inherited from BTS. If I recall correctly, in BTS the Stonehenge, which gives free Monuments, obsoletes with Astronomy which then erases all your free Monuments. Monuments themselves are also obsoleted by Astronomy in BTS, but the ones you built yourself still give you culture after that (as that is the only effect that does not obsolete and is their primary purpose). So in effect, getting Stonehenge would rob you of late game culture you could normally get from Monuments. This is the way the game handles free buildings by default as they differ from "real" building in the code.

I was just thinking that Python could probably handle this a lot better than the native free building feature.

 

[Leoreth]

I got a repetitive crash ending the first turn of Byzantium.

PYlayer 5 Civilization Gupta Empire City Tanjapuri was razed by Player 49
PY:City Razed Event: Tanjapuri
15:14:21 DEBUG: BugEventManager - event religionRemove: (5, 49, <CvPythonExtensions.CyCity object at 0x1A087BC8>)
15:14:21 DEBUG: BugEventManager - event improvementBuilt: (2, 91, 30)
15:14:21 DEBUG: BugEventManager - event cityLost: (<CvPythonExtensions.CyCity object at 0x1A087BC8>,)
15:14:21 DEBUG: BugEventManager - event cityCaptureGold: (<CvPythonExtensions.CyCity object at 0x1A087BC8>, 49, 3)

Removing the barbarian that capture Tanjapuri fixes the crash, so it is related to it. Seems they razed a city and it started calculating what was lost?

A fix for this is coming soon.

 

[Leoreth]

So I'm getting this bug which says every turn that my cities can't build castles anymore and despite being told that I can't build castles anymore they keep building uninterrupted and I keep getting the gold refund for lost production turn after turn (without losing the production). So this bug is incredibly profitable as I'm currently trying to build a half dozen or so castles but it seems to be a bug none the less.

The autosave is few turns away from this bug...keep your eye on the Egyptian cities of mine.



Quoting myself with this one. I think that when the effect of Krak des Chevaliers obsoletes, it should not erase castles away from the cities that wonder gave you (castles to all your cities). Because now it does. Only the defeated defenders retreat to your capital-effect should go away.

Which revisions are you both on? I tried develop and couldn't load, and the commit from the other of your saves I looked into didn't work either @Vekkimaster.

 

[Leoreth]

It is so RARE that a bug pops up in v1.15, that I wanted to report it directly.

Spoiler Bugged CvRFCEventHandler.onBeginGameTurn :

View attachment 517427

Attached is a save file from the turn right before it happens (just press Enter). If save files from previous turns are needed, I can provide them.

I looked into this and it's a bug that has since already been fixed. Thanks for reporting!