? I get it with Arabia too so I don't think it's that late.
Edit: Repeatable CTD in 568 AD. Not sure why though.
? I get it with Arabia too so I don't think it's that late.
Edit: Repeatable CTD in 568 AD. Not sure why though.
Huh. Strange indeedDid you try my save?
And yet I didn't make any modification on my install... I may try to reinstall some time later.
When the game crashes do you get anything in the other logs, like "audio.log"?
I don't think there is a problem with the new sound files
Updated to rev 750 and no ctd anymore (not tried Austria yet).
I think 749 was a graphics update.
749 was mine
I only changed small things, so I'm not really sure where could I have made a mistake
The only bigger thing was that I started to mess up with the attitude/memory modifiers in the leaderheadinfos.xml, but I don't see any mistakes there either...
XML mistake would cause the game to not load. Unless it is graphics or C++ it shouldn't crash mid-game.
Then my best guess is that it's somehow related to the crusades
I experienced this a couple times right after a crusade was called
Maybe it's connected to Jeruselam being a minor nation?
All crashes were around the time the AI could have conquered Jerusalem
I checked a few older revisions
The game crashes with 748
There is a crash around 600AD, but in some reloads/retries I was able to get through that
The other crash is around 1200AD, never got to the austrian start
Unfortunately I didn't find anything useful in the logs
But the new sounds definitely works fine (at least Simeon's diplomusic and the serbian chant), I'm sure of that
EDIT: revision 744 seems fine, I started 2 austrian games without any problems
I'm almost certain that there are no issues in the small updates of 745-747, so there is a good chance the bug is in revision 748
It's probably in the mercenaries.py changes there
Are you still working on that one? I tested it on the latest revision and if it works fine there, I declare the bug fixed and move on. Unless I have a consistent bug, there is little that can be done. Checking with old revisions is also a bad idea since we have old bugs that have been fixed. Last weekend (I don't remember which revision) I had a Merc bug that would constantly cause random infinite loops, but it has been fixed now.
When something is random like that, it is possible that the bug is due to broken linkage in the .dll file. When I test things, I sometimes have make partial rebuilds of the .dll 10 - 20 times, eventually this leads to broken linkage and I have to rebuild everything from scratch. I try to remember to do that right before I submit a revision, but it is possible that I forgot (and there is no way to check).
Test the file on the latest revision and unless you get a consistent bug there, move on.