4 swordsmen on turn 25 - AI cheating?

[NombreyApellido]

humankind AI is so advanced, mama ocllo was able to recruit 3 swords (4 in total, there's another in the city) by turn 25 and she did it while being ranked on 6th place with just one era star and with a Babylon size 4 on un remarkable location (i counted 15 food 31 industry with 1 attached territory, a copper mine, an iron mine and a horse ranch)

she had +10 (melody) and +14 (darkest depth) industry though

isn't this game just FUNNY?

 

[fortydayweekend]

humankind AI is so advanced, mama ocllo was able to recruit 3 swords (4 in total, there's another in the city) by turn 25 and she did it while being ranked on 6th place with just one era star and with a Babylon size 4 on un remarkable location

she had +10 (melody) and +14 (darkest depth) industry though

isn't this just FUNNY?

Wow I've never seen the

humankind AI is so advanced, mama ocllo was able to recruit 3 swords (4 in total, there's another in the city) by turn 25 and she did it while being ranked on 6th place with just one era star and with a Babylon size 4 on un remarkable location (i counted 15 food 31 industry with 1 attached territory, a copper mine, an iron mine and a horse ranch)

she had +10 (melody) and +14 (darkest depth) industry though

isn't this game just FUNNY?

Cripes. That seems buggy, seeing as nobody could have discovered iron yet

How much are they charging for a non aggression pact??

 

[NombreyApellido]

the strategy is valid but the timeframe doesn't fit... not only she was getting huge bonuses in research but it seems she has infinite gold available

never requested the non aggression deal... perhaps i need to look into it next time

i went on a surprise war after she attacked one of my scouts stacks

a surprise indeed

fortgot to mention she had a poorly placed astronomy house without adjacent farm

yay cheating AI!

 

[Taefin]

humankind AI is so advanced, mama ocllo was able to recruit 3 swords (4 in total, there's another in the city) by turn 25 and she did it while being ranked on 6th place with just one era star and with a Babylon size 4 on un remarkable location (i counted 15 food 31 industry with 1 attached territory, a copper mine, an iron mine and a horse ranch)

she had +10 (melody) and +14 (darkest depth) industry though

isn't this game just FUNNY?

The path to swords is a humble one indeed.

 

[Siptah]

the strategy is valid but the timeframe doesn't fit... not only she was getting huge bonuses in research but it seems she has infinite gold available

never requested the non aggression deal... perhaps i need to look into it next time

i went on a surprise war after she attacked one of my scouts stacks

a surprise indeed

fortgot to mention she had a poorly placed astronomy house without adjacent farm

yay cheating AI!

I suggest to report that as a bug. With the current bonuses/cheats the AI gets on Humankind, this shouldn't be possible.

 

[NombreyApellido]

i think it's working as designed. it's just that the design is a lousy one, meaning an AI that plays by a completely different ruleset

it's there to provide the illusion of an opponent using the same systems as the player when it actually isn't

the optimist in me wants to think it's just a placeholder

 

[Siptah]

AI factions follow the same rules as the human player, except for the Independents. They get the following bonuses on Humankind:

Spoiler :

-50% time to convert territories to religion / culture (rounded up)
-70% population food consumption
+25% science
+10 industry per population
+30 industry, money and food
+30 stability
-80% unit upkeep
-60% buyout cost
+10 fortification
+2 combat strength on units

A much better player than me said it should be possible to get swordsmen on turn 25 as Babylonians if you beeline for that (just 3 techs necessary).

 

[NombreyApellido]

it may be possible, yes. when having a very strong location, hitting very heavy cuiosities and knowing there's iron in your territory to begin with

BTW this cpu player not only got to standing army, but took carpentry and domestication too... not sure if calendar though

i maintain the AI of this game plays by a different set of rules: it is able to see the whole map, move faster and it would seem it's also able to get units on demand without population cost

i love the game, i'm very invested in it. but i think it's fair and honest to criticize it design flaws, its implementation shortcomings and attemps at solving its many issues

 

[fortydayweekend]

AI factions follow the same rules as the human player, except for the Independents. They get the following bonuses on Humankind:

Spoiler :

-50% time to convert territories to religion / culture (rounded up)
-70% population food consumption
+25% science
+10 industry per population
+30 industry, money and food
+30 stability
-80% unit upkeep
-60% buyout cost
+10 fortification
+2 combat strength on units

A much better player than me said it should be possible to get swordsmen on turn 25 as Babylonians if you beeline for that (just 3 techs necessary).

With all those bonuses it's really surprising that the player can so easily get ahead in population growth, even vs the Harrapans/.../Brazilians. Is it just because the AI is building/losing too many units??

 

[Siptah]

They lose a lot of units, and certainly choose and place districts very badly. Commons don’t give population slots, and the AI seems to build 90% commons (adjacent to commons, which renders them useless). However, one of the devs said recently on the G2G forums that they will improve that.

 

[fortydayweekend]

They lose a lot of units, and certainly choose and place districts very badly. Commons don’t give population slots, and the AI seems to build 90% commons (adjacent to commons, which renders them useless). However, one of the devs said recently on the G2G forums that they will improve that.

Good point, they are probably hitting the population cap for not having enough districts, while running a 300-food surplus

 

[Taefin]

+30 industry per city and +10 industry per pop explains a lot. With 6 pop that’s a free sword every turn on top of regular production sources.

The real question is what is the AI doing so wrong in mid-late game, when it struggles to field (and then rebuild) a reasonably sized army.

 

[NombreyApellido]

there must be a starting money and science bonus since they are able to build extractors and quarters the moment they place their capital

 

[Siptah]

hm... I've never understood what happens to the science you gather in the neolithic era. My impression so far was, that it is just lost (as usual with science when there is no tech available). Maybe the AI bonus science leads to something fishy there?

@The_J I think it would be nice to yank this discussion about the turn 25 Swordsmen and AI bonuses/rules out of this thread and into its own. What do you think?

 

[fortydayweekend]

hm... I've never understood what happens to the science you gather in the neolithic era. My impression so far was, that it is just lost (as usual with science when there is no tech available). Maybe the AI bonus science leads to something fishy there?

It's saved up and is spent on the first turn in Ancient. If you have a long neolithic you can sometimes get a couple of techs immediately.

 

[The_J]

@The_J I think it would be nice to yank this discussion about the turn 25 Swordsmen and AI bonuses/rules out of this thread and into its own. What do you think?

Moderator Action: -> moved.
Let me know if I forgot to move any posts (simply report missed posts, they'll be moved to this thread).

 

[NombreyApellido]

correct, but ai don't seem to do curiosities, and even so they don't provide money for buyouts

or am i missing something?

 

[Gedemon]

it may be possible, yes. when having a very strong location, hitting very heavy cuiosities and knowing there's iron in your territory to begin with

BTW this cpu player not only got to standing army, but took carpentry and domestication too... not sure if calendar though

i maintain the AI of this game plays by a different set of rules: it is able to see the whole map, move faster and it would seem it's also able to get units on demand without population cost

i love the game, i'm very invested in it. but i think it's fair and honest to criticize it design flaws, its implementation shortcomings and attemps at solving its many issues

The AI code is not hidden and is planned to be fully moddable, it should be possible to tell *soon* if this kind of behaviors are AI bugs or intended AI cheats (and change it if the latter)

On a side note, I do think that asymmetric gameplay is not a bad idea for civ-like games, but it has to be well integrated.

 

[fortydayweekend]

From the graphs it looks like Scientist AI can run Collective Minds in Ancient for a couple hundred science/turn, without any loss of Industry (their money goes to zero). That could let them get Standing Army potentially by T25? It'd be obvious from the graphs if that was the case - a massive spike in science and zero money.

 

[Siptah]

From the graphs it looks like Scientist AI can run Collective Minds in Ancient for a couple hundred science/turn, without any loss of Industry (their money goes to zero). That could let them get Standing Army potentially by T25? It'd be obvious from the graphs if that was the case - a massive spike in science and zero money.

I‘m not entirely sure when the AI bonuses are applied - it may be that collective minds doesn’t take some of these into account because the conversion is applied before the flat bonuses per worker for example. I think I remember seeing some graphs where the AI obviously had land raiser or collective minds going while other resources did not diminish as much as I thought they would.