Next Beta Preview Changelog

[Gazebo]

Hey all,

Posting the current (WIP) changelog for dissection. Barring any issues a new beta later this week. Log:

General

• Major bugfixes and performance improvements for AI city gold purchase logic
• Reduced AI denounce rate slightly
• Added scaling value for war weariness from city capture based on # of times you've owned the city.
• Bugfixes for AI settlement site selection and settler movement logic
• Added UI element for sanctioned civs (the deal value bar will be replaced with a SANCTIONED warning)
• Improvements to building and unit production logic
• Improvements to city citizen food management logic
• Fixed broken Evangelism belief
• Adjusted Influence change UI logic for culture overview to be more accurate
• Improved AI tile purchasing proclivity and logic (will snap up tiles that link cities, or are of strategic value)
• Better RNG system for avoiding MP desyncs
• Soon: alpha MP build based on DaveNye's work

Balance

• Feature:
  • Revolutions system finalized:
    • Cities will liberate into free city-states if the losing player lacks an ideology and is not influenced by anyone with an ideology
    • This means that empires can overexpand and fracture into city-states well into the modern period (and beyond, if no influence is happening)
    • Free cities count as city-states for all gameplay purposes, and will respawn as the same city-state if the city is recaptured and then re-liberated once again
• General
  • Increased policy cost exponent to 2.22 (was 2.19) (ideologies more regularly appear after 1900, pushing culture victory back a bit)
  • Puppets: no longer penalize for food/production, but all other yields at 80% penalty
  • Increased happiness tech scaler to 2.0 (was 1.9)
• Civs:
  • Venice: UA now a 30% reduction of puppet penalties (so puppets are 50% of all yields except food/production, which are 100%)
• Beliefs
  • Church - now grants 2 Culture, 40% pressure mod (was 25%)
• Espionage:
  • Decreased value of early game spy power by 100 (50% weaker) - this value scales downward with Era, so spies naturally get better as the game progresses
• Units:
  • Moved GP scaling from owned improvements/works/themes to XML for modders
    • Scaling on owned improvements only applies to improvements you created yourself
  • Workers and Archaeologists can move through ice with Ice Breakers promotion (like work boats)
  • Increased air defense of guerilla to 8 (was 5)
  • Reduced archaeologist base faith cost to 225 (was 400)
  • Added option for GP planting on artifacts to disable it (BALANCE_CORE_ARCHAEOLOGY_FROM_GP)
• Buildings
  • Borodbudur: now grants just 2 free Missionaries (from 3), but all Missionaries get an extra spread.
  • Louvre:
    • theme bonus now 15 Culture
    • grants free Museum and 2 free archaeologists when completed
  • Uffizi:
    • theme bonus now 10 Culture
    • grants all great works +1 culture
  • Globe Theater:
    • theme bonus now 10 Gold
    • Artist, Musician, and Writer specialists in all cities produce +1 GAP
  • Slater Mill: now 4 coal (was 2)
• Improvements
  • Changed remove feature logic:
    • Marsh removal moved to iron working, increased base cost by 5 turns
    • Jungle removal moved to calendar, increased base cost by 4 turns
    • Forest removal cost increased by 4 turns
    • Cost of each is reduced by 2 turns at Machinery and Biology

Edit:

A few things we need to discuss:

AI Warmongers are really doing well right now. Like...really well. I'm worried war is too beneficial. The Zulu are especially scary.

• Zulu nerf: hit promotions, or UA?
• Early warmongers - do we need more base defense for cities/walls?
• Courthouse - too cheap, cost should scale a bit more when going wide?

I don't want to upend the cart, but war is definitely swinging the < turn 150 game for AI tests, so a soft touch is needed somewhere.

Cheers,
G

 

[Stalker0]

G, is this something you have seen very recently....aka after the change where population doesn't affect city strength anymore?

Could that be making cities too easy to take?

 

[pineappledan]

A few questions:

> Venice: UA now a 30% reduction of puppet penalties (so puppets are 50% of all yields except food/production, which are 100%)
1. Glad to see the Venice nerf. Felt like they were getting all the reward for none of the micro. Were they outperforming obviously?

> Moved GP scaling from owned improvements/works/themes to XML for modders
2. Perhaps I should just wait and see when it comes out, but can I get a little more info on what this entails? Is this a boolean 0/1 value, for whether the GP expending affects bulbing, or is it more customizable than that?

• Could a modder make a UGP that gives double scaling?
• Could a modder make a UGP that scales bulbing regardless of what action they do? ie. could a Unique GEngineer scale bulbing by using its Hurry Production action?
• Could a modder make a UGP that scales bulbing on multiple different GPs? ie. Could a unique engineer's tile scale for both Great Engineers and Merchants?

> Workers and Archaeologists can move through ice with Ice Breakers promotion (like work boats)
3. Could I get more detail on this? Is this a promotion that is unlocked at a certain tech?

> AI Warmongers are really doing well right now. Like...really well. I'm worried war is too beneficial. The Zulu are especially scary.
4. In your opinion, what is the more likely culprit? Is this due to recent changes in AI, them getting better at war, or is this the change to city defense scaling?

 

[tu_79]

Overall I like the new changelog.

Regarding warmongers, there has been a few changes that can affect them.

My most hated change was the 0 unhappiness for puppets. Does the warmonger AI use puppets? It allows for a much bigger aggressive empire, with the only drawback of not adding more units to help securing borders.

The other one, already pointed at, the change to city wall strength. Progress civs that usually overextend will not have too much population in their frontier cities. Easy prey?

 

[pza]

On prince (up until 3/31, didn't manage to update yet), I never see the AI sucessfully wage war* anymore, but their strategy in war (against humans) seems to have become quite good, so probably the enemy AIs are just defending (too?) well. Probably a reason why AI-controlled civ's like Rome seem to constantly fall behind.

*With never see sucessful war i mean only in the rarest cases I see a city change ownership, and haven't seen a forceful vassalization or defeat in aaaages. I might be playing to land-rich pangeas.

• Better RNG system for avoiding MP desyncs
• Soon: alpha MP build based on DaveNye's work

I stood up and made a little dance.

 

[Paramecium]

A few things we need to discuss:

AI Warmongers are really doing well right now. Like...really well. I'm worried war is too beneficial. The Zulu are especially scary.

Yes that I am experiencing in my current playthrough and my last one. Warmonger especially with early boni are eating at the moment late bloomer in that game. Like I said, on my last playthrough (patch 4-11-18) I played Austria tall on standard communitas and I played peacefull the until the last game, meaning that I just defended my realm. On my continent I had carthage, maya and aztec, who fighted each other a lot, but carthage reached a point where they vassalized the other both.
But the other continent was a huge battleground and graveyard. It had the other 6 civ. When I reached that continent, I just saw the last moments of Brazil before getting eaten by Japan, the only religion on that landmass was founded by German, but it already disappeared, conquered by the Zulu. And the last civs were Denmark, who ate Spain. Over the course of the game, first Japan dominated, but in the end, Denmark crushed the Zulus, nearly Japan and even invaded my continent crushing the maya. I have to admit, that it was quite rare for me to see a AI civ getting 6 of 10 capitals ...

 

[Owlbebach]

Glad to see the Venice nerf. Felt like they were getting all the reward for none of the micro. Were they outperforming obviously?

In my last Deity game i vassalized 2 AI before turn 120 with Venice... Starting from turn80 i was around top4 in tech.

Though in Renaissance i screwed up my attack on Russia, but she was top1 in both tech, army and production.

 

[wobuffet]

Looks fantastic, G!

• Increased happiness tech scaler to 2.0 (was 1.9)

Does this mean the Unhappiness-to-Techs curve is steeper (more Techs discovered means higher Needs) than before?

G, is this something you have seen very recently....aka after the change where population doesn't affect city strength anymore?

Could that be making cities too easy to take?

This seems very plausible. Possible solutions:

• Local war efforts:
  • Give all cities +10 Defense but also -1 per Local Unhappiness
  • so Cities with more infrastructure are harder to take over and newer/less-developed ones more vulnerable — could be an interesting dynamic, esp. early-game (harder to REX/early-game spread)
  • also indirectly makes Pillaging more powerful(!)

• Empire-wide Military Supply effects:
  • Give empire-wide City Defense bonuses depending on Military Supply, evenly split among Cities
  • e.g., if your Military Supply is 20 and you have 5 cities, they each get +4 Defense
  • gives a natural, player-controlled defensive advantage to small, turtle-y empires

• Military training buildings:
  • Barracks, Armory, Military Academy increase City Defense slightly
  • gives player more granular control: e.g., build Walls to turtle up (defense only) or build Barracks for some defense and some offense (making units)

 

[CrazyG]

I would take a few % points away from some of the Zulu promotions. I haven't tracked data, but I would guess that the Zulu are by far the most common civ that I lose to.

I would guess that the changes are related to puppets are pushing warmongers higher.

 

[pza]

I'd blame the zulu, tbh. War, it's what they do, but having a ~30% stronger swarm of melee units that can run 3 tiles in any stage of the game (barracks are very early, giving the best promotions) is crazy. And then there's the Impi, one of the best UU's in the game. Zulu-Neighborhood is even for other early warmongers like askia "keep clear" zone, they defend like champions. (can't speak about their offensive warmongering, since like i said before, it doesn't seem to occur on prince effectively enough)

 

[Tantro]

In my current immortal game Mongolia is set to dominate the map. He is on a rampage to say the least. Anyway, out of his 22 cities nine are puppets. I believe the problem lies there.

 

[theMisterRud]

In the last 3 months I have been playing warmonger-only, with various Civs (Denmark, Zulu, Sweden especially) and I noticed that on King Mining -> Pottery -> Bronze Working with Statue of Zeus rushing is just too good and consistent, if you have a decent source of income (unless you are Denmark): you hardly need Archers to take 6-11 defense cities. I'm wandering if the Spearmen buff is to blame

The Zulus are just .... too many of them : is like having a Zerg Rush with Protoss units in Starcraft. If they are really better than other warmongers, I would look at the UA.

3 other things (throwing them out there)

- in the last patches, poverty and literacy have been my main sources of unhappiness, there is always that awkward "I have to wait Banking to get constabularies" gap, but aside from that, happiness management seams okay (though on 4-11 I had to manage specialists manually more, especially regarding literacy)
- Tributes and Pledges of protection: seams that the AI is very lenient if you play bully, I know overtuning might lead to human exploits, but maybe a slight adjustment is desirable ? Food for thought
- Even pacific Civs tend to declare war a little to much in the early game, having Arabia declare war on you for territorial disputes with Warriors while you have 1-promo Spearmen is ... funny.

As always, thank you all for your gigantic effort. I do sincerly believe Vox Populi is the best 4X in the market right now.

 

[civplayer33]

Courthouses are pretty cheap...increasing their costs is reasonable.

Also, how about simply reducing the AI supply handicap? The defender would be affected as well, sure, but he can create units in the war zone while the attacker has to move them there first, which means this would be an indirect buff for the defending AI.
Edit: this would also indirectly buff cities, in a way, since now less units will be assaulting them per turn.

On a different topic, is it possible to increase the number of turns for which instant yields are averaged to 20 or 30? I feel like 10 turns doesn't reflect the real bonuses well enough since the instant yields are crazy large sometimes and 10 turn averages won't smooth them out enough to get a good sense of their contribution IMO.

 

[tu_79]

is it possible to increase the number of turns for which instant yields are averaged to 20 or 30?

@Gazebo, could this be the same issue that you found in the influence trend code?

 

[Enrico Swagolo]

Changes seem cool to me, but is happiness getting reduced again? It's pretty harsh as it is if my environment is unpleasant, I hope the change won't be too considerable.

I think Celts also might need a nerf. They always do way better than Zulu in my games, I've seen Shaka get crushed but never Boudicca (unless I did the crushing). She always runs away with more policies and techs, especially if she's not acting stupid and takes a better pantheon. Like take Dagda or any other weird named thing, then look at the UB come Medieval. Take a long, calm peek, especially Dagda though. The Dagda UB alone pretty much means you don't have to care about city unhappiness on a civ with a guaranteed, parasitic religion that's always the best in the game but only for you so you don't have to fear someone stealing it or even adopting it, like Byzantium if she goes for wide, yield-per-pop religion. All that when the Pict comes earlier, eats everything and provides a ton of Faith. This is absolutely broken.

 

[tu_79]

The Zulus are just .... too many of them : is like having a Zerg Rush with Protoss units in Starcraft. If they are really better than other warmongers, I would look at the UA.

They might be, but Zulus have not been changed recently, that I remembered, so it's not their UA what causes the domination imbalance.

 

[Gokudo01]

Courthouses are pretty cheap...increasing their costs is reasonable.

Also, how about simply reducing the AI supply handicap? The defender would be affected as well, sure, but he can create units in the war zone while the attacker has to move them there first, which means this would be an indirect buff for the defending AI.
Edit: this would also indirectly buff cities, in a way, since now less units will be assaulting them per turn.

On a different topic, is it possible to increase the number of turns for which instant yields are averaged to 20 or 30? I feel like 10 turns doesn't reflect the real bonuses well enough since the instant yields are crazy large sometimes and 10 turn averages won't smooth them out enough to get a good sense of their contribution IMO.

I don't know if courthouses are to blame for <150 snowballing warmongering.
I 'm not against any zulu nerf ( a slight nerf to the flanking and defence bonus wouldn't be too much) but we have to review the changes of the last months :
- AI is more efficient at warring : better tactic better unit management.
- Spearman buff : it really changed the early war : less vulnerable to range shot ( the difference between the archer strength and spearman is huge now even without cover ) deal lots of damage to city(with drill), and you can spam them. I won't talk about pictish warrior.
- No more city strength from population.
- No more unhappiness from puppet.

all those things made AI better at managing early war and quick (military ) expansion, not just the zulu. It's more efficient on the zulu because they are the best at war and with zero economic bonus. while the huns, rome or mongalia have still got some powerful "economic" bonus when they are not fighting.


I would like a small nerf to spearman ( and don't forget the UU, they are too strong too ) : don't nerf the pikeman.
barrack and armory could give a really small strength boost ( like constabulary, for example 2/3 extra strength )

 

[civplayer33]

I don't know if courthouses are to blame for <150 snowballing warmongering.

I didn't say they were...it's one of the points in Gazebo's OP so I affirmed that the change makes sense.

The AI supply bonus nerf I proposed would counter the AI's better unit management and tactics to some degree (it may be better to look at it as the AI requiring better management and tactics to do well with less units so this nerf would be fairly "natural" IMO); it would also reduce the unit carpets to make cities last a bit longer on the whole (not necessarily individual cities, though, but it would give the defender more time overall).

I wouldn't oppose a small def bonus for barracks and maybe armories, though.

 

[raphaell666]

I don't want to upend the cart, but war is definitely swinging the < turn 150 game for AI tests, so a soft touch is needed somewhere.

I believe that the nerf to puppet cities output penalty from 75% to 80% should be a heavy enough touch for this, it translates into a reduction of 20% in the yield output of puppets (from overall output 25% to 20%).

I also agree that puppets were relatively too strong, but I am not sure if nerfing their yields output is really the main issue. I feel that happiness is, but I am not sure on how to address it.

I had an idea right now, what if we tried making puppets closer to vassals, as in they initially give the owner no gold, but you can tax them increasing unhappiness? Without changing the values of the other yields. Example: 0% tax with 0 unhappiness, 10% gold output with 1 unhappiness, 20% gold with 2 unhapiness, etc. Reducing the gold available to warmongers would help reduce their effectiveness, I think.

 

[ElliotS]

AI Warmongers are really doing well right now. Like...really well. I'm worried war is too beneficial. The Zulu are especially scary.

• Zulu nerf: hit promotions, or UA?
• Early warmongers - do we need more base defense for cities/walls?
• Courthouse - too cheap, cost should scale a bit more when going wide?

I don't want to upend the cart, but war is definitely swinging the < turn 150 game for AI tests, so a soft touch is needed somewhere.

Actually I think people are missing the forest for the trees.

The best strategy for a player has ALWAYS been early and opportunistic war. Even tradition civs are recommended to pillage and harass (If not always puppet or annex) neighbors.

So it's very possible that AIs have just been improved to the point where they're competent enough to consistently get advantage from wars, just like players.

Like in an all human multiplayer game if I heard that half of the players won a war (+50 warscore) and half lost a war (-50 warscore) pre-turn-150, I would give a big advantage to the winners of the war. (85% chance that someone from the winner group will win.)

This would be true this patch, the patch before, the patches where puppets are bad, the patches where puppets are even more OP than they are now and so on.

So while it's possible that the mechanics are unbalanced, I think we need to consider if changing any of them can actually help without making war detrimental to the victor (which would be dumb obviously).