Be a bug detective!

[unmerged(6777)]

Okay...so we're pretty sure that some change made between the June 3 and August 15th beta patches is causing a sudden increase in the instances of CTDs but both Byakhiam and I have been through the changes and can't see anything obvious. It's possible that the reason is simply that we're so familiar with the files that we simply don't notice the error and that a "fresh" set of eyes will spot something that we won't...so here's your chance to be a bug detective!

Attached is a rar file. Download it and unpack it TO A NEW FOLDER (not to your CK game folder).

You will then see a list of html files that you can click on to open in whatever your default browser happens to be. This is a comparison report of all changes made to all database and event files between the two patches. The June 3rd version is shown on the left and the Aug 15th version is shown on the right. Changes are highlighted, and a few lines of "context" are shown around each change. You will also see the line number of the file if you want to go to the file and look up the entire event yourself.

If anyone can spot an error in there, please post back the details here.

If we can't find any errors then I guess we'll move on to the scenario and character-related files in case there's something in there that's causing the crashes, but with the majority of your crash reports ending with a rpvince_effect call, the most probable bug will be in an event that is changing province effect.

Good luck, and happy bug hunting.

 

[Schlieffen]

Oh I so found this.
battle_events.txt, event id = 7024
That's it. Don't know what it's doing wrong - "random_enemy" can't be the province event refered to maybe - but it's causing the crashing. Triggered it manually for myself five times with a crash each time (only when option A is picked), set the MTTH to one to have a crash within a month three times as the AI got it, and the exact same last line in the debug reports each time. In addition, the effects of the event can be seen in all of the reports posted so far as well:

Processing an 377 (_CEUEVT_PRESTIGE_) to 2.10
Processing an 432 (_CEUEVT_PIETY_) to 2.10

Both of these lines are pretty close to the bottom of each crash report so far, including my own. Thus I deduce this event to be the culprit.

EDIT: Yeah, it's something about the event triggered by option A. I commented it out and went crazy with it, pushing the character into triple-diget negative piety and nothing went wrong. Probably something about your "random enemy" being a province that it doesn't like.


Yay me!

 

[unmerged(21937)]

Excellent work Schlieffen! It's not a problem with random enemy though. 7024 just triggers 7021 in option A and 7021 is a province event, which can't be triggered by other events.

 

[unmerged(2456)]

I thought province_x_province and character_x_character events did not work well with csc conditions. i think it may be the cultural conversion event now trying to fire.

 

[unmerged(21937)]

CSCs work fine in P_x_P events, as long as they are correctly scripted.

 

[Riddermark]

i can confirm cultural event works for ai at least i had a savegame which i continued and had several 2-3 events of cultural conversion with no crash.

 

[Ayeshteni]

Yay me!

Three beers for that man!
Ay.