1.9.5 Feedback

Joined
Jun 5, 2010
Messages
394
Location
Houma, LA
>>>Download Dune Wars Patch 1.9.5 Here<<<

Like all the previous patches, this can be installed onto a clean 1.9.1 install(download here), or any patched version after it.

While there are some minor balance fixes here, the main aspect of this new patch is the AI work that has resulted in a much more economically competent AI, so you can expect it to be more challenging to build a huge tech lead, especially at higher difficulties. At the same time, the combined effort with the AI code and handicap rebalancing should reduce unit spam issues while still keeping the AI rather dangerous. We've also solved the issues with AI transports dropping their cargo when invading which by itself makes them much more competent at war. I hope everyone enjoys the work we've accomplished here, and I look forward to hearing feedback from players, even those who normally don't post much :).

There's also a lot more going on between ongoing balancing to make this great mod even better, improvements to air combat that I hope everyone will enjoy, a general reworking of the Ixian Mech units that keeps much of what was fun about them before and makes them even better, and several refinements in religion spread and founding, just to name a few areas.

Patch Notes
  • Considerable AI overhaul:
    • AIs much less likely to tank economies early on
    • More competitive in the tech race
    • Worker AI improvements, especially concerning spice production
    • Transport danger/dropoff AI fix
    • Numerous other smaller fixes/tweaks
  • Handicaps also overhauled:
    • Less 'artificial' AI bonuses, more focus on a more natural difficulty curve throughout the game.
    • considerably less 'unit spam' at high difficulty levels and in the late game
      • (this doesn't mean to expect them to be pushovers though ;))
  • Siege Weapons further rebalanced (base combat limits and collateral damage reduced)
    • more likely to survive but do less maximum damage
  • Siege Promotions also tweaked to encourage more specialization
    • Drill encourages survivability at the cost of collateral damage
    • Barrage needed for heavier collateral damage, but sacrifices survivabilty
  • Overhaul of Air Combat
    • All air combat (air-air and air-ground) revamped to account for promotions/terrain/etc. in a way more in line with what a player would expect
    • This includes including anti-hornet promotions (AA Rockets) and defensive ground/fortification bonuses into the combat model
    • Air units receive fractional xp for successful bombing and airstrike missions based on interception risk
    • Air units intercepted by ground units can inflict small damage to the intercepting unit and both receive xp based around damage dealt to the other unit in the exchange
  • Religion fixes:
    • Bug preventing free missionaries from spawning sometimes when founding religions fixed
    • Imperial no longer removed from holy city if another religion is spread to it
    • Small refinements in religion wildfire spreading (on founding certain religions) and holy city placement
    • Tleilaxu automatically convert to their religion when it is founded and can't switch to no religion
  • Ordos TechByConquest Ability
    • Upon capturing an enemy city, they can steal tech points in any researchable tech(s) the other team has based on how much more advanced the other team is
  • Mech Overhaul:
    • Less restrictive upgrade path, more encouragement for using a diversity of mech types at the same time
    • Late game promotions that can boost the strength of all mechs with it by +2:strength:
    • Unit number limits for all mech classes
    • See This Post for more information
  • Misc. Fixes/Tweaks:
    • Various text fixes
    • AIs now value Sapho Juice(if they can build mentats) and ginaz training in trades
    • Political gives 2 free culture instead of 3
    • Golden Ages slightly longer
    • Ducal Guard and Order of Agamemnon require the unit to have fought 1 or 2 battles respectively to give them to a unit (no fresh recruits)
    • 'Fanaticism' Tech can be reached from 'Faith' or 'Defense Tactics'
    • Pilgrimage Sites moderately more expensive (170:hammers:->210:hammers:)
    • Fixed bug allowing Fremen to build all aircraft, now restricted to just Wasp Interceptor as intended
    • Siege Units display XP from winning battles correctly in the Combat Odds
    • Increased Civic costs (Organized now much more useful)
    • Unit costs increase the higher a player goes over their free unit cap
 
Excellent! I look forward to trying it.
I would add "increased civic maintenance costs" and "increased unit upkeep costs", these are significant. Also add: change to
I had another idea for Political trait: since we have reduced the culture yield, perhaps we could mimic Religious, and give a free great noble if you found Imperium and a free great merchant if you found CHOAM?
I also wonder if we should tweak the tech tree to add an extra requirement to Great Houses, so that it is not so easy to found both Imperium and CHOAM in a single beeline.

'Fanaticism' Tech can be reached from 'Faith' or 'Defense Tactics'
Good idea; this should make early AI rushes with Bladesmen more feasible.

Golden Ages slightly longer
Why?
Golden ages are already more powerful than they are in vanilla, because nearly every tile produces hammers and most produce gold. So the economic boost of a golden age is huge.
 
Wow, just saw a nasty early bladesman rush by Ordos. Very tough to fight. This really changes the game it forces you to get a combat unit (thopter, quad, bladesmen) early on - at least on Epic. Nice!
 
Golden Ages slightly longer
Why?
Golden ages are already more powerful than they are in vanilla, because nearly every tile produces hammers and most produce gold. So the economic boost of a golden age is huge.

The change log is everything from 1.9.4 to 1.9.5, so a number of these are stuff introduced in the 1.9.4 beta patches. This is the same small increase from one of those that we already discussed.

'Fanaticism' Tech can be reached from 'Faith' or 'Defense Tactics'
Good idea; this should make early AI rushes with Bladesmen more feasible.

Wow, just saw a nasty early bladesman rush by Ordos. Very tough to fight. This really changes the game it forces you to get a combat unit (thopter, quad, bladesmen) early on - at least on Epic. Nice!

Yep, again carried over from the beta patches, I've been very pleased with the results of this change. I had to quit one Emperor game when I was at the sore end of a bladesman rush from two opponents at the same time from different directions. And loved it :lol:.

I had another idea for Political trait: since we have reduced the culture yield, perhaps we could mimic Religious, and give a free great noble if you found Imperium and a free great merchant if you found CHOAM?

That's a great idea! I'll be sure to work it in.

I also wonder if we should tweak the tech tree to add an extra requirement to Great Houses, so that it is not so easy to found both Imperium and CHOAM in a single beeline.

This is something I've been thinking about too. I'll have to take a good look at the tech tree and see where this can be improved. One certainly shouldn't lead directly into the other so easily.

I would add "increased civic maintenance costs" and "increased unit upkeep costs"

done, and done.
 
the main aspect of this new patch is the AI work that has resulted in a much more economically competent AI, so you can expect it to be more challenging to build a huge tech lead, especially at higher difficulties.

I found the opposite to be true in my first few Deity games with the new patch. In 1.9.4, it took a particularly fast start or lots of spice to get a tech lead and even then most religions and wonders took a beeline and some luck to get. Now a routine start is enough to get the edge in tech by turn 100-150 (normal speed) and it's possible to get as much land as the AI, found the mighty CHOAM and build a few wonders while doing so.

On a more positive note, the decreased tech cost makes it possible for any civ to build a worker, research Water Conservation and finish the tech just in time to build a wind trap without any wasted worker turns. This helps a lot to bring those civs that don't start with the tech up to par with those that do. Further, the faster tech rate makes skipping Mysticism a less attractive move: you could probably drop the political office without affecting the game much.

Barbarians also seemed weaker with fewer waterstealers early. I haven't played enough games to tell if it's just a quirk of the games that I did play, but perhaps the AI changes have lead it to fogbust more. Fewer barbs means an easier game, of course, but given how many people play on Epic where the barbs are worse, perhaps it's for the best.
 
I found the opposite to be true in my first few Deity games with the new patch. In 1.9.4, it took a particularly fast start or lots of spice to get a tech lead and even then most religions and wonders took a beeline and some luck to get. Now a routine start is enough to get the edge in tech by turn 100-150 (normal speed) and it's possible to get as much land as the AI, found the mighty CHOAM and build a few wonders while doing so.
Interesting! I wonder if this is because on Deity the AI got such significant bonuses to maintenance, that its old strategy of more reckless expansion was profitable at Deity?
Also, did we cut down some of the bonuses at Deity?

Further, the faster tech rate makes skipping Mysticism a less attractive move: you could probably drop the political office without affecting the game much.
I'm certainly open to considering this. The other early game techs give you something very important, the political office makes mysticism feel a bit weak.

Barbarians also seemed weaker with fewer waterstealers early. I haven't played enough games to tell if it's just a quirk of the games that I did play, but perhaps the AI changes have lead it to fogbust more. Fewer barbs means an easier game, of course, but given how many people play on Epic where the barbs are worse, perhaps it's for the best.
Also worth considering. I only play Epic, and the barbs there feel about right. What experience are others having on Normal speed?
 
I found the opposite to be true in my first few Deity games with the new patch. In 1.9.4, it took a particularly fast start or lots of spice to get a tech lead and even then most religions and wonders took a beeline and some luck to get. Now a routine start is enough to get the edge in tech by turn 100-150 (normal speed) and it's possible to get as much land as the AI, found the mighty CHOAM and build a few wonders while doing so.

One thing I aimed for with setting up the handicap difficulties was try to set up a little smaller steps than before, and I probably took out too many AI bonuses at Immortal and Diety, and I haven't play tested above Emperor (which makes this very important feedback for balancing this for Immortal/Diety players). At the same time, the threat of early economic tanking (which was the primary thing the new AI work is designed to deal with) is much lower at these very high difficulty levels. In other words, the new AI code has the most effect on AI performance where the AI has fewer bonuses, less effect at Diety.

I'm attaching a zip file with some changes to Immortal and Diety that brings them somewhat closer to their original bonuses, try it and let me know how it goes :)

Interesting! I wonder if this is because on Deity the AI got such significant bonuses to maintenance, that its old strategy of more reckless expansion was profitable at Deity?
Also, did we cut down some of the bonuses at Deity?

Right, precisely what I said above in a different fashion, it's bonuses made the economic downturn from over expansion too early seen in lower difficulties much less significant or never happen at all at Diety, so it's more or less unaffected by the changes that by contrast can have a large impact at Monarch for instance.

With that in mind, the attached XML moves the bonuses closer to before, but with targeted adjustments for less late game unit spam still.

I'm certainly open to considering this. The other early game techs give you something very important, the political office makes mysticism feel a bit weak.

It depends a lot on your strategy. If you only need the first ring of expansion, then you can get by with political offices-that's what they were designed for. If you really want to pursue cultural expansion (to get spice, compete with nearby enemy cities, pick up resources outside of the first ring) you'll still need something more potent.

A thought I did have, you'd suggested a while back Ahriman to give them +1 espionage, I wouldn't want to do that in general, but it would be a nice perk for organized leaders maybe?

Barbarians also seemed weaker with fewer waterstealers early. I haven't played enough games to tell if it's just a quirk of the games that I did play, but perhaps the AI changes have lead it to fogbust more. Fewer barbs means an easier game, of course, but given how many people play on Epic where the barbs are worse, perhaps it's for the best.

I haven't changed anything in barbarian spawn rates (I just double checked). It's of course largely dependent on the nature of your starting spot, what map size you use, land percent, etc. because if you have a lot of open land around you at the start... you're gonna see a lot of barbs. If you're starts have been in tight spaces with little room for them to spawn near you, that's the most likely cause.
 
+1 espionage, I wouldn't want to do that in general, but it would be a nice perk for organized leaders maybe?
Sounds reasonable to me.

I haven't changed anything in barbarian spawn rates (I just double checked).
But it seems possible that cutting down on the number of scout thopters the AI builds means that they do more exploration with military units (which are only on land) and so more fog-busting occurs.
 
But it seems possible that cutting down on the number of scout thopters the AI builds means that they do more exploration with military units (which are only on land) and so more fog-busting occurs.

To be specific, the AI looks at UnitAI_Explore and UNITAI_ExploreSea as separate things. In vanilla civ, AI_Explore is exemplified by the scout unit. As a general rule the only AI_Explore units you will ever see in Dune Wars is the initial soldiers given to the AI at the start, I don't think they are building any after that, which makes perfect sense since ExploreSea units can explore land and sea unlike in vanilla civ. So for all intents and purposes, other than the initial soldier(s) that will wander around until killed (there's no place where they are switched to another unitAI), the only "exploring" units built are any ExploreSea units (scout thopters at first, but these will be the best thopter unit available at the time).

The AI also has no specific coding to 'fog bust'. What they do have is coding to start building an attack stack once they are running out of good spots to found cities with which they will try to take Barbarian Cities if any are around, but that's the extent of it. I'm sure you've all seen them doing so and this works well, but only kicks in if there aren't more good city sites.
 
The new tweaks noticeably improve the performance of the Deity AI, but still leave it weaker than it was. I think that the elimination of the human's 30% research handicap is the culprit. The AI does do a better job of improving spice early, but not enough to offset the faster human research rate. A player beelining the key economic techs can quickly take the lead in population, hammers and research.

I haven't seen any of the Bladesman rushes that are in others' games either. That might be a speed issue (doesn't the AI roll the dice once each turn to decide whether it will prepare for war?), a product of just not playing enough games or the result of trying to keep neighbors at pleased. It's usually easy to keep them happy just by not adopting a different state religion and then piling up the smaller bonuses. Since most religious benefits come from founding or having the religion and don't need it to be the state religion, this strategy isn't very costly.

Interestingly, I saw a Deity AI crash its economy in my last game. Around turn 80 (still normal speed), Scytale had 66 turns to go to finish researching Desert Plantation. His research rate tripled in the next few turns and eventually recovered fully, but he remained the lowest in score by a wide margin for next hundred turns (I haven't played past that, but expect the trend to continue), so his over-expansion clearly did significant damage.
 
The new tweaks noticeably improve the performance of the Deity AI, but still leave it weaker than it was. I think that the elimination of the human's 30% research handicap is the culprit. The AI does do a better job of improving spice early, but not enough to offset the faster human research rate. A player beelining the key economic techs can quickly take the lead in population, hammers and research.

This is a common misconception of the <iResearchPercent> tag in the handicap infos, it actually does not only affect the human, it affects all players the same by increasing the cost of each tech by this percentage. That's why I removed it, it has no real bearing on AI difficulty and only serves to create odd differences in the tech pace (that don't need to be there) between lower and higher difficulty levels.

Interestingly, I saw a Deity AI crash its economy in my last game. Around turn 80 (still normal speed), Scytale had 66 turns to go to finish researching Desert Plantation. His research rate tripled in the next few turns and eventually recovered fully, but he remained the lowest in score by a wide margin for next hundred turns (I haven't played past that, but expect the trend to continue), so his over-expansion clearly did significant damage.

The sudden drop then recovery of the research rate usually means something occured where the AIs 'target gold', how much gold it wants to have in it's treasury, suddenly went up (possibly due to wars, upgrades available for units, etc.), so it drops it's research temporarily to reach that gold target, then shifts back into it. The tleilaxu are actually usually one of the better 'tech'-ers (I know, not really a word, just keep with me here :crazyeye:) in my no-tech-trading games, but they often suffer with tech trading on. Aside from that, there are always winners and losers among the AI civs just based on starting conditions, did they lose a few settlers early, etc., etc.

Another cause of AIs that never really 'take off' is early wars that stalemate, with both sides pouring resources into a non-winnable war and falling behind as a result. Hopefully the changes bringing in offensive units earlier and preventing the silly 'infantry-rush' tactics they were trying before should help this. I should probably look into encouraging peace if neither side is getting anywhere, but ultimately things like this are unavoidable at a certain level.

If you want to help out with getting a little insight into the AI's economic growth/decisions, in your BTS ini file, put in the 'chipotle' cheat code (this won't affect your ability to play normally), I've added some code that gives you some economic information on an AI if you hold 'ctrl' while hovering over their name in the score board. The most important things it shows is the gold and research output and a number called 'funded percent', which is a good indication of the economic health of a player. A good number is basically anywhere above 70, the higher the better, once it gets around 50 or lower (this indicates a high costs to commerce ratio) it's a bad thing. This is usually handled better now with my changes, but the total size of the economy (net commerce) still often lags behind a good human player.

It'll also show information about how many units the AI has and how many free units it can support, the total unit costs, and the percentage of netCommerce those unit costs represent.

The real thing I'm trying to figure out right now is what is the biggest differences between what a good player does and what the AI does that allows a player to build up a larger economy for the same land area consistently compared to AIs, even with considerable AI handicap bonuses. I've largely handled over-expansion issues, now the focus of my observations is going into research choices, build choices, and maybe improvements as well.
 
It's usually easy to keep them happy just by not adopting a different state religion and then piling up the smaller bonuses.

I'm strongly considering making having no state religion equivalent to a different state religion, though maybe with a little less diplomatic backlash. Obviously it wouldn't give you diplo bonuses to other non-state religion players though.
 
I'm strongly considering making having no state religion equivalent to a different state religion, though maybe with a little less diplomatic backlash. Obviously it wouldn't give you diplo bonuses to other non-state religion players though.
I'd lean towards supporting something like this. Testing required, of course, but I've always found staying deliberately no-state religion to feel exploitative and gamey. I think it is good to encourage the player to get off the fence and pick a side.
 
Another thing I've been thinking of; maybe due to the wildfire spread of Mahdi or other changes in the teching (AIs are not prioritizing economy techs more over Feudalism for religion founding so Imperial is getting founded ), I haven't noticed Imperial spreading as much, and I'm a bit worried that it will be underpowered. One possible adjustment tweak; we could bring the Imperial temple building (Feudal estate?) which is required for the Imperial missionary unit (which is national limit 1) back to Feudalism tech from its current position at Imperialism, which is really a solid midgame tech.
We could also check if the Imperial missionary unit has the Suspensor travel promotion or not, and consider if it should be added in the latter case.

Of course, adding another tech pre-req for Great Houses might also delay CHOAM and so leave more time for Imperial spread.
Or we could move CHOAM founding to Offworld Trade or something.
Or we could make Great Houses require Spice Industry.

Though I worry that these changes might make Offworld Trade too powerful and Great Houses and Imperialism underpowered.

Also, is the water souk too expensive in hammer terms? It always feels a bit odd the way it has much higher hammer cost than its technological contemporaries.
 
We could also check if the Imperial missionary unit has the Suspensor travel promotion or not, and consider if it should be added in the latter case.

It doesn't, and I'd support adding it. Right now it's just useless micromanagement to use them that I doubt players appreciate.

My thoughts on the CHOAM question are a little different (for better or worse, I don't know). I'm thinking along the lines of placing Great Houses right next to, instead of after, Feudalism, so that picking one or the other will probably be necessary, as an AI will probably pick up the other by the time you've got one (if you get one that is). Imperial would be the better diplomatic choice, while CHOAM would be the better economic, but I like the idea of them being early competitors of each other. Because of CHOAM's slow spread rate, it's of course rare to see it become prevalent outside of the founder's borders, so I don't think it's a problem for Imperial spread.

Right now the issue as I see it is every logically viable prereq for Great Houses (if we take CHOAM from it and move it to another tech, Great Houses virtually becomes an empty tech, we'd do better to ditch it in that case) passes through Feudalism, so the first to it still has a leg up in trying to found both.

One possible adjustment tweak; we could bring the Imperial temple building (Feudal estate?) which is required for the Imperial missionary unit (which is national limit 1) back to Feudalism tech from its current position at Imperialism

The feudal estate occurs at caste system, which requires great houses. I agree it seems there is too much distance between founding imperial, and actually being able to build missionaries. If we did move Great Houses next to Feudalism, we could set Caste System right after Feudalism. Protected Trade and the rest of it's line would still extend off of Great Houses.

I'm really not set on anything yet in terms of any of these tech changes proposed, but I did want to say where I've been leaning for the discussion. It's the kind of thing where it's all too easy to create new problems fixing old ones, so we have to be careful making any big changes. It's an important part of the tech tree and most of it works quite well.

maybe due to the wildfire spread of Mahdi or other changes in the teching (AIs are not prioritizing economy techs more over Feudalism for religion founding so Imperial is getting founded ), I haven't noticed Imperial spreading as much, and I'm a bit worried that it will be underpowered.

Once the free great nobleman for Political leaders is included, political leaders will have a big leg up in spreading imperial early. Building a Holy City 'shrine' essentially doubles the 'spread factor' for that religion in connected cities. Since Imperial has such a high spread factor, the imperial palace is the most powerful shrine in terms of spreading religious influence in the game. Getting it earlier should have a big impact.

Relevant part of Tech Tree
Spoiler :
 
I'm thinking along the lines of placing Great Houses right next to, instead of after, Feudalism
Hmm. Worth considering. My worry is that bringing CHOAM founding earlier might limit spread, but earlier Imperial spread with missionaries might counteract this. It would be an interesting strategic tradeoff to go for diplomacy/culture/civics or economy/trade.
But I agree that it is very unfortunate to force Feudalism in order to found CHOAM, and currently Feudalism is too much of a bottleneck.

One possibility:
Feudalism requires Fanaticism or Education (as current)..
Feudalism enables Hereditary Rule and founds Imperial, and is required for the Feudal Estate. The Imperial diplomat (missionary) requires the Feudal Estate and has suspensor travel.
Feudalism is required for Law of Arrakis and Caste Systems.
Great Houses requires Light Manufacturing and Dune Topography (basically: industry and trade/navigation techs are required in order to get produce and use trade goods, for a trade religion - and these are currently dead-end techs).
Great Houses is required for Spice Economy and Protected Trade.
Caste Systems is required for Serfdom civic, Faufreluches civic.
CHOAM Emissary (missionary) requires the CHOAM outpost building and the protected trade tech, so the missionary still comes a bit later.

Harsh Conditioning requires Fanaticism AND [Feudalism or Great Houses]

Move wet planet conservatory? Maybe to an ecology tech? Planetary Ecology?
Or leave Serfdom at Imperial but move Hereditary Rule to caste systems?

Building a Holy City 'shrine' essentially doubles the 'spread factor' for that religion in connected cities.
Are you sure?
I thought there were separate spread factor parameters and that these were additive. In vanilla Civ4 every religion has spread factor of 100. In Dune Wars, Imperial religion has spread factor of 150. I don't think the shrine doubles, I though it increased by 50? But I can't see the XML tag for the shrine effect, so I might be wrong.
 
That's pretty close to how my mind has thinking on it, with a few additions that I happen to like. I definitely agree too many techs were bottlenecked through feudalism, and that was part of my logic with moving Great Houses back parallel with it. Only part I'm not sure about in principle is moving the Wet Planet Conservatory, I see it a lot like the wierding room at the Arrakeen Palace (I always assumed that's kind of what it was supposed to be, just on a bigger scale) and is kind of a symbol of Imperial power in a way. I'd leave it alone I think, just first impression.

I'm going to try shifting things around to see how this looks and we'll take it from there.

Are you sure?
I thought there were separate spread factor parameters and that these were additive. In vanilla Civ4 every religion has spread factor of 100. In Dune Wars, Imperial religion has spread factor of 150. I don't think the shrine doubles, I though it increased by 50? But I can't see the XML tag for the shrine effect, so I might be wrong.

You know me, I do my homework:

Code:
if (pHolyCity->isConnectedTo(this))
{
	iSpread = pLoopCity->getReligionInfluence(eLoopReligion);
	iSpread *= GC.getReligionInfo(eLoopReligion).getSpreadFactor();

	if (iSpread > 0)
	{
		iSpread /= std::max(1, (((GC.getDefineINT("RELIGION_SPREAD_DISTANCE_DIVISOR") * plotDistance(getX_INLINE(), getY_INLINE(), pLoopCity->getX_INLINE(), pLoopCity->getY_INLINE())) / GC.getMapINLINE().maxPlotDistance()) - 5));

		iRandThreshold = std::max(iRandThreshold, iSpread);
	}
}

Religion Influence is 1 for a holy city, and all the shines add 1 more. Influence is multiplied by the spread factor. Then it's decayed according the distance of a city from the holy city (as a percentage of the map size). The holy cities are also the only cities with influence, and therefore the only cities that can cause spread of a religion.
 
Just an aside I just realized looking at that code again, so I'll mention it--religion spread is not adjusted for game speed at all...

The 'iRandThreshold' shown above is compared against a random number roll from 0-1000, so a city within 5% of the total map size of the imperial holy city (so no distance decay) without shrine has a 15% chance each turn of converting to Imperial, 30% chance with the shrine. This means there will be a lot more spread before other religions come into play at slower speeds (since their are more turns to spread in), favoring Imperial and Shai'halud early spread, and conversely less chances for spread at faster speeds before other religions are founded.

This is something that should probably be scaled by game speed--e.i. increase the spread % at quick, decrease it on epic/marathon--so that you expect roughly similar observations regardless of game speed
 
Top Bottom