Plans for V28

[Koshling]

Koshling, would culture be able to spread through a mountain pass before mountaineering?

What precisely do you mean by 'a mountain pass'. My view is that (in terms of the Civ map) that would be a non-mountain hill between mountain tiles, in which case the answer would be 'yes', but if you mean something else...

 

[Taxman66]

There are occasions where you can move between mountatins to get to a square on the otherside of the range. Usually this occures on diagnol moves.

 

[Koshling]

There are occasions where you can move between mountatins to get to a square on the otherside of the range. Usually this occures on diagnol moves.

That will be fine - diagonal spread will work for culture (it will slow it a little but not much). You'll also be able to actively encourage it by building a route from your city through the pass.

 

[Dancing Hoskuld]

Ok, I have a concrete approach to rewriting realistic culture spread in mind, which I'll do once V27 is out. The characteristics of it will be:

• Every spread point WILL cause at least one tile spread
• Spread rate will be somewhat tech dependent. In particular it will not be possible to spread THROUGH (to will be possible) mountains until you have mountaineering, or THROUGH (but again to is possible) ocean until Navigation. Once you get those techs it will tend to release very significant culture flows in 'catch-up' areas, so these will have extra strategic benefit
• At each spread event a pool of points will be allocated proportional to the square of the spread level. Tiles adjacent to the current spread borders will be claimed in lowest cost order, until the spread pool is exhausted, with the first tile always taken regardless of whether that is over the points pool size or not
• The 'cost' of spreading to a tile will be based on:
  • Its distance from the city
  • Its terrain type (as with current RCS)
  • Whether it involves crossing a river (as with current RCS)
  • Whether it contains a revealed bonus (that will significantly reduce the cost to reflect the effort a culture would make to 'capture' a bonus it knows about). This reduction won't be so much that your stone-on-hill-across-a-river will b preferred to an immediately available grassland with no river, but won't be far behind (I'll also make the reduction percentage a tunable global define so we can fiddle easily)

The effect should be guaranteed spreading (to at least some degree), following terrain channels (the 'realistic' bit), a degree of 'reaching out' for attractive tiles (revealed bonuses), and the ability to spread further more quickly if channeled into narrow regions (i.e. - some compensation for that mountain range near you, by squishing the spread wider in the other directions)

It is not. This may be worthwhile to add, but I'm not sure it's really acting as much of an impediment since, in most cases, the buildings are quite attractive anyway so it anyway builds them most everywhere. If you are aware of specific examples where the AI seems to definitely be suffering from this let me know (with save game) and I'll look into adding some smarts for it.


No. There is a whole other thread about this somewhere, which was pretty active a couple of weeks back. I agree that it does need tackling, and should really be something of a V28 priority

Again my suggestions and concerns on this have been ignored. Maybe I should give up and get my morning coffee. To me the biggest problem is in the prehistoric era. This mechanism is fine for an agricultural nation where it makes sense that cleared land that you can farm is acquired before any others but before agriculture it just means that you acquire useless plots when more useful plots are available eg that forested hill with obsidian.

1) Is the AI aware that it need to build Ex 4 Tabletmakers to be able to build the Library-Wonder? So that it decides: "ok I really want this wonder, so I need 4 more Tabletmakers... Those 4 cities would be the fastes to build them, so I will just do this!"

The idea that a wonder requires x buildings before you can build it is used extensively in vanilla civ so I would have thought that the AI would know how to use it. The Forbidden Palace requires 10 courthouses is the one that comes to mind.

 

[ls612]

My first steps in V28 will be to make a tag for orbital buildings (think SMAC), and set up Cyberwarfare. I'll detail those two things more fully in due time, but I think I can do it (cyberwarfare anyways, the orbital tag may end up requiring some complicated SDK).

 

[Koshling]

Again my suggestions and concerns on this have been ignored. Maybe I should give up and get my morning coffee. To me the biggest problem is in the prehistoric era. This mechanism is fine for an agricultural nation where it makes sense that cleared land that you can farm is acquired before any others but before agriculture it just means that you acquire useless plots when more useful plots are available eg that forested hill with obsidian.

Did you actually read it? Two things both make it pretty easy to get obsidian/stone on that hill:

• Revealed resources will significantly reduce the cost of a tile
• Routes will reduce costs

So once you reveal the resource the hill will cost about the same as (and therefore happen as soon as) grassland (tuning via global define can vary that, but that's about my intent). If you build routes to the tiles you want that will also cause culture to flow there faster.

How is this ignoring the concern?

 

[LumenAngel]

But in prehistoric era, a tile with forest is often better than a tile without forest...
Feature are used for this calcul too? because a tile with an oasis is really good, but one with some nuclear pollution (or worst : indigenous ppl), not really...

 

[Koshling]

But in prehistoric era, a tile with forest is often better than a tile without forest...
Feature are used for this calcul too? because a tile with an oasis is really good, but one with some nuclear pollution (or worst : indigenous ppl), not really...

I have no intention of making 'realistic' mean 'maximum output' or 'optimal'. It's easier for culture to spread where habitation can/does, so I'm only planning on biasing it towards places that are harder to reach for extra-ordinary attraction (aka revealed resources). However, you have the option of building routes if you want to encourage certain spreads (to forests for example, though forests will be spread into fairly easily anyway).

 

[wolfensoul9]

I have no intention of making 'realistic' mean 'maximum output' or 'optimal'. It's easier for culture to spread where habitation can/does, so I'm only planning on biasing it towards places that are harder to reach for extra-ordinary attraction (aka revealed resources).

These two sentences seem to be contradictory to me.

Besides, the fact is, maximum output IS realistic. Cities will naturally expand towards land that will benefit them. They're not stupid.

Making an exception for resources, but not for other high yield tiles is arbitrary.

 

[LumenAngel]

Yes, in a desert, I really think everyone will rush for this oasis ^^
A problem in my actual game was with tar pit : it take lot of turn to have it in my culture border, bu I was really in need for it

 

[Koshling]

These two sentences seem to be contradictory to me.

Besides, the fact is, maximum output IS realistic. Cities will naturally expand towards land that will benefit them. They're not stupid.

Making an exception for resources, but not for other high yield tiles is arbitrary.

Fair enough, but 'high yield' is subjective. If you mean 'total of hammer + food + commerce count' then I can agree (in which case grassland = forest). This would indeed cause the oasis to get a boost. I can go with that, but I have trouble with more subjective measures (hammers > commerce say) which are implied by something that would boost a forest over grasslands.

How about if I use total unimproved tile output (as I defined it above) as a cost reducer (revealed bonuses would still be a further reducer - I especially want to keep them special to address the DH point of the canonical stone-on-hill issue). That way you get things like oasis naturally and can still 'steer' with routes if you really want to promote (say) forests...?

 

[JosEPh_II]

Can you "steer" RCS with trails and mudpaths now in any way?

JosEPh

 

[Koshling]

Can you "steer" RCS with trails and mudpaths now in any way?

JosEPh

I don't believe so, but I'm not 100% sure. If you can it has little effect on the current system based on my experience.

 

[JosEPh_II]

In the game I'm testing RCS in, I had a tile become part of my Culture after I won a combat on it. Might have been just coincidence though. Because I don't have FB On in that game.

Trying test games with a lot these Options I don't normally use. And also trying to be more aggressive. I'm a defensive builder type usually. Although this mod has made me be more aggressive than I ever used to be.

JosEPh

 

[Koshling]

In the game I'm testing RCS in, I had a tile become part of my Culture after I won a combat on it. Might have been just coincidence though. Because I don't have FB On in that game.

Trying test games with a lot these Options I don't normally use. And also trying to be more aggressive. I'm a defensive builder type usually. Although this mod has made me be more aggressive than I ever used to be.

JosEPh

That's not RCS, that's culturally driven war (or whatever it's called). It happens with out RCS too, and is a totally independent mechanism

 

[JosEPh_II]

IDW yeah, hard to keep up with all these options anymore.

JosEPh

 

[wolfensoul9]

Fair enough, but 'high yield' is subjective. If you mean 'total of hammer + food + commerce count' then I can agree (in which case grassland = forest). This would indeed cause the oasis to get a boost. I can go with that, but I have trouble with more subjective measures (hammers > commerce say) which are implied by something that would boost a forest over grasslands.

How about if I use total unimproved tile output (as I defined it above) as a cost reducer (revealed bonuses would still be a further reducer - I especially want to keep them special to address the DH point of the canonical stone-on-hill issue). That way you get things like oasis naturally and can still 'steer' with routes if you really want to promote (say) forests...?

Sounds great with the total unimproved tile output focus. Though, forests naturally have a higher yield than grasslands...two food one hammer vs. two food.

 

[Koshling]

Sounds great with the total unimproved tile output focus. Though, forests naturally have a higher yield than grasslands...two food one hammer vs. two food.

Ah yes, I was thinking grassland had a commerce, but that's only next to a river isn't it.

 

[Necratoid]

One thing I haven't seen mentioned is that there should be a faster spread in the tiles around the ones your actively using. Working a tile means people are actively in the area and thus will be more likely enter the surrounding tiles. When we are talking the prehistoric we should also add in bonus spread weight to tiles that have a 'sacrifice gatherer to build' improvements... you know as people are by default hanging out on the tile.

 

[Thunderbrd]

One thing I haven't seen mentioned is that there should be a faster spread in the tiles around the ones your actively using. Working a tile means people are actively in the area and thus will be more likely enter the surrounding tiles. When we are talking the prehistoric we should also add in bonus spread weight to tiles that have a 'sacrifice gatherer to build' improvements... you know as people are by default hanging out on the tile.

Good idea