AND2 and SVN Bug Reports - A New Dawn 2 ONLY

[dudeSDCA]

About the Storms, this has been reprted in C2C as well. But Dancing Hoskuld says it's related to the Storm Events and not the BUG Storm Option. I'm not entirely sure if that's 100% true or if it's the BUG Option.

If you can remember next game after you set the Options when the new game starts hold the Shift Key down while the Mod loads. After it loads go into BUG and See if No Storms is checked or unchecked. If it's either way set it to No Storms (checked) and then save that turn twice. And Yes, this is the Only known surefire way to save BUG Options.

Then play awhile and see if the Storms keep popping up. Also when you spot one check the Log in game to see if a Storm Event has occurred.

Lots of stuff to remember but really doesn't take much time to do at all.

Now for your UN vs Apostolic Diplo Victory, I'm not sure what is happening there. I rarely use either and so I've missed that. Another one for the List.

Thanks for reporting.

JosEPh

Thanks - I think the UN issue has to be in the DLL -- I can't find anything in the XML that looks off.

I commented out all the storm events... we'll see if that clears the issue. I'll also clear cache and see how BUG looks. Thanks for the suggestion!

 

[dudeSDCA]

Minor question: love the new speed-dependent treaties/agreements. However, one side effect -- advance start on marathon starts with 40 turns of peace now... if playing a small map or want to play early-rush, well, 40 turns of peace is a long time. Especially with advance start.

 

[JosEPh_II]

But isn't that a fair cost for a slower game speed?

JosEPh

 

[dudeSDCA]

I don't have a problem with it... just seemed like an unintended side effect, not necessarily WAD...

 

[Sgtslick]

So your saying advanced start causes your civ to sign a peace treaty? Normally you meet a civ and you have the option to be friends or not etc. But your saying that it automatically signs peace?
I don't play advanced start so idk.

 

[JosEPh_II]

Most Advanced Starts have a "Grace" period before you can become aggressive. It's been ages since I've done one. Especially since you really can't in C2C, but you can in RoM/AND.

40 turns in an Adv Start in a Mid to late Era on Marathon isn't that long. You probably will only be getting your 1st tech researched by then. And maybe found your second city. But if you try to Rush a neighbor you will have spent most of that time building Mil Units not settlers or workers.

JosEPh

 

[45°38'N-13°47'E]

Thanks - I think the UN issue has to be in the DLL -- I can't find anything in the XML that looks off.

Yes, the relevant code is in the dll and python but previous changes shouldn't have caused the problem you're reporting. I'd really need a savegame to check that; if it happens to you again (or it happens to someone else), please report it here with a savegame. I suspect the problem might be related to the UN option at the start of the game rather than the victory conditions, but until I can test it, it's just my guess.

 

[45°38'N-13°47'E]

After some test, I'm more convinced the problem lies in the UN option which in fact was introduced to be used with Mastery Victory IIRC (and it works perfectly with MV); using UN option by itself disabling diplomatic victory causes this problem, but I'll have to dig deeper to solve this issue.

 

[Vokarya]

I've been playing a few games with Start as Minors turned on, and I noticed there seems to be a problem with how the AI is set up to deal with this. The AI interprets the "Always War" part of minor civilization status as "go on the offensive", but at that time, the civilization doesn't have the offense to make anything really happen. All that results is a stack of a couple Warriors and maybe some other units at best camped out on a good defense tile and at worst slaughtered on flat ground by a Surround & Destroy. Would it be possible to have the AI adopt a more defensive posture while it is still in minor civilization status?

 

[45°38'N-13°47'E]

I've been playing a few games with Start as Minors turned on, and I noticed there seems to be a problem with how the AI is set up to deal with this. The AI interprets the "Always War" part of minor civilization status as "go on the offensive", but at that time, the civilization doesn't have the offense to make anything really happen. All that results is a stack of a couple Warriors and maybe some other units at best camped out on a good defense tile and at worst slaughtered on flat ground by a Surround & Destroy. Would it be possible to have the AI adopt a more defensive posture while it is still in minor civilization status?

I frankly don't know. I suppose it's possible but I really don't know how to do it. Everything related to AI behaviour is a very complex matter and I'm not so good at coding (yet, at least). My idea was to import AI behaviour code from C2C as it's surely more advanced than AND code, and in fact part of the code was already imported. But it's a loooong task. So I guess the answer is "No, for the moment", I'm sorry.

 

[45°38'N-13°47'E]

Reporting minor bug in v613 (However, I started the game in v598ish).

Only victory condition was Conquest.
- Building Apostolic Palace triggered election (under Diplomatic victory) - however, as expected, the game continued (no Diplomatic Victory)

- HOWEVER, I turned on the UN option (I like the UN resolutions). Never caused Diplomatic Victory before... but somehow I won Diplomatic Victory even though that wasn't an option.

I went into the new game options to double-check, and yes, Diplomatic was still turned off.

Previously, if Diplomatic Victory is turned off, no one should get the option of the U.N. Election (Diplomatic Victory), but people can still be the Secretary General and pass resolutions.

I would attach saved games, but I screwed around in WB a few times trying to discover a CTD (I think something with the mini-map...but not sure).

Maybe this is due to changes to victory conditions in the DLL that came in around rev 605?

I've done some more testing and these are my conclusions:
- Building Apostolic Palace triggered election but the game continued because it doesn't grant a Diplomatic Victory, so that's ok.
- Building UN triggers UN Elections for Secretary General (and that's ok) but then shouldn't trigger Diplomatic Victory vote, but in fact it does.
What I'm not sure of, is whether this problem was there in previous versions but my guess is that it was. IIRC United Nations option was made to be able to use UN resolution with Mastery Victory. And in fact it does work perfectly with MV. Even if you get the chance to vote for a Diplomatic Victory and you win, you won't win if MV is enabled (Diplomatic Victory vote is still there, but it's useless). Something similar should happen when United Nation option is chosen, even if MV isn't selected. But in fact it doesn't happen. I'm digging the code to understand if I can apply MV trick to other games with different victory conditions as well. In theory, the best solution should be that when you chek "United Nation Option" you only get the Diplomatic Victory working if "Diplomatic Victory" was selected.

I'm currently testing some changes to United Nations and I will try to solve this issue on victory conditions too, but I'm not sure it's possible without rewriting a long part of the code.

Edit: I've looked again at the code and the only part changed regarding victory conditions was in the Mastery Victory section, so it wasn't that causing this issue. It's probably been there for a long time before dudeSDCA discovered it.

 

[Vokarya]

I frankly don't know. I suppose it's possible but I really don't know how to do it. Everything related to AI behaviour is a very complex matter and I'm not so good at coding (yet, at least). My idea was to import AI behaviour code from C2C as it's surely more advanced than AND code, and in fact part of the code was already imported. But it's a loooong task. So I guess the answer is "No, for the moment", I'm sorry.

OK. It's not particularly game-breaking at the moment, just something I've seen several times.

 

[JosEPh_II]

SaD still has it's AI problems too according to Koshling. It's on his list last I heard, but not a top priority.

So much AI work he has done it's tough to keep up with it all. Too bad we couldn't "hire" him for a week!

Or find someone with similar skills to help merge the good AI work from C2C into AND.

JosEPh

 

[mbrady10]

Storms: Its definetly and event issue.

In regards to importing stuff from C2C, I would very kindly ask that caution be used. Why? While I'm all for things that improve ai behaviour, ATM there is just no comparison to the turn times of C2C and AND. I am not bagging C2C at all, becauae it is probably doing more things behind the scene, but on the huge maps I play, at similar stages through the game (modern era) AND has amazing 15-20 sec, while C2C is heading towards 2 and a half minutes. For this fact alone, AND has become the only big mod that I can play. Amazing.

 

[JosEPh_II]

The importation is Not assets but Dll AI improvements. C2C is almost 2.5+GB in size, while AND is still around 1GB just because of C2C's content.

As for Storms yes it's been reported in C2C as well. But Ori is not available and he did the SDK and Python work on the Events that ae used in AND and C2C.

JosEPh

 

[Vokarya]

As for Storms yes it's been reported in C2C as well. But Ori is not available and he did the SDK and Python work on the Events that ae used in AND and C2C.

JosEPh

I don't think it would be difficult to get the Storm Events to work properly with the No Storms option. Every event can have a Python callback that allows it to check conditions that aren't available through the XML (for example, the Blessed Sea quest event will not trigger if One City Challenge is on). I think I see how the Python checks BUG Options as well. So if we set up callbacks that tell the events not to trigger if the No Storms option is on, that should stop them. Do you think this is viable?

 

[45°38'N-13°47'E]

I don't think it would be difficult to get the Storm Events to work properly with the No Storms option. Every event can have a Python callback that allows it to check conditions that aren't available through the XML (for example, the Blessed Sea quest event will not trigger if One City Challenge is on). I think I see how the Python checks BUG Options as well. So if we set up callbacks that tell the events not to trigger if the No Storms option is on, that should stop them. Do you think this is viable?

Good suggestion, I'm trying it now and I'll see how it works. Basically I've changed Civ4EventInfos.xml for every Storm Event (there are 4 of them) using <PythonCanDo>canTriggerStorm</PythonCanDo>
and defining canTriggerStorm in CvRandomEventInterface.py

def canTriggerStorm(argsList):
kTriggeredData = argsList[0]
map = gc.getMap()

if GC.getGameINLINE().isModderGameOption(ModderGameOptionTypes.MODDERGAMEOPTION_NO_STORMS):
return false
return true

it should work, I'll do some test.

 

[Istanbul_Kop]

I've just 'upgraded' to Windows 8.

Installed the SVN again like in Windows 7 and got everything set up.

However, when I run the mod from within BTS...the game reboots (like normal) then crashes:

Here is the crash report:

Problem Event Name: APPCRASH
Application Name: Civ4BeyondSword.exe
Application Version: 3.1.3.1
Application Timestamp: 48317e8a
Fault Module Name: ntdll.dll
Fault Module Version: 6.2.9200.16384
Fault Module Timestamp: 5010ae7a
Exception Code: c0000005
Exception Offset: 0004f44d
OS Version: 6.2.9200.2.0.0.256.103
Locale ID: 2057
Additional Information 1: 7969
Additional Information 2: 79696b5d2f258950c757ee4efcabcb64
Additional Information 3: 721e
Additional Information 4: 721e54d3e296196931ed445390396e91

ps. I even had copied my old SVN folder over from my WIN7 drive that still works fine on my WIN7 dual boot, but I get the same error.

 

[JosEPh_II]

Application Version: 3.1.3.1

Might want to patch to 3.19.

JosEPh

 

[Vokarya]

Good suggestion, I'm trying it now and I'll see how it works. Basically I've changed Civ4EventInfos.xml for every Storm Event (there are 4 of them) using <PythonCanDo>canTriggerStorm</PythonCanDo>
and defining canTriggerStorm in CvRandomEventInterface.py

def canTriggerStorm(argsList):
kTriggeredData = argsList[0]
map = gc.getMap()

if GC.getGameINLINE().isModderGameOption(ModderGameOptionTypes.MODDERGAMEOPTION_NO_STORMS):
return false
return true

it should work, I'll do some test.

That's exactly what I was thinking, but it looks to me like there are FIVE Sea Storm events. I'm not quite sure what the difference between them is, but 3-4-5 look to have slightly different chances of sinking the ship.