Bug Reports and Technical Issues

Just to affirm some recent posts, I'm getting "DLL attaching" also.
Recently I switched to map, loaded into one game to make some map changes, then accidentally loaded a save, then back to develop, which is when I started getting the issue.

Turn times are REALLY long on 1700, like unreasonably long (from ~5 seconds to over 30).
Looks like I accidentally pushed a debug DLL, sorry.

[Overflow bug]

No, I don't have an earlier save, sorry.

But Ninua? Really? Because when I load the save (I checked again before replying here) the overflow is in Iol Caesarea.
Could be that more cities are affected, it's just that I could pinpoint it to Ninua at least. I assume Iol Caesarea is also independent?

The issue with the bug is that when it is noticed, it is already too late. There is a discrepancy between the actual properties of the city and what the game keeps track of what their commerce rate is (either insanely high or insanely negative), so at some point some calculation must go wrong and introduce the error, which then never goes away. My current suspicion is that it is related to city flips, likely during collapse, where the new city of the new owner (strangely always independents so far) has the error.

I encourage you to enable autosaves with a generous retention rate so that the last 20 turns are covered at least (doesn't have to be every turn). If you suspect a city is affected, please submit a save from before it flipped to independent right along with it.

I think I may also temporarily force the game to create a save before a civilisation collapses outside of the autosave system, so that we are guaranteed to have saves of all those situations. That should help having the right situation on hand to look into.

How about two more to look at? Both are development version of DoC.

The straightforward one is a CtD in an India game. Attached is the turn before it crashes (155), and the two turns before that. Three times in a row it CtDed on me, even when backing up to Turn 153. I finally got around it by reloading the Autosave on Turn 152 and playing forward from there. The crash seems related to Rome's collapse.

Spoiler :
That was a viciously mean game that I eventually resigned, btw, so I offer the CtDs only as research; I never want to see that particular game again. The eastern Mediterranean powers never built roads, and China never opened its borders, so for the early game I could trade only with Persia. The barb spawn rate was thru the roof, sometimes three turns in a row, and never less than every other turn, a stack of them spawned. Byzantium never spawned, leaving me in AD 600 on with only the Turkmen for trading partners after conquering Chola. That's no place to try running a 10-city empire, even with two shrines to help support the economy.
Alright, I will look into it.

Then there's the goofy one, which might just be random roll thing and not a bug. Background: If you don't settle on the spot when you spawn as India, Indraprastha should flip to you the next turn, founding Hinduism there.

But in this one, if you delay, Indraprastha doesn't flip until the turn after that. Moreover, you can move the Chariot into Indraprastha while it is still an Independent city. I've attached Turn 54, which is the turn where you spawn. Move the Chariot west to find Indraprastha and observe the behavior.

I had another India game where Indraprastha flipped on me at the proper time, but Hinduism didn't found. I don't have a save of that one, because I didn't think to save the game on launching it. But I'll mention that weirdness in case it is relevant to the above case.

EDIT: I'm starting to think the goofy one is random happenstance and not a bug, because it's happened to me several times this evening on launching and relaunching an India game. (But it never happened in 1.16 to me, so would it be a change introduced in the development version?) I'll also note that on those goofy launches, Indraprastha will generate three Workers, but only sometimes will it generate an Archer to go with it.
Yeah, the spawn mechanism is sometimes irregular like that. I plan to rewrite it in 1.17, the new implementation should behave more logically and reliably.
 
Generally with the founding of Religion it is in the first city you own, also Workers get put in that city and the Capitol. I played with Harrappan making a city that flipped next to the spot Persia appeared. Then weirdly the Capitol got put not in Shush (which I expected) but a Babylon city 4n/1e Thuspa.

Persia doesn't have a flipzone for Thuspa but sometimes Byzantium also flips Egypt, so seems a flipzone can expand somehow sometimes, maybe depends on stability of the owner?

On another note, I always go to main menu before loading a save in game (saves time else loading takes ages for me) this save when I load I can just play it but when I go to main menu it crashes. Had this happen more especially with games in later ages and having a huge empire. Not sure what causes it (but there been several updates since I started playing it 2 weeks ago and I do update) A save a few turns before doesn't have the issue... Wow now it also CTD when going to main menu.

Update: Forgot about it, I had changed my graphic settings, to low and back to high and it resolved it... Maybe the Nvidia update caused something?
 
Last edited:
I got a repeating fixed crash it is the turn USA appeared (playing as Rome and expanded into North America). End turn and at the start before the autosave it crashes.

Deleting the Americans through world builder doesn't solve it.

Deleting the city Mascat solves it. Removing the Troops so the German can conquer it also solves it. Removing 1 independent arquebusier also resolves...

the skdDbg varies but shows up:

[276964.031] AI chose conquest: Shouer
[277028.094] AI chose conquest: Mascate
[277035.844] AI chose conquest: Shouer
[277157.719] AI chose spare: Mascate
[277164.656] AI chose conquest: Shouer
[277372.531] AI chose conquest: Mascate

Update 2: Deleting the Defender of Shouer (Japanese hold-Korean city) also resolves it. Suggesting it is related to the random seed?
 

Attachments

Last edited:
Another persistent CtD: When ending Turn 224, impossible to get past even reloading from four turns previously. I've also uploaded Turns 220-223.

Development version of the game.
 

Attachments

Hi Leo, the game is reporting the inconsistent commerce bug on the following turn for me. Hope this helps you sort that out!
Oh great! Thanks. I hope so too. Let's see what I can find out.
 
Hi Leo, the game is reporting the inconsistent commerce bug on the following turn for me. Hope this helps you sort that out!
Do you also have an earlier save?
 
[Overflow bug]



No, I don't have an earlier save, sorry.

But Ninua? Really? Because when I load the save (I checked again before replying here) the overflow is in Iol Caesarea.

How about two more to look at? Both are development version of DoC.

The straightforward one is a CtD in an India game. Attached is the turn before it crashes (155), and the two turns before that. Three times in a row it CtDed on me, even when backing up to Turn 153. I finally got around it by reloading the Autosave on Turn 152 and playing forward from there. The crash seems related to Rome's collapse.

Spoiler :
That was a viciously mean game that I eventually resigned, btw, so I offer the CtDs only as research; I never want to see that particular game again. The eastern Mediterranean powers never built roads, and China never opened its borders, so for the early game I could trade only with Persia. The barb spawn rate was thru the roof, sometimes three turns in a row, and never less than every other turn, a stack of them spawned. Byzantium never spawned, leaving me in AD 600 on with only the Turkmen for trading partners after conquering Chola. That's no place to try running a 10-city empire, even with two shrines to help support the economy.


Then there's the goofy one, which might just be random roll thing and not a bug. Background: If you don't settle on the spot when you spawn as India, Indraprastha should flip to you the next turn, founding Hinduism there.

But in this one, if you delay, Indraprastha doesn't flip until the turn after that. Moreover, you can move the Chariot into Indraprastha while it is still an Independent city. I've attached Turn 54, which is the turn where you spawn. Move the Chariot west to find Indraprastha and observe the behavior.

I had another India game where Indraprastha flipped on me at the proper time, but Hinduism didn't found. I don't have a save of that one, because I didn't think to save the game on launching it. But I'll mention that weirdness in case it is relevant to the above case.

EDIT: I'm starting to think the goofy one is random happenstance and not a bug, because it's happened to me several times this evening on launching and relaunching an India game. (But it never happened in 1.16 to me, so would it be a change introduced in the development version?) I'll also note that on those goofy launches, Indraprastha will generate three Workers, but only sometimes will it generate an Archer to go with it.
No crash for me.
 
Great (in a sense). I take it that this basically happened right after you spawned? Keeping the turn of the spawn around might help already. I'm kind of suspecting city flips screw things up somewhere so it may have been your spawn flip from India.
 
Another persistent CtD: When ending Turn 224, impossible to get past even reloading from four turns previously. I've also uploaded Turns 220-223.

Development version of the game.
Any chance of a save from turn 213 or earlier?
 
Any chance of a save from turn 213 or earlier?

A-ha! So that's your game, is it? Always ask for an earlier save!

Well, I'm on to you, mister! I've got an auto-save of Turn 212 and an auto-save of Turn 208!

CtD that didn't crash for you: I'm not surprised. My ghost at work.
 

Attachments

Hi, I have a question about religion stability as 3000 BC China Monarch/Epic game (git 12/10/19):

I'm getting a -5 penalty for Religious Disunity, in the following circumstances: my state religion is Taoism, and I have it spread to all of my cities. Confucianism is spread to a few cities. I had first converted to Confucianism when it appeared, but I didn't get many free spreads and then Taoism appeared in a much better production city to get the missionaries out, so I swapped to Taoism to get the Clergy bonus everywhere. In the Stability Influence Factors documentation, I read that Taoism and Confucianism tolerate each other, which makes sense, but it doesn't seem to be the case here, unless the penalty is coming from another source?

EDIT: Alternatively, is it possible that stability hasn't been checked since I swapped into Taoism, and the disunity is when my state religion was Confucianism but I had too many cities with only Taoism and not Confucianism that were counted as atheist/pagan population?

Spoiler Screenshots :

F6UtjLN.png

BIbOE8p.png

yn0irlU.png



EDIT 2: Well whatever it was, it seems to be fine now, and now I'm getting +5 from Religious Unity. All is well :)
 
Last edited:
A-ha! So that's your game, is it? Always ask for an earlier save!

Well, I'm on to you, mister! I've got an auto-save of Turn 212 and an auto-save of Turn 208!
Yeah, sorry, I'm getting tired of asking that question too. Great to hear you have one though.

To further explain, it looks like it's related to additional yields/commerce from resources (like +10% production with copper) being off, but the affected city (Ankara) had no building that provides that. I also noticed the infamous Solar Plant in the city. It flipped last in 214 so I want to check if the error was introduced then.

So I actually got a little further here but with getting further comes more need for savegames :)

Hi, I have a question about religion stability as 3000 BC China Monarch/Epic game (git 12/10/19):

I'm getting a -5 penalty for Religious Disunity, in the following circumstances: my state religion is Taoism, and I have it spread to all of my cities. Confucianism is spread to a few cities. I had first converted to Confucianism when it appeared, but I didn't get many free spreads and then Taoism appeared in a much better production city to get the missionaries out, so I swapped to Taoism to get the Clergy bonus everywhere. In the Stability Influence Factors documentation, I read that Taoism and Confucianism tolerate each other, which makes sense, but it doesn't seem to be the case here, unless the penalty is coming from another source?

EDIT: Alternatively, is it possible that stability hasn't been checked since I swapped into Taoism, and the disunity is when my state religion was Confucianism but I had too many cities with only Taoism and not Confucianism that were counted as atheist/pagan population?

Spoiler Screenshots :

F6UtjLN.png

BIbOE8p.png

yn0irlU.png



EDIT 2: Well whatever it was, it seems to be fine now, and now I'm getting +5 from Religious Unity. All is well :)
Sounds like the calculation eventually behaved as it is meant to. I'm still curious how the -2 came about. If you have the save from before converting to Taoism I can look into it.
 
The calculations can sometimes be a few turns old, I think they update every 5 turns? and if you switch government features but not get anarchy (Cristo Redendor) I notice it updates immediatly.
 
The thing is, both switching civics and state religion should trigger a check that updates the score. So there shouldn't be a delay.
 
There was a Solar Plant too?! OMG ...

I've noticed religious disunity lingering after it shouldn't, and like Nyayr I assumed it was an artifact of a delayed check. I've also noticed that the "obsolete civics" (or whatever that penalty is called) sticks around even after updating to contemporary civics.
 
I'd be happy to provide saves - as I've modified a few files, would it be easier if I made a git branch as well?
 
The save should still be compatible. If you have the one before switching your state religion, that should be enough.
 
Back
Top Bottom