1.9.6 Feedback

Love the mod mod :goodjob:

One thing that I always see is that the AI seem to make little to no effort to settle and use spice, and they routinely delay the Spice Extraction tech for extended periods too.
Even quite late in the game they seem to leave spots that could claim extra spice unsettled and don't improve spice tiles :crazyeye:
Would it be possible to increase AI decision weighting towards spice settling and getting the tech?

Also, what does the AI base building decisions on? I hadn't thought of using WB to check till just now and my only active game is passed turn 200 so they had time to build many buildings. Even this late though several AIs haven't one of a monastery or a refinery in the capital yet, and a fair number of smaller cities have nurseries and dew extractors which are contributing nothing but -1 water.
The mod favours very heavily commerce development of the capital early due to the power of spice, how do other players see the AIs performance with this?
 
Even quite late in the game they seem to leave spots that could claim extra spice unsettled
This isn't necessarily a bad policy. Extra cities add significantly to upkeep costs (city maintenance and civic upkeep), and spice is temporary, so if there aren't good water resources near by, a city can easily be net drain on the economy.

AI settlement policy in Dune Wars is heavily driven towards water resources, with other resources and spice as secondary draws, and it likes sinks as a tertiary draw. I think this is appropriate, it leads to the AI picking pretty decent city sites; until this code was re-written (ie back when it was still using vanilla values) the AI settlement pattern was terrible; it packed cities far too closely and had lots of weak cities that would never grow. The most significant shift from vanilla was to get the AI to stop valuing tiles that were already within the BFC of an existing city.

I think the main thing slowing the AI building harvesters is worker time; it prioritizes water improvements before spice, which is probably also appropriate. I'm not sure how else things can be changed, there is a risk that having them focus too much on spice would leave them without sufficient military or without sufficient growth.

I don't know enough about how the AI picks buildings to comment; it is basically just using the vanilla Civ4 AI.
 
Could anyone tell me where I find the relation between unit class and category? Example: "Hardened Bladesmen" belong to the category "melee". How can I change them to, let's say, "thopter"? (I know, bad example :D)
 
in the unit xml (not unit class), for each unit you'll find towards the top the tag <combat> which defines it, so for a melee unit it says:

Code:
<Combat>UNITCOMBAT_MELEE</Combat>

and you would change that unit to a thopter unit by changing it to:

Code:
<Combat>UNITCOMBAT_THOPTER</Combat>
 
Is it intended that Bladesmen (other than Hardened Bladesmen) do not receive the "ignore city defenses" bonus? Noted that when attacking a fortified city. Bladesmen: 10 % chance of victory. Same Bladesmen upgraded to Hardened Bladesmen: 90 % chance of victory.
 
Is it intended that Bladesmen (other than Hardened Bladesmen) do not receive the "ignore city defenses" bonus?
All melee units should ignore city defenses (note that this tag means they ignore city walls - I forget the precise name - and force shields, not cultural defenses). Are you sure they don't?

Bladesmen: 10 % chance of victory. Same Bladesmen upgraded to Hardened Bladesmen: 90 % chance of victory.
This seems potentially possible just from the strength differential alone; the hardened bladesman has 40% higher strength, and the Civ4 combat calculator can be very sensitive to strength ratio differentials.
 
All melee units should ignore city defenses (note that this tag means they ignore city walls - I forget the precise name - and force shields, not cultural defenses). Are you sure they don't?

Damn. Checked this again, and it seems you are right. Which is actually a good thing ;)
 
After some time of distraction, I too am back playing this mod, and I have to say it fits like a glove. Just wish to express my sincere gratitude to all that have worked on it thus far. Great job guys, your awesome.

I'm enjoying the updated version, looking forward to some more pedia work.
 
the pedia has some holes. the list of promotions in the promotions section ends after the letter "D" and if i click on any promotion no info pops up in the main pedia frame

I think I figured out why the Promotions pedia is broken in 1.9.5 and 1.9.6. It is to do with the iMaxPlayerInstances attribute on PROMOTION_DUCAL_GUARD and PROMOTION_ORDER_OF_AGAMEMNON - a C++ exception is thrown when trying to display any promotion that has this attribute. If you remove this restriction from the two promos in the XML then the Promotions pedia works fine.

My guess is that it because the code below in CvGameTextMgr.parsePromotionHelp, which is used in the pedia when you jump to a promotion, doesn't work because there is no ActivePlayer - you can view the pedia without a game being in progress. This is the only code in parsePromotionHelp to reference the ActivePlayer. We should change the code to just report the MaxPlayerInstance value when there is no ActivePlayer.

Code:
	if (GC.getPromotionInfo(ePromotion).getMaxPlayerInstances() > 0)
	{
		CvUnit* pLoopUnit;	
		int iLoop;
		int iPromoCount = 0;	
		PlayerTypes eActivePlayer = GC.getGameINLINE().getActivePlayer();
		CvPlayer& kPlayer = GET_PLAYER(eActivePlayer);
		for(pLoopUnit = kPlayer.firstUnit(&iLoop); pLoopUnit != NULL; pLoopUnit = kPlayer.nextUnit(&iLoop)) 
		{
			if (pLoopUnit->isHasPromotion(ePromotion))
			{
				iPromoCount ++;
			}
		}
		szBuffer.append(NEWLINE);
		szBuffer.append(gDLL->getText("TXT_KEY_PROMOTION_NATIONAL_LIMIT_LEFT", GC.getPromotionInfo(ePromotion).getMaxPlayerInstances() - iPromoCount));
	}
 
Hey all. I've been working on a new update. I have Chris' latest changes and hopefully I'll be able to release something soon.

I've attached a couple of screenshots to whet your appetite...
 

Attachments

  • Civ4ScreenShot0062.jpg
    Civ4ScreenShot0062.jpg
    146.6 KB · Views: 5,213
  • Civ4ScreenShot0064.jpg
    Civ4ScreenShot0064.jpg
    106.9 KB · Views: 5,220
Cool. I'd remove the core +2 happy from the Missionaria Protectiva; that was always just a placeholder value.
 
This is good news! :goodjob:
Still, I have one question: The old text for Missionaria Protectiva told something about "penalties for having multiple religions". Several posts here in the forum brought up this, too. But I never noticed such a penalty in the game itself (currently 1.9.6). Did I miss something or were such penalties planned, but never implemented?
 
This is good news! :goodjob:
Still, I have one question: The old text for Missionaria Protectiva told something about "penalties for having multiple religions". Several posts here in the forum brought up this, too. But I never noticed such a penalty in the game itself (currently 1.9.6). Did I miss something or were such penalties planned, but never implemented?

Yes, there is no penalty for multiple religions in Dune Wars at present (1.9.6). I am going to introduce a -2 happy penalty for each non-state religion with the forthcoming patch. The Missionaria will negate this penalty - which is actually what you see happening in the screenshot - the 2 unhappy conflict of belief penalty is negated by the +2 happy from the Missionaria. There is no fixed happiness from the building - just the non-state religion bonus.
 
Yes, there is no penalty for multiple religions in Dune Wars at present (1.9.6)
Really? I thought there was 1 unhappy for having non-state religion. But your design with 2 sounds excellent.
 
Ouch. That would allow for missionaries as weapons - have them join your beliefs and they will rebel as mad.
If the AI is bad at using inquisitors, this could be heavily exploited. Will it be possible to turn the penalty off, at least for the AI?
 
Missionaries *should* be a form of weapon (and already are to some extent - look at Qizarate missionaries), and there should be risks from open borders. It isn't that bad though, because getting the extra religions lets you construct the basic building, which counteracts some of the unhappiness.

Extra religions already cause 1 unhappiness in vanilla BTS IIRC, so all you're doing it is increasing by 1, and missionaries are much more restricted in this mod than they are in vanilla, because of the national limits and because some religions don't have missionaries.

I like the mechanic, because of both the flavor purpose (these philosophies are very different, you should have big problems if you have believers of multiple ones), and because it encourages you much more to pick a religion and stick with it rather than trying to spread lots of religions around to get benefits from all of them (one of the things I dislike most about vanilla).

It also helps make BG and BTl more unique, because they don't suffer from this internal dissension.
 
I agree with that. A penalty in exchange for the advantages of having multiple religions is surely welcome. I just fear that the AI will have its difficulties with it. I would never call it an 'exploit' against a human player.
 
I agree with that. A penalty in exchange for the advantages of having multiple religions is surely welcome. I just fear that the AI will have its difficulties with it. I would never call it an 'exploit' against a human player.

There is Python AI code for Inquisitors which should mean that if an Inquisitor is built it should find its way to a city with a non-state religion and purge it. The part I am less sure of is whether the AI prioritises researching Divine Mandate and building Inquisitors if it has any cities with non-state religion unhappiness. It shouldn't be too hard to tweak the AI to take this into account.
 
Top Bottom