Fall Further 050 Bug Report Thread

Was a DLL based fix, so unless FFlat is distributing a DLL with the patches (might be, since he had included some hotfixes before FF released them as a patch) you should have gotten the fix. If he is releasing a DLL with patches, then it is most likely out of date.


OOS issues are fairly tricky to track down, unless you are lucky and someone used "Active" calls somewhere, those are generally dead giveaways for OOS. About the best you can hope for is a repeatable (or easily reproducible) OOS trigger, then start cutting chunks of code and hoping you make it go away. Then narrow down the size of your cuts till you isolate a single line (or function) and then rewrite it until you solve the issue.
 
I'm not sure if this is from FF or basic FfH, but when I get the angelic procession event the text says the Armageddon Counter is reduced by two but it only drops one step.
 
Was a DLL based fix, so unless FFlat is distributing a DLL with the patches (might be, since he had included some hotfixes before FF released them as a patch) you should have gotten the fix. If he is releasing a DLL with patches, then it is most likely out of date.


OOS issues are fairly tricky to track down, unless you are lucky and someone used "Active" calls somewhere, those are generally dead giveaways for OOS. About the best you can hope for is a repeatable (or easily reproducible) OOS trigger, then start cutting chunks of code and hoping you make it go away. Then narrow down the size of your cuts till you isolate a single line (or function) and then rewrite it until you solve the issue.

In other words, try it myself and see.

I'll go ahead and do that. Change a couple lines of python and see if I get an OOS error.
 
I'm not sure if this is from FF or basic FfH, but when I get the angelic procession event the text says the Armageddon Counter is reduced by two but it only drops one step.

The AC is displayed as a percentage of the armageddon value. For instance, reducing 102 out of 200 by 2 is enough to lower the AC by one percent.
 
So, in trying to squash that OOS bug, I found another one. The log isn't terribly helpful, but this is what happened:

The game crashed at the start for computer 2. Normal for multiplayer. Anyway, it rejoined and I got started. 40 turns in, I still didn't have an awakened! So I exited from computer 1, changed my config file so I could see the pyton error, and then rejoined. Then I got an OOS error. I checked the OOS logs and the only thing different is the seed. No other clues as to why rejoining caused the problem.
 
The OOS error for scion awaked spawning fixed. I took the code from onBeginPlayerTurn to onGameBeginTurn... I modified it just a little to work in the different context, but here it is:

Code:
# This code is now in onGameBeginTurn instead of onPlayerBeginTurn

# SCIONS START - doTurnScions has the Awakened spawning code
		for nPlayer in range(gc.getMAX_PLAYERS()):
			player = gc.getPlayer(nPlayer)
			if player.isAlive():
                		if player.getCivilizationType() == gc.getInfoTypeForString('CIVILIZATION_SCIONS'):
                			cf.doTurnScions(nPlayer)
			
# SCIONS END

For best results, all civilization's special "docivstuff" should be moved in this way. Simply add all of the "if civ = X" lines here and it'll cut down on quite a few potential OOS issues.
 
So, in trying to squash that OOS bug, I found another one. The log isn't terribly helpful, but this is what happened:

The game crashed at the start for computer 2. Normal for multiplayer. Anyway, it rejoined and I got started. 40 turns in, I still didn't have an awakened! So I exited from computer 1, changed my config file so I could see the pyton error, and then rejoined. Then I got an OOS error. I checked the OOS logs and the only thing different is the seed. No other clues as to why rejoining caused the problem.

Base issue with FfH since before I started to mod, so have to comb through ALL pre-existing changes to find that one. But rejoin is NEVER an option with FfH based mods till it is found, all player must exit to the lobby or you OOS instantly.

One more bug: The score display button hides the score list and the manabar both, which is rather weird because the mana bar has it's own hide button.

Manabar was made by Kael before anyone really messed with understanding the whole screen system too completely, and before DLL modification was possible, and he just slipped it into another widget. Was initially placed above the score list so it probably made sense then. Anyway, when I overhauled the interface it just never occurred to me to grant it a widget seperate from the scorelist. Wouldn't be hard to do at all.
 
I found and fixed the error for svartalfar shrouded woods. It's in CIV4CivilizationInfos.xml under (what else?) the Svartalfar building section:

Code:
				<Building>
					<BuildingClassType>BUILDINGCLASS_CIV_BUILDING1</BuildingClassType>
					<BuildingType>BUILDING_SHROUDED_WOODS</BuildingType>
				</Building>
should be:
Code:
				<Building>
					<BuildingClassType>BUILDINGCLASS_HUNTERS_HALL</BuildingClassType>
					<BuildingType>BUILDING_SHROUDED_WOODS</BuildingType>
				</Building>
 
While I'm at it, the following is missing from the Bannor Units section in the same file:

Code:
				<Unit>
					<UnitClassType>UNITCLASS_CLERIC</UnitClassType>
					<UnitType>UNIT_CLERIC</UnitType>
				</Unit>
 
I went looking for cannot be built by anybody tags, and found that amurites can't build their wizard halls or battlemages and hippus can't build a field of the horselords. I fixed everything, but I have nowhere to host it, if anyone wants the file.
 
there is definatly a bug where a gobblin archer in a gobin fort will cause an OOS when they shoot at someone using opportunity fire. In this game it seems to be 100%. Unless it only happens when it is visible and fires and causes an OOS.
 
SO I CAN.

I thought it was too big, maybe I was looking at the wrong file. Anyways here it is, extract it to \Beyond the Sword\Mods\Fall Further 050\Assets\XML\Civilizations\

The file is taken from Patch N and fixes the shrouded woods, clerics, wizard halls, battlemages and fields of the horselords.
 

Attachments

While I was at it, I fixed the hunter event to give myself one less reason to open the world builder. The EVENT_HUNTER_1 through 5 sections in Assets\XML\Events\Civ4EventInfos.xml were missing the following line:
Code:
			<bPickCity>1</bPickCity>
 
A little on gobblin archer oos, it happened every time an AI player came in range and it attacked the AI player. When the AI player died we would keep on playing until another unit came in range and we started with the oos again until the AI retreated or was killed. Then no more oos until they attacked the gobbie again. Player 1 finally killed the gobbie no more oos. That wasn't the first gobbie to have that happen all gobbie archers caused this. I am thinking we can only verify it happened on the server side, not the client side.
 
Back
Top Bottom