AI city distance?

[SamBC]

When I have time, I'll muck around with these as well, and try to decide which to include in my mod... I can see why reversing the value of the rings would have the desired effect, but it might reduce the intelligence of AI city locations in ways other than spacing.

What sort of spacing do you get with this?

 

[blitzkrieg80]

OK, I have updated my last post with some screenshots... and some ones to compare below
____

I have discovered that these variables do not seem to be necessary for the successful ICS-spacing I have been observing (thus I am thinking of not changing them in the future, in case they have unforseen effects that takes away from the original developers' intention)

GlobalDefines.xml
SETTLER_DISTANCE_DROPOFF_MODIFIER

GlobalAIDefines.xml
AI_STRATEGY_AREA_IS_FULL_PERCENT
AI_STRATEGY_MINIMUM_SETTLE_FERTILITY

*The default (post-Dec patch, no tampering) values in the files seems to work... with the CITY_RING_MULTIPLIER changes in my previous post and the '-1 million' ALREADY_OWNED_STRATEGIC_VALUE, possibly being the critical elements... I tried SETTLER_EVALUATION_DISTANCE=100 and the ICS was really nice / more prominent at 4 spaces, but there were too many disconnected / far-flung cities...
___

screenshots of 150 turns on Autoplay... Pangaea map with 8 City-States (all without changing the default variables I just mentioned)

 

[alpaca]

These screens look pretty much like the distance 4 ICS grid I get without changing the ring modifiers

 

[SamBC]

Yeah, in the sense used by the defined minimum, those are distance 3 (it measures the number of empty hexes in between, rather than the 'distance' per se.

I find having the evaluation distance high helps with getting past 'barriers' of low-quality land or city-states or so on, but it doesn't want to be too high. After all, some far-flung cities is a sensible thing.

 

[blitzkrieg80]

yeah, it's true... still not much changed from the MIN RANGE

it seems like there's not much can be done without drastic overreaction by the AI...

BUT, I find that the behavior with the rings is that they try to place themselves more often next to resources, rather than JUST the 3 minimum range... where if it was just the 3 min range, it seems like they go anywhere, regardless of crap-tiles.... it probably has more to do with the ALREADY OWNED value, but it is a nice find you guys made

 

[blitzkrieg80]

OK... a little wrong on the SETTLER EVALUATION DISTANCE - at high values it seems pretty nifty, but there is definitely a fine balance... did some more experimenting and I've noticed some interesting 2-part (split with another civ between) empires... with SETTLER_EVALUATION_DISTANCE @ 80.

seemed pretty reasonable.

anybody have more information on the DROPOFF modifier?
______

VERY INTERESTING, imo (still the 3/4 space ICS as others have accomplished... but much better than default post-Dec patch ICS)... screenshots of 100 turns on Autoplay... Pangaea map with 8 City-States

using these variables:

Spoiler :

UPDATE Defines SET Value=2 (at default - no brute force this time)
WHERE Name="MIN_CITY_RANGE";

UPDATE Defines SET Value=-1000000
WHERE Name="ALREADY_OWNED_STRATEGIC_VALUE";

UPDATE Defines SET Value=2
WHERE Name="CITY_RING_1_MULTIPLIER";

UPDATE Defines SET Value=6
WHERE Name="CITY_RING_2_MULTIPLIER";

UPDATE Defines SET Value=12
WHERE Name="CITY_RING_3_MULTIPLIER";

UPDATE Defines SET Value=75
WHERE Name="SETTLER_BUILD_ON_COAST_PERCENT";

UPDATE Defines SET Value=[FONT="Arial Black"]60[/FONT]
WHERE Name="SETTLER_EVALUATION_DISTANCE";

UPDATE Defines SET Value=8
WHERE Name="SETTLER_PRODUCTION_MULTIPLIER";

 

[DragonBlood87]

On the center-right of the mini-map, is that a lone Persian city separated from the rest?

 

[blitzkrieg80]

good question I happened to notice that myself, in case it was worth including when taking the screenshots...

but no, that is actually a City-State with a similar color (Kutaisi, btw... my European Mod includes Colchis/Georgians )

 

[blitzkrieg80]

ok, won't spam too many more screenshots, but thought this last test was interesting / worth sharing...

same variables as described last: vanilla Drop Off and Min_Range... 60 Distance, flipped Ring emphasis... 100 turns Autoplay

*Notice the Norse have 4 cities: 2 split into 2 groups and pretty early in the game!

 

[Sneaks]

Are the values you listed with Min City Distance still at 3?

 

[blitzkrieg80]

no, for the last 2 groups of screenshots.... MIN_CITY_RANGE = 2

 

[BomberEscort]

Hard part is there's no way to know what they used to be set to...

I have a copy of all XMLs for every patch. Let me know what you need... or PM me your email and I'll send them to you.

 

[Sneaks]

Bomber is pretty cool!

 

[BomberEscort]

Bomber is pretty cool!

You can be president of my fan club!

 

[Tatran]

In stead of using city distance, I've tried to use terrain types for settling.
So, founding cities on forest, jungle, marsh and oasis was not possible.
Everything worked as intended, but after the first test one huge problem occured.
Several city states were conquered very early in the game.
This happened because every city state starts as a settler on a fixed location.

Spoiler :