Research: Fixing the trait bugs

[therother]

Please post possible fixes for the bugs in the triggers for various traits. For background info on some bugs already identified, please see these threads:

• Traits that are bugged in 1.2 with fixes. Part 1: Scarred, Farmer, Trader
• Would 1.2 be better with traits removed altogether?
• HA, Scarface fixes for download
• Double Combat Traits?

Bear in mind that this is a research thread - please post only fixes that you have tested yourself, or comments directly related to the feasibility of the suggested fixes. Please provide as much information as possible on: the proposed bug you are fixing, how you believe the trigger should work, what you have done to fix it, what parts of the file require editing, and so on. Also note that this is for bugs only, i.e. traits that do not work as CA intended, rather than traits you think could do with improving.

Anything not adhering to these criteria will be removed from this thread.

Eventually, the intention is to come to a consensus for solutions to the bugs, and to release a patch to implement them.

[Kraxis]

I can only urge people to look through the export_descr_character_traits file. There are a lot of small oddities, such as the progression of traits give the wrong bonus. I found one where the bonus dropped in the middle of a 5 step bonus only to jump up to the correct value in the next. Obviously nr 3 was wrong.
Sadly I fixed this on the run and can't remember which this was, but I think Drillmaster is wrong (should be -1 to morale rather than +1). But beware, not all traits progress liniarly, some have a penalty for going too far. So open the text file associated with it (descriptions of the traits).

I have found a small bug with risky battle
This is how it should look.

Trigger V0110_Standard_Risky_Battle_Attacker_Clear_Victory_VnV_Trigger
    WhenToTest PostBattle

    Condition WasAttacker
          and IsGeneral
          and WonBattle
          and BattleSuccess = clear
          and I_ConflictType Normal
          and BattleOdds < 0.5
          and BattleOdds > 0.16

    Affects GoodRiskyAttacker  1  Chance  100

But it looks like this

Trigger V0110_Standard_Risky_Battle_Attacker_Clear_Victory_VnV_Trigger
    WhenToTest PostBattle

    Condition WasAttacker
          and IsGeneral
          and WonBattle
          and BattleSuccess = crushing
          and I_ConflictType Normal
          and BattleOdds < 0.5
          and BattleOdds > 0.16

    Affects GoodRiskyAttacker  1  Chance  100

That is the case for all four Clear_Victory risky battles. That means if you win a clear victory you won'y get any bonus, while if you win a crushing victory you gain in the unmodded game 3 points towards GoodCommander.

Thus change all the crushing to clear, but remember to only do it to the clear battle, not the crushing or else it will only be reversed.

[Pode]

I *believe* the consensus from the other thread was that the vanilla governing10 and governing11 triggers (the ones that control good farmer and good trader tests when a building is finished) were working as designed (WAD), but the thresholds for these traits needed adjustments. Based on the amount of traits generated over 50 years for a governor remaining in place building a town from scratch, I recommended thresholds of 3, 18, and 36 for the levels of good farmer.

One point I'm still unclear on from that discussion, is whether the < = operator is functioning properly. We know this board doesn't like it, but does it work right in game?

[Bob the Insane]

I *believe* the consensus from the other thread was that the vanilla governing10 and governing11 triggers (the ones that control good farmer and good trader tests when a building is finished) were working as designed (WAD), but the thresholds for these traits needed adjustments. Based on the amount of traits generated over 50 years for a governor remaining in place building a town from scratch, I recommended thresholds of 3, 18, and 36 for the levels of good farmer.

Well I guess that was what you posted...

Actually I do not mind that the GoodFarmer traits are really for those who possess farming temples (possibly why the default thresholds are so high).

I think the real problem here are the triggers which cause BadFarmer and BadTrader points to be generated. In the default files you can get BadFarmer and/or BadTrader points simple by building anything that is not a farm or trade building, even if you have already maxed out on farm and/or trade buildings. I believe resolving this would resolve the problems with the GoodFarmer and GoodTrader traits...

[Bob the Insane]

For the Scarface thing... Does this mean we are using:

;------------------------------------------
Trigger battle1
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle < 0.7
and not CultureType roman

Affects BattleScarred 1 Chance 30
Affects Brave 1 Chance 15

;------------------------------------------
Trigger battle1b
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle < 0.7
and Trait Berserker >= 1

Affects Berserker 1 Chance 30

;------------------------------------------
Trigger battle1R
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle < 0.7
and CultureType roman

Affects BattleScarred 1 Chance 30
Affects RomanHero 1 Chance 15

;------------------------------------------

instead of the default:

;------------------------------------------
Trigger battle1
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle > 0.3

Affects BattleScarred 1 Chance 30
Affects Brave 1 Chance 15

;------------------------------------------
Trigger battle1b
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle > 0.3
and Trait Berserker >= 1

Affects Berserker 1 Chance 30

;------------------------------------------
Trigger battle1R
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle > 0.3
and CultureType roman

Affects BattleScarred 1 Chance 30
Affects RomanHero 1 Chance 15

;------------------------------------------

[therother]

For the Scarface thing... Does this mean we are using: [...]

If it is widely regarded as the best solution, I'll put this in as our first settled fix. Do you know who devised the solution?

[Pode]

delete the battlescarred effect from the battle1r trigger. Otherwise a roman general tests for scarring once under battle1 and again under battle1r. There's enough double traits from playing out battles, we don't need more. :)

And you're right, Bob, that was only what I posted. I thought it was the general consensus, and by claiming it as such I thought I could draw out the last of the undecideds :)

[Bob the Insane]

Do you know who devised the solution?

Yikes, who originally thought this one up?? I guess you would have to pour through the forums here and at TWCenter to see who posted it first... Or you could say the resolution developed by the consensus in the RTW community...

delete the battlescarred effect from the battle1r trigger. Otherwise a roman general tests for scarring once under battle1 and again under battle1r. There's enough double traits from playing out battles, we don't need more. :)

Much neater that way...

;------------------------------------------
Trigger battle1
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle < 0.7

Affects BattleScarred 1 Chance 30 
Affects Brave 1 Chance 15 

;------------------------------------------
Trigger battle1b
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle < 0.7
and Trait Berserker >= 1

Affects Berserker 1 Chance 30 

;------------------------------------------
Trigger battle1R
WhenToTest PostBattle

Condition GeneralHPLostRatioinBattle < 0.7
and CultureType roman

Affects RomanHero 1 Chance 15 

;------------------------------------------

[player1]

delete the battlescarred effect from the battle1r trigger. Otherwise a roman general tests for scarring once under battle1 and again under battle1r. There's enough double traits from playing out battles, we don't need more. :)

That's really not needed.
There is already "and not CultureType roman" around first battlescarred check.

Why edit files more then necessary?

[Bob the Insane]

That's really not needed.
There is already "and not CultureType roman" around first battlescarred check.

Why edit files more then necessary?

By default the "and not CultureType roman" is not there, I added it above to try and fix the issue (not orginially my idea I must point out)... Pode was just pointing out that removing the line in Trigger battle1R was more elegant than adding extra lines of code to achieve the same effect...

[Sinner]

I thought it might be helpful to include a few notes on how the triggers work given some of the apparent confusion.

I'll use the probably now familiar GoodFarmer trigger as my example...

Trigger governing5
WhenToTest GovernorBuildingCompleted

Condition SettlementBuildingFinished >= farms

Affects GoodFarmer 1 Chance 100

First we have the trigger's identity, governing5, which is nothing more than a convenient label to make it easier for people to locate a specific trigger. I could rename this trigger as get_badfarmer and it would have no effect on its functionality, equally I could change its name to governing1, duplicating that of another trigger, and it will have no effect - both triggers will still function correctly.

Next is the event when the trigger is tested, GovernorBuildingCompleted. Many events can be determined from the examples already available in the traits file, while others can be found in the Events, Conditions, & Commands thread. It's important to note that only CA can edit existing events or create new events, so we can only work with what is available, no amount of modding will get around this.

In the case of the GovernorBuildingCompleted event, it occurs when a building is completed in a settlement that has a governor.

Next we have the condition, SettlementBuildingFinished. Despite how many of them appear conditions are not mathematical equations. Conditions are functions that return a true or false answer given one or more parameters. Some of these parameters are variables that are automatically passed to the function, and are otherwise hidden, while others have to be specifically declared in the trigger and are thus visible and amendable.

In the case of this trigger, the condition function SettlementBuildingFinished is intended to test a building that has just been completed. It is passed a variable parameter that contains the identify of the completed building, plus two declared parameters - >= and farms - that the function tests that building against. This trigger will return true if the completed building is Land Clearance or any other higher-level farm.

All conditions for a trigger in effect have to return an answer of true for the trigger to cause any results. I say 'in effect' because we have the possibility of 'not' switches, such as and not SettlementBuildingFinished >= farms, which would mean you'd get GoodFarmer for completing anything other than a farm. The 'not' switch simply reverses the true/false answer from the function, so you'd need the function to return false for that condition if the trigger is to cause any results. Confused? You will be!

A trigger can have multiple conditions... for example if we only want Julii characters to gain GoodFarmer when they complete a farm, we'd add and FactionType romans_julii under the Condition SettlementBuildingFinished >= farms line. In the case of multiple conditions, all conditions must return true, after any 'not' switches if appropriate, for the trigger to cause any results.

Finally we have the results caused, Affects GoodFarmer 1 Chance 100 if all conditions are met. In this case the governor's GoodFarmer trait has a 100% chance of gaining 1 point. Obviously if the chance is less than 100% then it's possible that random chance will result in no points gained even if all the conditions have been met. If the points gained are enough to equal or exceed the threshold of a trait level then the character will gain the appropriate effects.

As a note up to 10 traits can be effected by any one trigger, but the trigger can be duplicated and different traits effected if more than 10 are required for the same trigger event and conditions.

[player1]

By the way, has anyone investigated what Camp Freddie has done in regards to Coward trait.

I refer to changes done in his RTWv12_bugfixes patch.

In line:

    Condition not GeneralFoughtInCombat
          and PercentageEnemyKilled = 0

    Affects Coward  1  Chance  10

Add not before "PercentageEnemyKilled"

[player1]

Also, he made some changes to shieldbearer and heroic saviour ancillaries, since they also use GeneralHPLostRatioinBattle (and were maybe bugged).

[Bob the Insane]

Also, he made some changes to shieldbearer and heroic saviour ancillaries, since they also use GeneralHPLostRatioinBattle (and were maybe bugged).

These are definately still broke as they are still expressing a percentage...

i.e. GeneralHPLostRatioinBattle >= 80

[player1]

These are definately still broke as they are still expressing a percentage...

i.e. GeneralHPLostRatioinBattle >= 80

I guess that one should be:
GeneralHPLostRatioinBattle < = 0.2
(or seriously wounded general)

[player1]

Ok, I just realised that fix for Coward trait, done in Camp Freddie bugfix patch is not working.

It appears that GeneralFoughtInCombat condition is buggy.

So I tried instead of:

Trigger battle4
    WhenToTest PostBattle

    Condition not GeneralFoughtInCombat
          and PercentageEnemyKilled = 0

    Affects Coward  1  Chance  10

this:

Trigger battle4
    WhenToTest PostBattle

    Condition GeneralNumKillsInBattle = 0
          and PercentageEnemyKilled > 0

    Affects Coward  1  Chance  10

"GeneralNumKillsInBattle = 0" is the closest thing to "not GeneralFoughtInCombat" that works.

Also there is a > instead of = for "PercentageEnemyKilled", which is close to original fix done by Camp Freddie (to prevent getting the trait when enemy runs away, giving you no chance to fight it).


I tested this with "Affects Coward 8 Chance 100", when general properly got highest coward trait anytime when not fighting battle.

Only in case when enemy lost zero units trigger didn't started (tested by retreating at start of the battle).

[Kraxis]

Have I understood it correctly that Coward with the old fix is impossible?

[player1]

Since GeneralFoughtInCombat will always be true, you won't get Coward with such trigger.

P.S.
Althouhg there is another trigger about general routing that can give Coward.


And of course, my fixed version of this trigger.

[player1]

If anybody is interested I made fixed version of coward and scarred trait at Scarface thread at twcenter:

http://www.twcenter.net/forums/index...&id=1811983775

[therother]

Have I understood it correctly that Coward with the old fix is impossible?

As I understand it, it will be impossible with Camp Freddie's fix if you fight all battles on the battlemap, as the condition GeneralFoughtInCombat is always true. However, it should be possible to get it from an autocalc battle.

If anybody is interested I made fixed version of coward and scarred trait at Scarface thread at twcenter

Any chance you could compile the changes made into one post?

[player1]

Ok, I'll try.
It's pretty much slightly modified version of Camp Freddie fixes to export_descr_character_traits.txt

these are modified battle1 triggers:

;------------------------------------------
Trigger battle1
    WhenToTest PostBattle

    Condition GeneralHPLostRatioinBattle < 0.7

    Affects BattleScarred  1  Chance  30 
    Affects Brave  1  Chance  15 

;------------------------------------------
Trigger battle1b
    WhenToTest PostBattle

    Condition GeneralHPLostRatioinBattle < 0.7
          and Trait Berserker >= 1

    Affects Berserker  1  Chance  30 

;------------------------------------------
Trigger battle1R
    WhenToTest PostBattle

    Condition GeneralHPLostRatioinBattle < 0.7
          and CultureType roman

;    Affects BattleScarred  1  Chance  30 
    Affects RomanHero  1  Chance  15 

;------------------------------------------

and modified battle4 trigger:

;------------------------------------------
Trigger battle4
    WhenToTest PostBattle

    Condition GeneralNumKillsInBattle = 0
          and PercentageEnemyKilled > 0

    Affects Coward  1  Chance  10 

;------------------------------------------

[player1]

quote the message if you wanna see correct formating

[Bob the Insane]

For the coward thing I am using:

;------------------------------------------

Removed due to being stupid...  ~D   

;------------------------------------------

If I follow this correct this is triggered if the general was on the battlefield but was withdrawn (GeneralFoughtInCombat returns false) and none of the enemy was killed...

From what I can see you should only get the coward trait when you go to the battlemap, if you withdraw from the strategic map a lot you should get IndecisiveAttacker instead. Is this what people see?

[player1]

You need to realise that after some testing I'm confident to say that GeneralFoughtInCombat is always true in battle.

But, if you do autocalc, it would be only true if general didn't actually fighted.
So GeneralFoughtInCombat is bugged. Use some other trigger instead.

Also, what you use (with added not), will trigger coward when doing autocalc and some enemies get killed, but not by general.

Better replace it with GeneralNumKillsInBattle = 0, so it works for both autocalc and manual battle.

[Bob the Insane]

player1 - I understand where you are coming from and I understand the problems you have pointed out... (I have removed the code I suuggested to avoid confusng people)...

But, one thing I noted with the example you provide, it will not trigger coward for simply withdrawing from the battlefield immediately (on the battlemap). Where GeneralNumKillsInBattle = 0 and PercentageEnemyKilled = 0 (I presume)... Was this intentional?

[player1]

The problem is that if you leave that, then you could get coward trait when enemy withraws right from start (and not just you).

[Pode]

Try adding a not WonBattle condition

[Camp Freddie]

But, one thing I noted with the example you provide, it will not trigger coward for simply withdrawing from the battlefield immediately (on the battlemap). Where GeneralNumKillsInBattle = 0 and PercentageEnemyKilled = 0 (I presume)... Was this intentional?

Player1's fixes to my attempt at solving the coward bug seem to be the best option. Generalfoughtincombat seems to be totally buggered. This could be affecting the Heroic Saviour ancilliary too, which should be changed to GeneralNumKillsInBattle > 0.

To answer your question, it will not trigger if you withdraw, BUT that isn't what the coward trait is about.
There's another trait (indecisive attacker) and 3 triggers which cover withdrawing before combat (100% chance when you are the attacker, 50% chance when odds are 0.8-1.5 and a 50% chance for a 2-step increase when odds >3). It has huge thresholds though (8, 16, 32) and has goodcommander and goodattacker as antitraits, so it's hardly ever seen.
Strangely, if odds are 1.5-3 and you aren't attacking, it's safe to withdraw - another trait glitch!
Therefore, in Trigger V0200, the > 3 should be reduced to > 1.5. If you want coward traits to be added, then put "coward 1 chance 50" in Trigger V0200, which would now fire when odds >1.5 and you withdraw. You could put "coward 1 chance 10" in Trigger V0190 as well, if you want to really persecute generals who avoid battle when they have a good chance of winning!
The only problem is that I don't know how often the AI general withdraws. If he withdraws a lot (he did in MTW) then he could be heavily penalised by coward traits. That might even be the reason why "indecisive attacker" has such high thresholds.

[player1]

Well, it's interesting to point that when using autocalc, if having enough troops, generals rarely fight, which already increases chances of gaining this coward trait for AI.

So no need to add more futher.
Fixed Coward trigger does even more then enough.

[player1]

By the way, Heroic Savior works fine, since one of the conditions is already generalnumberofkills.