New Beta Version - January 20th (1/20)

[Aristos]

Mind you I'd have to ask why anyone still uses IE!

bad habits don't go away easily...

 

[Funak]

Seeing double food on holy-sites from Sacred Garden, not sure if it isolated to just that wonder (the AI is horrible at building their religious wonder). Anyways my holy-sites provide +10 food.

 

[merill]

the AI is horrible at building their religious wonde

me too

 

[Moi Magnus]

If there's a part of 'human behavior' that I've missed, let me know.
G

Some "human behavior" that are maybe not in the current version :

+ Building chains : sometime, some not-so-usefull building are needed for more usefull ones. (Example : if I remember correctly, Forge is sometime useless but unlock some really good buildings).

+ Useless spectialist slots : if the city has not suffisantly pop potential, Guilds are uselless, and specialist-building are less usefull.

+ Free buildings : Some wonders, and some policies (mostly ideologies) give free buildings. I will not build them if I'm almost sure to have them free.

+ Delaying Settler : If the city grow in 2 turns, I wait 2 turns before starting a Settler.

+ Production / Tech synchronisation : "I unlock writing in 4 turns, and I would like to build as soon as possible a library (or the Great Library) in my capital, so I will not begin a world wonder or anything really long". Similarly, if I'm about to unlock a powefull unit, I will maybe build a lot of the outdated unit to upgrade them immediatly and attack, or at the contrary wait for the tech if I can't afford upgrading a lot of units.
=> This kind of behaviour is probably difficult to code, and I'm not sure it will really improve the IA.

 

[Gokudo01]

Hi Gazebo,

Is Authority scaler supposed to be 3 or 4 percent ?

Because your tooltip says 4% per policy but the xml files say 3% per policy.

 

[Enrico Swagolo]

Am I the only one who thinks Building costs should be scaled down?

It's a bit weird for me that a Pioneer/Conquistador raised city is going to have more constructions than pretty much anything in your civ excluding capital and maybe 2nd/3rd city if they've got a good location.

I play on Epic.

 

[Gazebo]

Hi Gazebo,

Is Authority scaler supposed to be 3 or 4 percent ?

Because your tooltip says 4% per policy but the xml files say 3% per policy.

3%. I'll fix.

G

 

[phantomaxl1207]

Am I the only one who thinks Building costs should be scaled down?

It's a bit weird for me that a Pioneer/Conquistador raised city is going to have more constructions than pretty much anything in your civ excluding capital and maybe 2nd/3rd city if they've got a good location.

I play on Epic.

I think Building Purchase Costs should be cheaper in the Early game and a bit cheaper compared to Units, as it will make early gold look better and you only get half the Building.

Purchase Costs should scale more with # of Cities than w/ # of Techs for the sake of smaller Civs w/ Gold Focus.

 

[Gazebo]

I think Building Purchase Costs should be cheaper in the Early game and a bit cheaper compared to Units, as it will make early gold look better and you only get half the Building.

Purchase Costs should scale more with # of Cities than w/ # of Techs for the sake of smaller Civs w/ Gold Focus.

I agree - building gold costs are a little to high at present.

G

 

[Funak]

Am I the only one who thinks Building costs should be scaled down?

It's a bit weird for me that a Pioneer/Conquistador raised city is going to have more constructions than pretty much anything in your civ excluding capital and maybe 2nd/3rd city if they've got a good location.

I play on Epic.

I don't see this at all, anything settled in ancient/classical era is going to be ahead of pioneer cities easily. Sure, settling cities by the time you're at civil service isn't necessarily a good move (you're still going to be more likely to get a good location however), but still.

 

[Atlas627]

Some "human behavior" that are maybe not in the current version :

+ Building chains : sometime, some not-so-usefull building are needed for more usefull ones. (Example : if I remember correctly, Forge is sometime useless but unlock some really good buildings).

+ Useless spectialist slots : if the city has not suffisantly pop potential, Guilds are uselless, and specialist-building are less usefull.

+ Free buildings : Some wonders, and some policies (mostly ideologies) give free buildings. I will not build them if I'm almost sure to have them free.

+ Delaying Settler : If the city grow in 2 turns, I wait 2 turns before starting a Settler.

+ Production / Tech synchronisation : "I unlock writing in 4 turns, and I would like to build as soon as possible a library (or the Great Library) in my capital, so I will not begin a world wonder or anything really long". Similarly, if I'm about to unlock a powefull unit, I will maybe build a lot of the outdated unit to upgrade them immediatly and attack, or at the contrary wait for the tech if I can't afford upgrading a lot of units.
=> This kind of behaviour is probably difficult to code, and I'm not sure it will really improve the IA.

These are good ones.

To that I'll add

- Let the other players vote for or against something while you vote for/against something nobody else will.

- Propose something you don't care about but you know other people will have to vote on, sucking away their votes.

- Propose something just to appease other players.

- Avoid spying on or proposing a vote a potential/current friend will dislike.

- Avoid declaring friendship to avoid the hate of their enemies.

- Avoid starting a wonder if you're about to get a Great Engineer and want to rush it

- Save to purchase things that are more efficient and build things that are less efficient.

I can come up with more later if I get time. These are hard enough already.

 

[Gazebo]

1 Building chains : sometime, some not-so-usefull building are needed for more usefull ones. (Example : if I remember correctly, Forge is sometime useless but unlock some really good buildings).

2 Useless spectialist slots : if the city has not suffisantly pop potential, Guilds are uselless, and specialist-building are less usefull.

3 Free buildings : Some wonders, and some policies (mostly ideologies) give free buildings. I will not build them if I'm almost sure to have them free.

4 Delaying Settler : If the city grow in 2 turns, I wait 2 turns before starting a Settler.

5 Production / Tech synchronisation : "I unlock writing in 4 turns, and I would like to build as soon as possible a library (or the Great Library) in my capital, so I will not begin a world wonder or anything really long". Similarly, if I'm about to unlock a powefull unit, I will maybe build a lot of the outdated unit to upgrade them immediatly and attack, or at the contrary wait for the tech if I can't afford upgrading a lot of units.

- Let the other players vote for or against something while you vote for/against something nobody else will.

6 Propose something you don't care about but you know other people will have to vote on, sucking away their votes.

7 Propose something just to appease other players.

8 Avoid spying on or proposing a vote a potential/current friend will dislike.

9 Avoid declaring friendship to avoid the hate of their enemies.

10 Avoid starting a wonder if you're about to get a Great Engineer and want to rush it

11 Save to purchase things that are more efficient and build things that are less efficient.

Good lists!

1, 5, 10, and 11 would be a little too intensive for the CPU, as right now the functions are restricted to things you have/can do right now (or have already done).

5 and 6 are hard to do because of predictive behavior, but the other WC ones make sense.

The others are already present.

 

[Enrico Swagolo]

I don't see this at all, anything settled in ancient/classical era is going to be ahead of pioneer cities easily. Sure, settling cities by the time you're at civil service isn't necessarily a good move (you're still going to be more likely to get a good location however), but still.

I agree that they're ahead in Pop, but in buildings? Not in my case at least. Capital, okay, no problem. Second/third city - okay (easily if in a very good spot). Fourth? Leagues behind (unless the location is extraordinary), even if the fourth I settle by late very early classical or end of ancient (which I do often).

I feel as it could partially be due to technological growth being a bit too fast, because of which the buildings come too quickly.

Do you also play on Epic?

 

[Moi Magnus]

The others are already present.

I've just think of another detail :
During a game with team enabled (and research agreement), an AI declare wars on my teammate while I had a declaration of friendship (and a research agreement) with his teammate.
It was a very bad move :
+His teammate was forced to become a backstaber.
+They lose a research agreement
+We were two on his island, and his teammate was in another, so we were two against one, and we eliminate him.
=> I'm not sure teams are correctly handle by AI.
(This is probably not the good place to post this)

 

[Funak]

I agree that they're ahead in Pop, but in buildings? Not in my case at least. Capital, okay, no problem. Second/third city - okay (easily if in a very good spot). Fourth? Leagues behind (unless the location is extraordinary), even if the fourth I settle by late very early classical or end of ancient (which I do often).

You kinda have to work to keep the city relevant, focus on getting production-buildings up first, manually selecting tiles with decent production on them and not just let the AI run all specialists.

I feel as it could partially be due to technological growth being a bit too fast, because of which the buildings come too quickly.

I play on standard, but there shouldn't be a difference there.

 

[Gazebo]

I've just think of another detail :
During a game with team enabled (and research agreement), an AI declare wars on my teammate while I had a declaration of friendship (and a research agreement) with his teammate.
It was a very bad move :
+His teammate was forced to become a backstaber.
+They lose a research agreement
+We were two on his island, and his teammate was in another, so we were two against one, and we eliminate him.
=> I'm not sure teams are correctly handle by AI.
(This is probably not the good place to post this)

Preset teams are tricky, as the AI tends to look at playertypes more than teamtypes. That'd be quite the rewrite (my recommendation is to not use preset teams).

G

 

[Zanteogo]

City states are not declaring war when their allies declare war on other civs. (for the AI anyways)

View attachment nowar.Civ5Save

 

[Aristos]

City states are not declaring war when their allies declare war on other civs. (for the AI anyways)

strange... I just had Poland and all its 3 CS allies declare war on me, business as usual... are you sure you don't have some setup mismatch?

 

[Zanteogo]

strange... I just had Poland and all its 3 CS allies declare war on me, business as usual... are you sure you don't have some setup mismatch?

Pretty sure everything is setup normally.

Attached save to original post if anyone wants to look.