Dune Wars

Back around post 50-55 in this thread, we were discussing about the Force Shield suddenly becoming so huge, it blocks the view of the city. I had this happen in a game just now. Of course my local build will make save games incompatible with anybody else's. But I experimented with turning on and off buildings in WB. I think this is reproducible. In your city, build a few buildings and a force shield. Now add the Oracle of Hajra. Turning the OoH on and off makes the force shield turn infinitely large and back again. It does not happen if the city is otherwise empty, and I am not sure exactly what combinations of buildings does this. Probably any random buildings to fill up some space.

I have also noticed that with a different set of buildings present, turning on OoH causes the force shield to disappear. Not become really large, just vanish. I noticed the same with Observatory. But the OoH is the only one I saw turn the force shield huge.
 
Ahriman said:
Your economy tuning needs to remove the whipping/drafting, and then try testing.
Click to expand...

I know we have been through this before, but this tuning would be *much* more efficient if you chose to edit xml files on your own. Then you could directly try out your suggestions instead of waiting for me to edit the file for you.

For example, slavery is controlled by assets/xml/gameinfo/civ4civicinfos.xml. Turn this:
Code: 
	<Hurrys>
		<Hurry>
			<HurryType>HURRY_POPULATION</HurryType>
			<bHurry>1</bHurry>
		</Hurry>
	</Hurrys>
Into this, in the three places where it appears:
Code: 
	<Hurrys/>
In xml you do not need to worry about white space, but it is "nice" to preserve tabs.
 
Is unit maintenance the same for every type of unit? I've never looked into it.
 
Into this, in the three places where it appears:
Click to expand...

*three* places? I saw one for slavery and one for serfdom. What is the third? I left the gold hurry as is.
And I set the draft for imperial fealty from 3 to 0.

Once the autoplays are done, how do I view the economy output of each player?

Is unit maintenance the same for every type of unit? I've never looked into it.
Click to expand...

Yes. Well, I think it can be different for workers, but its the same for every combat unit unless you deliberately change it.
 
You need to have the cheat code set to equal 'chipotle' in your Civilization.ini, if not done already. Then you can do Ctrl-Z at any stage to reveal the whole map and see every players economic breakdown in the Financial Advisor by changing the name in the drop-down.
 
AHR30

From 250 turns of autoplay:

I'm not sure how to see the economies of the other players, but for the corrino player (probably a bad choice since they tend not to use slavery or serfdom):

Without whipping, economy was:
Spoiler :




And this was second to last in terms of score:
Spoiler :




With the original file (ie whipping allowed) from the same initial autosave (so same map, same start locations), economy was:
Spoiler :




And scores were:
Spoiler :




So: 92 worked tiles on turn 250 with the whip and draft disabled, with 47 worked tiles with the whip and draft enabled.

And the only reason that the economy-with-whip is even close to the same size, is because with the whip the corrino player happened to control more spice tiles (which is the result of random luck), and somehow had a large "building income" (a shrine?).
Their income from worked tiles was only *half* as high with the whipping.

The reason that I didn't test this before was that it was so obvious to me what the result was; I wasn't saying that someone else should suggest it to see if we should remove the whip/draft, I was saying that it was obvious that the whip/draft was a problem and that we needed to fix the AI or remove it.
 

Attachments

  • NowhipCorrino.jpg
    NowhipCorrino.jpg
    120.4 KB · Views: 350
  • NowhipCorrinoScore.jpg
    NowhipCorrinoScore.jpg
    175.3 KB · Views: 326
  • corrinowhip.jpg
    corrinowhip.jpg
    123.5 KB · Views: 352
  • corrinowhipscore.jpg
    corrinowhipscore.jpg
    144.9 KB · Views: 325
AHE30

With whipping and draft DISabled:

Atreides: Worked tiles = 96, commerce = 348
BTL: Worked tiles = 104, commerce = 482
BG: Worked tiles = 79, commerce = 508
Ordos: Worked tiles = 139, commerce = 762
Harkonnen: Worked tiles = 36, commerce = 222
Ecaz: Worked tiles = 84, commerce = 424
Corrino: Worked tiles = 92, commerce = 358

Average: Worked tiles = 90, commerce = 443


With whip + draft ENabled, we have:

Atreides: Worked tiles = 102, commerce = 471
BTL: Worked tiles = 111, commerce = 594
BG: Worked tiles = 40, commerce = 253
Ordos: Worked tiles = 77, commerce = 573
Harkonnen: Worked tiles = 74, commerce = 270
Ecaz: Worked tiles = 71, commerce = 302
Corrino: Worked tiles = 47, commerce = 331

Average: Worked tiles = 75, commerce = 399

So, with whipping ENabled, we have an average 20% reduction in population, and 10% reduction in commerce, relative to what we get with whip/draft DISabled.
Average score is ~6% higher with whip/draft DISabled.

The mean is misleading, because some factions (eg Atreides) do not adopt slavery, and so maintain strong economies even with slavery whipping enabled.
But whipping halves the size of the Ordos economy and the Bene Gesserit economy.
(Harkonnen economy is stronger with whipping here, because they suffer less from whipping because of their slave pit building, and because in the no-whip case they were invaded and lost a lot of land).
 
Thanks for the information. Perhaps it is clear, or not, but I am hoping you can set up a more direct feedback loop. Today, the process goes like, "High level improvements should cost more". Pause. New version with higher costs. Pause. "Now they cost too much". Pause. New version with slightly lower costs. Pause. Repeat. When I am trying to do this tuning, I make several intermediate local versions and discard many of them. I am encouraging you to do the same.

You have suggested raising the cost of this, lowering the cost of that, adding this yield, and seeing what happens. I am certainly grateful for the feedback, but if we had two development paths at once, I could focus on terraforming, and the religion rewrite, and several other things besides tuning the economic parameters over again. I would love to help by explaining bits of xml or where the existing explanations are, and I can package up and merge all the changes once you like them.
 
One thing we would like to do is make slavery a less interesting civic to adopt, or possibly make whipping less interesting once the civic is adopted.

I had a look into the AI code for both decisions. For adopting slavery, probably it gets a high weight because of the +1 hammer on spice harvesters. One of the value terms multiplies the hammer bonus by the number of improvements, which is high. So removing the hammer bonus for harvesters would probably make slavery adopted less often.

For choosing whipping, I do not have a clear view. There are about 20 "if" conditions for different reasons to hurry, like if there is an ocean threat hurry an ocean unit, if in need of gold hurry a gold building, etc. In order to determine why the AI is hurrying, we need to know what units it is building when it hurrys. There is no easy way to find this out. One possibility is to litter the decision function with print statements to see which ones happen. But the function is called once per turn per city, so there would be a large volume of stuff to look through. The related function is AI_doHurry in CvGameCoreDLL/CvCityAI.cpp.
 
I'm sorry, but I'm just not going to have the time to be doing testing, at least over the next few weeks. Maybe after that.

The feedback is just that.

And if you're honest, there are not too many changes where my suggested changes have been undone, and in most of those that there have been, it has been because of other changes at the same time. So for example, a much longer build time for aquabores was justified when they were giving ~+7 water or so, but not when they are giving +4 or +5.

Cottages at just +1 commerce were too weak, but cottages at +1 commerce, +1 hammer with fresh water are too strong when 2/3+ tiles have fresh water access.
 
AHR30

One thing we would like to do is make slavery a less interesting civic to adopt, or possibly make whipping less interesting once the civic is adopted.

I had a look into the AI code for both decisions. For adopting slavery, probably it gets a high weight because of the +1 hammer on spice harvesters. One of the value terms multiplies the hammer bonus by the number of improvements, which is high. So removing the hammer bonus for harvesters would probably make slavery adopted less often.
Click to expand...

I agree that making slavery less attractive is probably a good way to go: I suspect that one reason whipping isn't a big problem in vanilla is that the AI's move out of it as soon as they get something else. Removing the +1 hammer from harvesters is trivial.

We can also help by giving the civic passive boosts that the AI doesnt' understand, like giving it a 25% slave-generation chance. I suggested elsewhere reducing the Harkonnen slave-capture chance down to say 50%, and then giving anyone who adopts slavery civic a +25% slave-capture chance. This makes the civic stronger, but doesn't make it more desirable for the AI.
 
Some 1.5.4 thoughts:
AHE31
Polar desert waste is not attractive enough, given that you can't build improvements on it.

AHR32
Cottages and turbines are still buildable on mesa, making mines not worth bulding.

AHE33
Shai-hulad is spreading waaay too fast. Particularly without any competitors for a while. Shai hulad should have almost no spread without missionaries, and missionaries should be expensive.

I'm guessing that the religions are just generic atm, but we will need to seriously tone the spread rate down.

Also, the start techs don't match those in the updated design, I brought some of them earlier.
http://forums.civfanatics.com/showpost.php?p=8449556&postcount=37

I'm guessing you also deliberately left the Tleilaxu one later (it was intended to be founded at game start) since you don't yet have a way of preventing it spreading everywhere or others adopting it.

AHR34
Something is strange with fresh water irrigation; I have a cottage getting the irrigation bonus despite not being next to a well (or being Paradise civic). See screenshot.
Spoiler :




AHR35
The salt pan desalination comes a little early. There should be a long period of time where the pans are unusable. I suggest the Desert Industry tech should be required to build the desalinate sink improvement.
 

Attachments

  • cottage.jpg
    cottage.jpg
    193 KB · Views: 323
I've been trying to come up with new names for the 7 Eras. The best I have so far is to name the 7 Eras after stages in the life of a Fremen using Fremen terms:

1) Ayat - The signs of life
2) Mihna - The season for testing Fremen youths who wish admittance to manhood
3) Hajra - The journey of seeking
...
...
6) Tahaddi - Fremen challenge to mortal combat, usually to test some primal issue
7) Alam Al-Mithal - The mystical world of similitudes where all physical limitations are removed

Obviously I have gaps. Perhaps these can be filled from the Dune Appendix.
 
I've been trying to come up with new names for the 7 Eras. The best I have so far is to name the 7 Eras after stages in the life of a Fremen using Fremen terms:
Click to expand...

Sounds good.

Other possibilities:
Wali - an untried youth
Amtal - Testing to Destruction
Burhan - the Proofs of life
Ijaz - the undeniable prophecy

So how about:


1) Ayat - The signs of life
2) Wali - The untried youth
3) Mihna - The season for testing
4) Hajra - The journey of seeking
5) Tahaddi - The challenge of combat
6) Ijaz - The undeniable prophecy
7) Alam Al-Mithal - The mystical world
 
Ahriman said:
Polar desert waste is not attractive enough, given that you can't build improvements on it.
Click to expand...

It is coast, like regular desert waste. Are there improvements you can build on regular desert waste which you cannot build on polar desert waste?

I'm guessing that the religions are just generic atm, but we will need to seriously tone the spread rate down.
Click to expand...

No need to guess. That is what it says in the release note.

Something is strange with fresh water irrigation; I have a cottage getting the irrigation bonus despite not being next to a well (or being Paradise civic). See screenshot.
Click to expand...

I am not sure how the "irrigated" text shows up. Possibly it is automatic if the plot is adjacent to a city.
 
Are there improvements you can build on regular desert waste which you cannot build on polar desert waste?
Click to expand...

No, but you were comparing polar desert waste to rock, which you *can* build improvements on.

Polar desert waste should be better than rock. If polar desert waste is only 1w1h1c, then it isn't.
Rock with a turbine or solar farm or cottage is just as good.

That is what it says in the release note.
Click to expand...
Well, its not really playable. The entire world adopts Shai-Hulad, and then the game is a big happy-fest where no-one attacks anyone.

I am not sure how the "irrigated" text shows up. Possibly it is automatic if the plot is adjacent to a city.
Click to expand...
The issue is not the irrigated text, the issue is the irrigation bonus. The cottage tile is getting the +1 hammer bonus from fresh water access despite not being next to a well.

This is strange, because other tiles next to the city are not getting the bonus, so its not that the city is giving fresh water to every plot around it.
Is it possible that the invisible fresh water resources from the fresh water lakes (that you filled in with rock) are still being placed?
 
You must log in or register to reply here.

Similar threads

Blake00
Dune: Spice Wars
Replies
0
Views
314
Blake00
Blake00
amateurgamer88
Dynasties of Dune Dev Diaries
Replies
16
Views
2K
Blake00
Blake00
Blake00
Heroes of Might and Magic 2 Scenario Mod (& HoMM1 Addon) - Release Thread
Replies
6
Views
2K
CurtSibling
CurtSibling
PDX Firefly
Developer Diary | After the Demo
Replies
5
Views
965
Sephi
S
Blake00
A finished Dune Scenario
Replies
14
Views
3K
Kyriakos
Kyriakos
Top Bottom