incremental patch discussion

[Slvynn]

I don't really understand what you're saying here.

If in the first quote are you saying that if the human player gets a good start position they are unstoppable? That's always been kinda true.

In the second, are you saying that you want the AI to build more units? I don't think thats a great idea; by definition if the AI builds more units then it builds less economy. I don't think we want to weaken the AI economy, or it becomes very weak from the midgame onwards.

In the games where I can steamroll the AI its because I can rush for bladesmen and then hardened bladesmen before the AI can really stop me. Late game once they have good defenses combat can be much slower.
Maybe bladesmen and hardened bladesmen should cost more hammers, or the techs cost more beakers, or infantry units should gain a bonus vs melee?
Maybe we should chance bladesmen so rather than strength 6 +10% city attack they are strength 4 +60% city attack, or strength 5 +30% city attack? So its easier to stop them with quads.

I think the biggest improvement is likely to come from getting the AI to place its cities better. Its current city placement means they grow slowly and just add upkeep. The AI needs to learn to to spread its cities more, so that they aren't trying to share the same water resources.

I would say though that for a given difficulty level, on average this mod is not as hard as vanilla but is harder than most mods, including FFH and the like.

Yes i think you have a valid point. In DW its very important to place city right. Mesas have very hight importance because of windtraps. Also Spice, which should be just harvested. If city placed wrong it wont grow hight,
So far , i saw that AI does much much better on Arrakis map, and worse on Dunipelago (with medium desert level). Previous game Was quite challenging at start actually Mahdi Fighters helped me alot, because i was rexing Harkonnen with bad tech rate, so achieving mahdi fighters at moment when my max str units were 3 just my getting religion spread in my city was awesome , i played Feyd and Agg Chm combo. Spammed them, enormous amounts and gone fighting.

Now , in puprose of test , i started new , with Corrino Agg / Imp , Immortal level and started near polar, with 1 polar ice in capital bfc. The thing that in DW ai will develop his economy first and never dow early with some good SoD, they do it later, and till then Human player have alot of time to get ready his infrastructure and defence as well. There are no guys like monty/shaka. Perhaps Beast Rabban can be one of such Psycho-Shakas.

May be we can add option, for increasing challenge/difficulty puproses. Lets say some bonus through Offworld trading, that can be given to AIs, which will depend on difficulty level. Or just free units.

I playing on Immortal/Aggressive AI on. Perhaps i doing wrong? Is it possible that Aggressive AI just lower AIs development rate, and thus postpone acctually their attacks?

Started another immortal game with Agressive AI off. Rolled Atriedes. Rushed both Ecaz and Ordos and killed them. Now attached screenshots. You can see that i didnt got desert trade and already almost killed 2nd ai, that will be 2 ais dead in couple of turns.
Check Roma attani capital screenshot. That screenshot means that she not prioritizing Windtraps. That will be major fix - if Ai will prioritize terrain, resources and techs which will benefit him most. Thats a major issue, that can and should be fixed. Ignoring Windrap tech stall Ais growth.
Another issue that Ai (Ecaz) not defending well, and i killed him just with 2 Soldiers, with barracks prebuilt - and using Aggresive and barracks xp - got AntiGuardsman promotion.
AI should be less vulnerable to such attacks too. He sends all his soldiers scouting and wondering while leave his land unprotected.

 

[Slvynn]

another post about issue with resources on arrakis map - (Insect Farm resources lacking on Arrakis script) I remeber such issue with FFH once - some of maps were not recieving reagents there. Thats because of resource placements. There are rules, and script put resources one after other. With requirements of distances between them, given by script itself. If those requirements are not met, script stopping placing resources. So is something in bottom of queue and limit achieved - those resources will not be placed at all.

 

[keldath]

hey guys, i been reading your ai reports and its low aggression,

i have some thoughts on it:

the ai aggression, and startegy, is mostly controlled by its leader properties, you can set the percentage it will devote to unit /building/wonder construction, and the levels of war declaration.
perhaps the current settings are just to low for the ai to be aggressive?
if you wish i can take a look at it, im sure that with the proper settings, the ai can become a problem to attack .

another thought i had, do you guys remember the all land maps koma13 created for one of the early versions?
one good thing about those map where - that the ai was aggressive and did some good attacks - this is due to the defenition of the map as all land, without see - so it didn't needed to use ships, meaning - it knew how to manage troops across the map well. we all now the transport system is screwed, the 2 downsizes of those maps - where that - we couldnt make an all terrain transport - meaning a land transport, and the sand sea, was the same height as the land....

perhaps with the knew codes for transports that you guys inserted, maybe the ai will manage land transport better,
maybe we can try those maps all over again, with some modification.
one thing im sure at, thats the ai will be much more active and fun, and most of all , when we tell it to be aggressive, it will be aggressive


edit:
just an update on my plans for next patch:

- update the bbai
-update car mod 0.2>0.3
- hopefully update revdcm to 2.6

maybe get some useful sdk code inside - i will consult with you before, off course.


edit2:

just had another thought,

we all like the idea of buying units from the off-world screen, youll probably wont like this, but its just a thought i had,

what the difference between ordered units to regular units - costs - the first costs money and you get it instantly, and the second, you get after the build time,
how about - that ordered units will have a life spand - meaning after a while those units, will get obsolete - they will die/perish/gone,
this will create some balance and you wouldn't be able to base your entire army on ordered units, say you wanna commence an attack - you order a bunch of units , you enhance your army, attack and puff they are gone after a few turns, you have limited time to use the units you ordered right?

any thoughts?

 

[Slvynn]

I dont think it is related to transports. As fasr as i sa it - AI handles transportation well, the simple he isnt enough aggressive even on big continents. Actually distant ai can become thread, because HE IS coming later with huge stack of tranports and attack your core, undefended cities. AI dont check your Strenght rating , and wont attack if you have it too low even with minor force. Why need to build supperior force if you alread yhave x2 than your enemy? I miss Shaka and some shakish behaviour, at least from aggressive leaders.

 

[keldath]

Slvynn,
then im sure the aggression levels are due to the current leader settings,

if everyone agrees, i can modifies the leaders to be more aggressive and to handle war better,
just for testing purposes, maybe creating a second set of leaders - like paul atreidis and paul atreidis (aggressive).
in my older mods, ive made some really aggressive leaders - had complaints to lower their power .

just give me an ok guys.

 

[Slvynn]

Slvynn,
then im sure the aggression levels are due to the current leader settings,

if everyone agrees, i can modifies the leaders to be more aggressive and to handle war better,
just for testing purposes, maybe creating a second set of leaders - like paul atreidis and paul atreidis (aggressive).
in my older mods, ive made some really aggressive leaders - had complaints to lower their power .

just give me an ok guys.

Just modify that file that will be great, post it ,and i will start testing immidiately
And then we'll see results.
(I am sure that will fix part of a problem, esp for hight levels).

 

[keldath]

ok , done, starting to work now.


edit:

wow, now that im looking at the ai settings.....i can most certainly know why its a peaceful ai....

all of the war parameters for the ai are the same for most leaders - and set to low probability it wil declare war.

i will edit the leaders to be extra aggressive for starter, after you guys test it a bit, we can refine it more.


edit 2:

ok, heres the new leaderhead file with aggressive parameters,
theres more work to do here, but, basically, i made that the leaders would be more likely to declare war on players,
in the original file in 1.5 - its a wonder if the ai declares war at all.......
i tried to give the more aggressive leaders - the baron, raban, to be more aggressive and to match the character traits in the book,
but of course more tweaks are needed.

i would love to get feedback to know if its challenging enough now.

put the file in - dune wars/assetes/xml/civilizations

happy testing.

 

[Ahriman]

The thing that in DW ai will develop his economy first and never dow early with some good SoD

Mostly the distances between factions and AI and the lack of transports mean that prosecuting an early game war will be very hard, particularly for the AI. So I think that its *good* that the AI doesn't waste its time trying to wage super-early wars.
I guess you could have 1 very aggressive early AI player, but there is really no point at all in being aggressive until you have bladesmen. Nothing else can take cities from infantry if they have a decent number.

If you're playing as Atreides, you're benefiting from the IMO ridiculous free unit spawn mechanic, which probably makes an early rush easier. Which again is out of theme, and so is another reason to remove this mechanic.

Re: the Ordos shot: I'm guessing the issue is that she didn't even have the water conservation tech. The AI otherwise builds windtraps because they have a large yield.
It is an absolute no-brainer every game to get water conservation as your first tech, you desperately need windtraps/dew collectors and then wells to catapult your growth; maybe we need to increase the AI weight on this tech.

Another issue that Ai (Ecaz) not defending well, and i killed him just with 2 Soldiers, with barracks prebuilt - and using Aggresive and barracks xp - got AntiGuardsman promotion.

Really? You must be playing on very small maps, and starting very close to the AI.

AI should be less vulnerable to such attacks too. He sends all his soldiers scouting and wondering while leave his land unprotected.

Interesting, I hadn't noticed it (not saying it isn't happening, only that I hadn't noticed it). Any thoughts on some way to change this AI behavior?

Maybe increasing soldiers to +50% city defense would be enough?

Thats because of resource placements. There are rules, and script put resources one after other. With requirements of distances between them, given by script itself. If those requirements are not met, script stopping placing resources. So is something in bottom of queue and limit achieved - those resources will not be placed at all.

I think it is highly likely that this is the issue.
Barrel cactus hardly ever shows up either.

how about - that ordered units will have a life spand - meaning after a while those units, will get obsolete - they will die/perish/gone,

I don't like this.
The balance factor with purchased homeworld factors is the fact that there are only ever a handful of units you can buy, and your pool is slow to replenish. So there are never enough homeworld units for you to be able to pump up a big army of them. You can buy a few in an emergency, but thats it.

Making them temporary would just be terrible, and the AI could easily purchase and waste them.

It also wouldn't make any logical sense; why should troops disband just because they were elite guys who came in from your home planet?

I do think we need to tweak the mix and availablility of homeworld units, and ideally make the pool faction specific, but thats for another topic.

i will edit the leaders to be extra aggressive

Some variation would be good, particularly within a faction. Eg make Rabban aggressive, but Vladimir not so much (he's more of a schemer). Make Shaddam aggressive but not Irulan. Make Paul aggressive but not Leto. Make Alia aggressive but not Stilgar (or vice versa). etc.
BG and BTl and Ix should never really be particularly aggressive.

 

[keldath]

I don't like this.
The balance factor with purchased homeworld factors is the fact that there are only ever a handful of units you can buy, and your pool is slow to replenish. So there are never enough homeworld units for you to be able to pump up a big army of them. You can buy a few in an emergency, but thats it.

Making them temporary would just be terrible, and the AI could easily purchase and waste them.

It also wouldn't make any logical sense; why should troops disband just because they were elite guys who came in from your home planet?

I do think we need to tweak the mix and availablility of homeworld units, and ideally make the pool faction specific, but thats for another topic.

ok, just a random thought i had.
i havent tried the game yet to be frank, so im not sure how everything ticks yet.

Some variation would be good, particularly within a faction. Eg make Rabban aggressive, but Vladimir not so much (he's more of a schemer). Make Shaddam aggressive but not Irulan. Make Paul aggressive but not Leto. Make Alia aggressive but not Stilgar (or vice versa). etc.
BG and BTl and Ix should never really be particularly aggressive.

yap, exaclty what i was thinking, due to the fact that we have low count of factions, each factions need to have at least 2 eladers with different strategies - some more aggressive some less, some will attack more factions without shared border some will prefer to attack shared border, some will attack players with higher ranking some with low, a varied personalities for each will create new tactics for the ai with each game.

for start i lowered all the war attitude to all leaders - first i wanna have some feedback on how the leaders are working with my new settings, then get some more emphasis on adjusting the values for each leader according to its character personality just as you suggested.

so plz guys, help with the tetting.

 

[Slvynn]

Ix Leader trait is Aggressive, so why not ?

Re-Ordos screenshot - SURE!!!! Windtraps should be prioritized always. Thats necessary for development. It should be priority 1 tech. Otherwise AI just not grow.

 

[davidlallen]

Re: the Ordos shot: I'm guessing the issue is that she didn't even have the water conservation tech. The AI otherwise builds windtraps because they have a large yield.
It is an absolute no-brainer every game to get water conservation as your first tech, you desperately need windtraps/dew collectors and then wells to catapult your growth; maybe we need to increase the AI weight on this tech.

If it is an absolute no-brainer to get one tech, doesn't that mean the early techs need to be re-arranged? There should be some decision making involved. How can we make small changes to the first couple of levels so that more brain is required?

 

[Ahriman]

Well, there are still interesting things to do once you get water conservation. I guess a strategic way of adding a choice would be to make something else a requirement for water transportation (mining maybe, you have to dig the wells?), so that you either go straight for dew collectors/windtraps but then have a while to go before wells, *or* you beeline for wells and get wells earlier.

But in vanilla getting agrictulure for farms almost immediately is normally wise, and that isn't a problem, but the real no-brainer in Dune wars comes from the fact that tile yields are so incredibly low without wells or dew collectors or windtraps.

 

[davidlallen]

the real no-brainer in Dune wars comes from the fact that tile yields are so incredibly low without wells or dew collectors or windtraps.

Perhaps all three of dew collector, windtrap, well should be different tier 1 techs or at least not on the same path. For example, dew collector could move to desert survival, and well could move to mining.

 

[Ahriman]

Perhaps all three of dew collector, windtrap, well should be different tier 1 techs or at least not on the same path. For example, dew collector could move to desert survival, and well could move to mining.

Interesting.

Moving the dew collector to desert survival is a very good idea I think; that tech is currently very underpowered. Moving the well to mining... not so much I think. That would just make mining a no-brainer as the first choice; you nearly always start near 1-2 groundwaters, and a well has a higher yield than a windtrap, and a tech that gave both mines *and* well would be very powerful.

We could however make mining an AND requirement for getting water transportation; so it takes both mining and water conservation.

But even so, we will still need to change the tech weightings somehow to give priority to the techs that provide water generating improvements.

 

[Ahriman]

Thinking about it, shoudlnt' it be easier to dig a well than to build a windtrap?

Another alternative:
Move well construction to water conservation
Move dew collector to desert survival.
Rename water transportation to something else, maybe "Intensive water collection", or "windtraps", or "water preservation" or something, and move windtraps to that tech.

Another thing we would have to tweak though; the starting techs/units for each leader.

 

[davidlallen]

Moving the dew collector to desert survival is a very good idea ... We could however make mining an AND requirement for getting water transportation; so it takes both mining and water conservation.

Done, and done.

But even so, we will still need to change the tech weightings somehow to give priority to the techs that provide water generating improvements.

The tech flavors and weights are hopelessly scrambled since the tech tree redesign. In retrospect, we should have had more columns in the design spreadsheet which included the weights, and then updated them at the same time.

If you are interested, take a look into technologies/civ4techinfos.xml and search for <Flavors>. You will find that there are about six named flavors, such as religion, gold, culture, growth, science, production, military. Each tech has a few flavors with a value between 1-10 assigned. As you will see with a quick survey, there is absolutely no logic today about how the flavors go to the techs. This is because each tech started out in vanilla as a different tech; while the names and effects were totally changed, the flavors were left the same.

 

[davidlallen]

Lots of good discussion here related to 1.5 feedback, rather than discussion of the next patch or release plans. Back on topic.

Last night I renamed most of the unit and building directories and rebuilt the fpk file with all the voluminous new art since 1.5. This has no user-visible effect. A release containing the fpk file will be large, and subsequent patches over that will be small.

Today I have been cranking out a large number of small changes from the various threads, so I have about 30 small changes. I want to dig into the worm rider graphics problem pointed by Ahriman a little.

But, the next release will be 1.6, because it contains a new fpk file. This is a little unusual because an X.Y release implies more, bigger changes than an X.Y.Z release. This will be relatively small in gameplay, but large in file size.

Does it seem OK? Does anybody else have changes planned for the next few days? More new building art, etc?

 

[Ahriman]

I know the flavors are messed up, and I will look at redesigning those at some point (and the leaderhead flavor weightings), but are there separate AI weights other than the flavors? Or just the flavors?
Does the AI just look at flavors when choosing techs, or does it also try to evaluate how useful a tech is?

Can we also look into increasing settler cost? I know its hard but I think that would have a very large gameplay improvement.

 

[Deliverator]

Can we also look into increasing settler cost? I know its hard but I think that would have a very large gameplay improvement.

It not actually hard. It's just that it seems that you can't change it without affecting city growth time as well. This might not be a bad thing - but it needs testing. I'm going to have a play with it.

 

[koma13]

Does anybody else have changes planned for the next few days? More new building art, etc?

I want to finish the next homeworld update. And having a new fpk is my chance to clean up the homeworld art directory.