FfH2 Bug Thread

[swordsciv55]

Try the attached save file.

Unfortunately, I had to make an important change to fix the crash - I had to eliminate the Basium hero unit. I'm not sure if the Mercurians lose their characteristics if Basium is killed by the Worldbuilder, I think they do.

I tried to play a turn and put Basium back in the game via the worldbuilder, and it crashed again.

Best wishes,

Breunor

that's okay, i mean it would be nice to have basium, but he didn't really have an affect on me at all so thats okay. but do you mean that basium was removed from the whole game or just the save file? i would think just the save but i want to be sure. thanks for the help

oh and if you noticed i had those random signs and wondered what they were, i just put them as memory to remember where the horselords of the apocalypse were killed.

 

[swordsciv55]

umm im sort of experiencing the same problem as before. but now on a different turn. this one is on turn 596. still no basium around since u took him out so, idk what it is this time.

 

[Breunor]

Once again the Mercurians are the problem. I can fix it by wiping them out. Unfortunately, it will take longer to determine if a specific unit or group is at fault (its celarly from the main army in the Mercurain capital if it isn't something to do with their Civ). There are lots of tigers there - a few years ago, tigers were the kings of crashes. I'll get back when I can.

I did wonder about those signs, thanks for letting me know - I thought this may be some sort of modmod. Also I never got Gaelen before - he's from an event, right?

Best wishes,

Breunor

 

[swordsciv55]

Once again the Mercurians are the problem. I can fix it by wiping them out. Unfortunately, it will take longer to determine if a specific unit or group is at fault (its celarly from the main army in the Mercurain capital if it isn't something to do with their Civ). There are lots of tigers there - a few years ago, tigers were the kings of crashes. I'll get back when I can.

I did wonder about those signs, thanks for letting me know - I thought this may be some sort of modmod. Also I never got Gaelen before - he's from an event, right?

Best wishes,

Breunor

Yea, Gaelan is from going through the event (i guess event chain is more like it) of doing lab testing on people. and also, how do you know its the Mercurians again?

 

[Breunor]

Yea, Gaelan is from going through the event (i guess event chain is more like it) of doing lab testing on people. and also, how do you know its the Mercurians again?

If you go into the worldbuilder and eliminate the Mercurians, the game progresses. That's the easy part - now I have to figure out which unit of units that they have are the culprit.

Best wishes,

Breunor

 

[Breunor]

Ok, try this save:

The bug here is a 'accumulated mage' bug, probably the most prevalent in the mod now. Somebody once checked the code and said what was causing it, but I don't really remember. Unfortunately, it is a wierd bug and may happen again.

The good news is that (even though it took a while to find) I was able to fix it with only one unit removed, one (Mercurian) mage.

Best wishes,

Breunor

 

[swordsciv55]

Ok, try this save:

The bug here is a 'accumulated mage' bug, probably the most prevalent in the mod now. Somebody once checked the code and said what was causing it, but I don't really remember. Unfortunately, it is a wierd bug and may happen again.

The good news is that (even though it took a while to find) I was able to fix it with only one unit removed, one (Mercurian) mage.

Best wishes,

Breunor

thank you. i hope this bug is done now. i also noticed that that one mage seemed to cause 0.9 KB of the save file, cuz now its a bit smaller.

 

[kanban42]

Hi Breunor,

can you please elaborate a little on this 'accumulated mage' bug? Because it seems to be what bites me too right now. Is it in the python code or in the DLL? Is that the 'fix', kill some of the mages? (Unfortunately, I researched Blood of the Phoenix already. )

The game doesn't crash to the desktop, it just 'hangs' when I end the turn. I can track down the cause to one movement where I split a stack with more than 50 mages in it.

--

It feels good to see that FfH2 is still played a lot. It's much better than Civ5.

Best regards

 

[Breunor]

Hi Breunor,

can you please elaborate a little on this 'accumulated mage' bug? Because it seems to be what bites me too right now. Is it in the python code or in the DLL? Is that the 'fix', kill some of the mages? (Unfortunately, I researched Blood of the Phoenix already. )

The game doesn't crash to the desktop, it just 'hangs' when I end the turn. I can track down the cause to one movement where I split a stack with more than 50 mages in it.

--

It feels good to see that FfH2 is still played a lot. It's much better than Civ5.

Best regards

Unfortunately, I don't know what causes it; I just know we get a lot of 'hangs' behavior (often called 'WOC' meaining 'Waiting for other Civilizations' which flashes on the bottom of the screen) and occasional crashes when there are a lot of AI mages.

Best wishes,

Breunor

 

[kanban42]

Okay, if someone could tell me where to look for the problem, that would be great. Right now, I cannot even imagine which game mechanics are involved. Some sort of inconsistent state that is caused by a simple movement?

 

[Tholal]

Okay, if someone could tell me where to look for the problem, that would be great. Right now, I cannot even imagine which game mechanics are involved. Some sort of inconsistent state that is caused by a simple movement?

Likely related to repeated grouping and ungrouping of the same units. I don't know what the exact cause is in this case, but try my AI mod to avoid these sorts of hangups.

 

[kanban42]

Does that mean, it's the AI that crashes? That would make sense...


By the way, there is a small bug in Assets\python\pyWB\CvWBDesc.py showing up in non-English versions of FfH2 (and derivatives like Fall Further, Rise from Erebus, Orbis):
Line 1506 should be changed from

		f.write("\tcaption=%s\n" %(sign.getCaption(),))

to

		f.write("\tcaption=%s\n" %(sign.getCaption().encode(fileencoding),))

and line 1539 from

				self.szCaption = v

to

				self.szCaption = v.decode(fileencoding)

(All that explicit encoding and decoding could be left to python if the codecs module from the standard library were used. Just import codecs and replace both calls to file(...) with calls to codecs.open(..., fileencoding).)


Another note: Why does Divination give access to Life mana and not Alteration?

 

[Bati]

- Never mind -

 

[Tielby]

Another note: Why does Divination give access to Life mana and not Alteration?

In my humble opinion, the 4 initial magical sciences (Elemental, Divination, Alteration and Necromancy) had to be balanced to allow access to four types of mana each. (Shadow and Sun were added later to make a fifth to two of those spheres) I think it was a tough call to decide which mana types should go in Divination and which in Alteration. In fact there are few mana types which fit the definition of Divination and too many that fit Alteration, but it makes sense to have the four spheres balanced. Perhaps a name other than Divination could have been used, but that name is well-grounded in fantasy circles. No doubt any other name for a magical sphere would also invite some disagreement about what mana types fit that name.

 

[avalonnn]

Fix:
CIV4UnitInfos.xml , row 13169
move this line

<PythonPostCombatLost>postCombatExplode(pCaster,pOpponent</PythonPostCombatLost>

up to row 12934:

        </DamageTypeCombats>
        <PythonPostCombatLost>postCombatExplode(pCaster,pOpponent</PythonPostCombatLost>
    </UnitInfo>
    <UnitInfo>
        <Class>UNITCLASS_AXEMAN</Class>
        <Type>UNIT_SONS_OF_ASENA</Type>

Further needed fixes in version "o":

Chariots are (partly) missing -25% Ancient Forest, -25% all forest

<class>UNITCLASS_CHARIOT</class>
*<Type>UNIT_CHARIOT</Type>
*<Type>UNIT_CHARIOT_SCORPION_CLAN</Type>

<FeatureAttacks>
    <FeatureAttack>
        <FeatureType>FEATURE_FOREST</FeatureType>
        <iFeatureAttack>-25</iFeatureAttack>
    </FeatureAttack>
    <FeatureAttack>
        <FeatureType>FEATURE_JUNGLE</FeatureType>
        <iFeatureAttack>-25</iFeatureAttack>
    </FeatureAttack>
    <FeatureAttack>
        <FeatureType>FEATURE_FOREST_ANCIENT</FeatureType>
        <iFeatureAttack>-25</iFeatureAttack>
    </FeatureAttack>
</FeatureAttacks>

Animals missing flags:
Baby Spider -> <Invisible>INVISIBLE_ANIMAL</Invisible>
Giant Spider -> <bNoDefensiveBonus>1</bNoDefensiveBonus>
Griffon -> <bNoDefensiveBonus>1</bNoDefensiveBonus>

Strange flag(s) on units:
Herald (Mercurians) -> -20% CityAttack ... dunno why? (from -25% in "n" IIRC)

1) Herald is the Mercurian beastmaster, recon unit, - city attack

2) cannot find the unitinfo file. what subfolder?

-- never mind found it in the XML folder, units

3) that explains the baby spiders suddenly appearing!

 

[kanban42]

Hi Tielby (sorry for this "cross-threading"),

In my humble opinion, the 4 initial magical sciences (Elemental, Divination, Alteration and Necromancy) had to be balanced to allow access to four types of mana each. (Shadow and Sun were added later to make a fifth to two of those spheres) I think it was a tough call to decide which mana types should go in Divination and which in Alteration. In fact there are few mana types which fit the definition of Divination and too many that fit Alteration, but it makes sense to have the four spheres balanced. Perhaps a name other than Divination could have been used, but that name is well-grounded in fantasy circles. No doubt any other name for a magical sphere would also invite some disagreement about what mana types fit that name.

I'm not sure I understand that. In my version, there are 18 types of mana. If you subtract Metamagic and Ice, the rest could be divided evenly among the four initial magical sciences. In my version, Divination enables five types of mana while Alteration enables three. Each magical science should enable the four types of mana that are necessary to build its wonder, I think.

I wonder if there's something wrong with the update process because such an imbalance seems very obvious.

(ModMods like Fall Further add another three types of mana. Leaving out Metamagic, each magical science enables five types of mana. They filled up Alteration and moved Life mana to Divination. Maybe that could explain the problem - vanilla FfH2 just lacks these additional types of mana but still has Life mana in the same place.)

 

[deanej]

That might be it. Base FFH was supposed to have them but nobody was able to come up with spells for them that Kael liked before he stopped development.

 

[swordsciv55]

sorry i havent posted in a while. i got distracted with another game for some time. anyway im having another problem with crashing, this one at turn 675

 

[Breunor]

sorry i havent posted in a while. i got distracted with another game for some time. anyway im having another problem with crashing, this one at turn 675

Ok, try this. The bug is listed in the spoiler:

Spoiler :

Once again, it is the Mercurians and the multiple mage problem. Maybe it is a greater problem with the Mercurians? Anyway, there were 5 mages. I deleted 3. With 16 turns left in the game it hardly matters. Anyone adventuresome can try deleting just one or 2 mages and see if it crashes. This is a very large file, it takes quite a while to experiment.

Best wishes,

Breunor

 

[swordsciv55]

i crashed again at 13 turns left