Thunderbrd
C2C War Dog
Ah... just means you need to update the unit schema in the folder in which Florence is located. Simply copy the core unit schema over and rename it to the schema name you're using in that file.
Ah... just means you need to update the unit schema in the folder in which Florence is located. Simply copy the core unit schema over and rename it to the schema name you're using in that file.
<Combat>UNITCOMBAT_HERO</Combat>
<SubCombatTypes>
<SubCombatType>UNITCOMBAT_CIVILIAN</SubCombatType>
</SubCombatTypes>
On that second panel it looked like the formatting might be a little funny but I couldn't see the full xml on the entry. It should read:
Code:<Combat>UNITCOMBAT_HERO</Combat> <SubCombatTypes> <SubCombatType>UNITCOMBAT_CIVILIAN</SubCombatType> </SubCombatTypes>
Despite the message you're getting, the schema looks correct to me off hand.
Ok... figured it out. You had the same schema name in use elsewhere and that schema had not been equally updated. Oddly, this causes problems and we've run up against it being an issue before. I'll push your files in just a bit here.
@Koshling:
I did some more testing and it turns out that the autosave size does not inappropriately expand before revision 4803, which was the one in which you made optimizations to save time.
I'm stumped then. I can se no way at all that those changes could cause it, and I cannot reproduce it (played several hundred turns with auto save every turn this morning without getting it)
Can you do a debug then on the savedgame above, maybe that will be something(if not sorry) i am like you i hate wasting precious TIME., like i said it took over 5 minutes for it to go to the next turn, might have been a loop problem??
Is anyone else noticing doubled turn times, or is it just me again?
Double, not me, maybe a 5-6 second one(all depending on Era and # of civs), but thats it at least for me. . . SO
Again, i think it is depending on where you are on YOUR map, how many civs, speed, map size, viewports etc.
Based on looking at your save that you said took over 5 minutes, the massive turn time there is down the Python errors happening tens of thousands of times per turn. However, I cannot see anything wrong with the Python, and for some of the errors it hasn't even been changed in many months. I **think** event or turn initialization Python is failing in some way (but not logged) and that all the errors we then see are due to init functions never having been run. I am currently suspicious of TB's trait changes, since they removed some Python interfaces from the DLL but in at least some cases left calls to them in the Python. I'm investigating still, but if you are seeing long turn times check your Python error logs and see if you are getting Python errors...
Dang... yeah, you're right I shouldn't have done that. Sorry. Good to see its been largely ruled out as the source of trouble though.I am currently suspicious of TB's trait changes, since they removed some Python interfaces from the DLL but in at least some cases left calls to them in the Python. I'm investigating still, but if you are seeing long turn times check your Python error logs and see if you are getting Python errors...