Acken's Minimalistic Balance for singleplayer (and AI improvements)

[Acken]

Well strategic ressources play a role here too, swords are very strong because they require Iron. Horsemen are 14 but with 4 movement and withdrawal and the penalty for city attack is lower than vanilla. While horses have a less powerful attack they are in general easier to save than melee.

Archers still work well in their respective eras as a second line or in defense where they should be. There is a limit due to 1UPT of how many front liner you can field, this is what acts as balance between melee and ranged. I want to also remind you that while archers are weaker they are also cheaper and cover is down to 25%. yung carl even made the claim that he still preffer Xbows to melee units in medieval. I want melee to act as the main meat grinder, range units as extra damage and horses as hit and run units. Unique Units are an exception and things that got hit hard by the changes like the Impi or the CKN will get some love back when I work on them.

For pikemen you have a point that they could go to strength 17 or 18 due to their later position which is a change entertained for the next version. The main reason why pikes are at chivalry is that they are the counter to knights and should come at the same time at the earliest. The second reason is that Civil Service already is a key tech and probably shouldn't give better defense on top of better economy.

Overall I should do another pass on units to keep the ratios consistent though. Probably should make warriors unavailable at IronWorking instead of MetalCasting too.

Here are a few strength ratios:
LS / Knight = 1.15
Sword / Horseman = 1.14
Knight / Pike = 1.25
Horseman / Spear = 1.17
LS / Pike = 1.44
Sword / Spear = 1.33

If I want to keep these ratios consistant it would appear a Pike buff is necessary to have Knight / Pike and LS / Pike be consistent with the previous era. Pike at 17 achieves this with a slight cost reduction (probably -5).

 

[noto2]

Acken, you joined a thread I created about the lacklustre AI in the unmodded game, maybe two months ago, and suggested I try your mod.

I took a 2 month break from Civ5 but I'm back, and just started playing your mod. At the moment I have paired it with another mod that tweaks the civs slightly, because I find some of the civs UA's horribly stupid/boring/pointless to play. The other mod is called "weak Civ AI balance" or something like that.

Anyway, it does certainly seem like you have drastically improved the AI. I have played 2 games and am on a third now and have noticed it's quite common to see one AI start to steamroll the others and become a superpower by the late medieval period. Japan seems to do this often. When I do go to war I find I'm up against 2x the number of troops I'd normally expect. So good job on your mod. I look forward to seeing more balance tweaks in the future.

Your changes to the social policies are really good - I like that you didn't change a lot, just enough to balance them out better - like improving piety.

I'd like to see some of the civs tweaked next. Some of them have pointless UA's, like the Ottomans. Some of them have weak UA's, like the Celts. Am I the only one that doesn't like the English UA? Why do they get a speed advantage? Why not an actual combat advantage to ships? Ugh, maybe it's just me...

 

[Acken]

Acken, you joined a thread I created about the lacklustre AI in the unmodded game, maybe two months ago, and suggested I try your mod.

I took a 2 month break from Civ5 but I'm back, and just started playing your mod. At the moment I have paired it with another mod that tweaks the civs slightly, because I find some of the civs UA's horribly stupid/boring/pointless to play. The other mod is called "weak Civ AI balance" or something like that.

Anyway, it does certainly seem like you have drastically improved the AI. I have played 2 games and am on a third now and have noticed it's quite common to see one AI start to steamroll the others and become a superpower by the late medieval period. Japan seems to do this often. When I do go to war I find I'm up against 2x the number of troops I'd normally expect. So good job on your mod. I look forward to seeing more balance tweaks in the future.

Your changes to the social policies are really good - I like that you didn't change a lot, just enough to balance them out better - like improving piety.

I'd like to see some of the civs tweaked next. Some of them have pointless UA's, like the Ottomans. Some of them have weak UA's, like the Celts. Am I the only one that doesn't like the English UA? Why do they get a speed advantage? Why not an actual combat advantage to ships? Ugh, maybe it's just me...

Hey thank you for playing and compliments are appreciated.

Is this this mod ?
http://forums.civfanatics.com/showthread.php?t=528434

Note that the next versions will slowly work on the civs themselves. It's voluntary that they haven't changed yet as I want the "core" game to be as close to finished as possible.

 

[Lord Yanaek]

Something i noticed playing (well, loosing might be a better word ) a few games with your mod. The psychos from G&K are back

Now, that's not necessarily a bad thing. The ultra peaceful AIs from BNW got boring after too many games, but is it on purpose you reverted to the other extreme or are you still tweaking them? I would guess the later as you stated that random aggression will be lower in a future version. It's a bit strange when some of the most peaceful AIs (Theodora in one of my games) goes Honor and start annexing CSs

Also it might be just some joke from RNGesus but it seems like on any map i start most AIs (like 6 out of 7) will choose the same starting policy tree (Liberty or Honor).

 

[yung.carl.jung]

People have been saying this a lot, but it isn't true at all. The AI is not more aggressive, fact of the matter is that the AI now has a lot more military strength compared to the player which means they are far more likely to DOW you and you also cannot bribe them in most cases, which means you don't get out of a horrible situation for a few GPT like in the base game.

I'm trying out Deity just for fun and was DOWed on T27 by Bismarck. His army consisted of about five or six warriors, six archers, a chariot and a spear. That was just what I saw, he probably had more units lurking in the fog of war. It should come to no surprise to anyone that the AI acts more "psycho" when they're fielding almost twice the units they do in the unmodded game.

Even keeping your unit score up won't help you much if they covet your lands and or have a casus belli against you. In my current game I had a military score of about 250.000, which means about two musket and a gatling for every city I had (7) and I still got DOWed.

 

[Acken]

Well both are true actually. The mod does increase the diplomatic impact of land dispute and competitiveness which lead to a more agressive AI.
The mod also decreases the impact of economy in war valuation. The AI used to value your economy as a factor for not attacking you ! So if you have a great economy it also had a similar impact to having a great army.

But yes I'm still tweaking it especially those double DoWs where one AI don't even participate. I'd also like them to be more cautious in breaking a DoF.

 

[yung.carl.jung]

I get more double DOWs then I get single DOWs nowadays, you get used to it pretty quickly, most of the time I know they're coming

 

[Phukit]

Also it might be just some joke from RNGesus but it seems like on any map i start most AIs (like 6 out of 7) will choose the same starting policy tree (Liberty or Honor).

Given that this mod is geared towards DomV and early aggression Liberty and Honor are clearly superior to either tradition or piety as opening trees since they help facilitate quick early expansion and/or highly promoted cheaply promoted units early on.Both of which are conducive to early war.

 

[Phukit]

I get more double DOWs then I get single DOWs nowadays, you get used to it pretty quickly, most of the time I know they're coming

Loading this mod is the first and best clue that a double/triple DoW is coming by T60, all other factors are irrelevant

 

[Lord Yanaek]

People have been saying this a lot, but it isn't true at all. The AI is not more aggressive...

Military strength is a factor of course, but i didn't really noticed more units in the hands of the AI, quite the opposite. Since they tend to be at war with someone or another about 80% of the time, they don't accumulate a large army, yet they still DoW everyone everywhere. I was playing Demigod but even comparing to the army Immortal AIs can accumulate in vanilla CiV when they don't have a lot of wars, it didn't look like they had more at any given time.
And even in vanilla CiV, any Deity AI has a vastly superior army to a CS, yet i've never seen Theodora capture 3 of them

The mod also decreases the impact of economy in war valuation. The AI used to value your economy as a factor for not attacking you ! So if you have a great economy it also had a similar impact to having a great army.

Was it all economy? I always thought is was raw stockpile which made sense as you could easily buy lots of troops at once.

But yes I'm still tweaking it especially those double DoWs where one AI don't even participate. I'd also like them to be more cautious in breaking a DoF.

Good. Actually it was a refreshing (if surprising) fact to encounter far more aggressive AIs, but i fear if they all act like psychos all the time it will become boring at some point. Looks like finding the right balance isn't easy thought, Firaxis obviously gave up on this so "good luck commander"

Given that this mod is geared towards DomV and early aggression Liberty and Honor are clearly superior to either tradition or piety ...

Is it? I haven't seen this in the description.
Again, variety is good. It's nice that AIs don't all go Piety as it makes it possible for the player to actually get a religion if they want to, but i don't think it's all that fun when every single AI goes Liberty (or when every single AI goes Honor). Ideally, some mix of starting trees would lead to the greatest diversity, and if tall vs wide is balanced, there is nothing wrong with AIs going tall either, provided they do it like a player would (growing cities, using those specialists) which i think i read they would.
And btw, when i told they all choose Liberty or Honor, it meant that one game they all went Liberty and the next one, they all went Honor

 

[Acken]

All of them chosing the same tree is pure RNG. Basically the AI just multiplies its flavors by the flavors of policy, get a total score and then randomize between the top 2.
Piety is underused due to a mistake in v3 though.

 

[yung.carl.jung]

I still play Tradition most of the time actually. Fact of the matter is that if you settle more than three cities you're pretty much guaranteed to get DOWed very early, which makes the whole thing counterintuitive.

I also disagree, as Acken has already noted, that melee units are the new go-to.

For me, roles have reversed, sort of. While in the unmodded game all you need to defend are a few upgraded Archer-line units, you now need a ton of melees with citadel and range support.

Archers are now extremely vulnerable, they can get oneshot by Spears and those are running around from turn 15 onwards.

Composite Archers suffer from the same problem. I think Liberty Composite rushes was by far one of the most viable strategies in the unmodded game, but they are incredibly hard to pull off now.

I still think Crossbows are the premier military unit until Artillery if you want to take cities. For me, nothing has changed concerning the importance of Crossbows. Almost all of my timed pushes are done with Crossbow upgrades.

Other military techs I would value highly: Swordsmen, Artillery, Bombers, XCom. Frigates and Battleships for water-based maps, of course.

Bomber tech has a really special role since it is on the path to hubble and gives you

a) Strong offense with Bombers
b) Strong defense with Paratroopers, who are much more durable than GWI, but don't require deviation from the standard tech path
c) The availability of a Bomber counter unit, although I value Anti Air Guns a little higher

 

[Phukit]

All of them chosing the same tree is pure RNG. Basically the AI just multiplies its flavors by the flavors of policy, get a total score and then randomize between the top 2.
Piety is underused due to a mistake in v3 though.

Yes but Liberty and Honor are the top 2 95% of the time.

I am astonished in V3 anytime the AI opens tradition.

Thats not to say they dont take tradition at all,but its usually taken as a filler after liberty or honor by the AI

 

[beetle]

Given that this mod is geared towards DomV and early aggression Liberty and Honor are clearly superior to either tradition or piety as opening trees...

I am hoping Acken might refute your assertion, but I understood that a point of the mod was to make Piety more viable. That worked out for me with this v1 game, with a SV no less. But I see that you saw that...

 

[Acken]

I think it does. Maybe Piety isn't the best to open with but I see a lot of value in it.

I have had good results with the following opener in a recent Byzantium game to test their changes:
Open Tradition
Open Piety
Legalism
Organized religion
Finish Tradition
Finish Piety

These policies give a ton of culture and faith in 4 cities allowing you to quickly finish both trees while developping a religion. You get a lot of free/cheaper stuff in every cities, monument shrine aqueduct temple walls, then you have both the best gold output of all starting trees allowing you to help field an army or make new cities. You also have the best happiness and growth of the 4 trees.

The fact that the AI doesn't take tradition though is unrelated to the strength of the trees. The AI doesn'T think like that. It'd be just a matter of unbalanced flavor.

 

[Phukit]

I am hoping Acken might refute your assertion, but I understood that a point of the mod was to make Piety more viable. That worked out for me with this v1 game, with a SV no less. But I see that you saw that...

Dont get me wrong,tradition and piety are both viable...but they are also both clearly inferior to liberty and honor as far as opening trees go imo.

 

[Phukit]

The fact that the AI doesn't take tradition though is unrelated to the strength of the trees. The AI doesn'T think like that. It'd be just a matter of unbalanced flavor.

This doesnt change the fact that tradition and piety are clearly inferior opening trees than liberty and honor.

 

[Phukit]

All of them chosing the same tree is pure RNG. Basically the AI just multiplies its flavors by the flavors of policy, get a total score and then randomize between the top 2.
Piety is underused due to a mistake in v3 though.

it would appear the same mistake applies to tradition as well,in V2 i saw a pretty even mix between the 4 opening trees....now its just liberty and honor with very rare exceptions.

 

[Acken]

No it's not the same reason. Piety was due to a programming mistake dividing religion flavor by 2 while tradition is probably due to the tree not having expansion flavor while the AI was given more expansion flavor in v3, I've added some to see if it helps.

 

[Lord Yanaek]

OK. I messed up a new game (my fault forgetting to grow enough while trying to adapt to some new stuff), but this time the mix is better. 2 tradition, 2 honor, 3 liberty and makes sense (Gandhi goes Tradition, Bismark and Suleiman Honor). Piety is the only tree totally ignored.

I still think Crossbows are the premier military unit until Artillery if you want to take cities. For me, nothing has changed concerning the importance of Crossbows. Almost all of my timed pushes are done with Crossbow upgrades.

Bows (as a group) are weaker, swords are stronger so unit to unit balance has changed. However, cities are also weaker now so bow to cities balance hasn't really changed. I often considered giving bows the same penalty against cities as horses have (a bow was never meant to take down walls, and even flaming arrows were mostly used to draw people out of flammable buildings so they could be killed more easily so it would probably make sense.

This doesnt change the fact that tradition and piety are clearly inferior opening trees than liberty and honor.

Are they? I think it depends on how you play the game. Tradition was always better for peaceful games and almost useless for early warmongers. I don't think this mod changes enough to make things different so if you tend to be aggressive, you will get better results with Liberty or Honor. I've often been a peaceful or late warmonger so Tradition was often better for me. Will try it again next time and see if it's really worse now

EDIT : @ Acken : yep, seeing how Harun has expanded like crazy in my last failed game while he opened tradition, you can safely give it some expansion flavor