Anyone know how it chooses what tile to expand to?

[Ivan Hunger]

I definitely think it factors district adjacency bonuses into its decision, especially districts of which you haven't built many yet.

Using the original post as an example, I'm guessing the game has noticed that you haven't built very many holy sites in your empire, so it's grabbing that bananas tile first to encourage you to put one there. It's adjacent to two natural wonder tiles, so it's a good spot for one.

 

[Hactar]

Why crabs before pearls though?

The important question... if there were both Truffles and Pearls, would it choose pearls before swine?

 

[nzcamel]

The important question... if there were both Truffles and Pearls, would it choose pearls before swine?

Smartass

 

[Trengilly]

it seems there is a random element decision with weighting.

Well that explains why it never picks the tile I want it to!

 

[craney1987]

Would it be too much micromanagement for us to be pick ourselves? A culture civ with this ability might be a nice perk, if it isn't too tedious

 

[ezysquire]

Nice looking weather for Leeds by the looks of it. As for the tile accrual, it'd be nice if the random generation was linked to the city governor priorities so you could give it a bit of a steer.

 

[cvb]

1) 2nd ring tiles available? After them: 3rd ring
2) special resource available? (luxury/strategic > bonus?)
3) number of yield icons, if tie: higher weight on science, culture & faith than food & production (gold?)
4) still tie: current priority towards different yields (set manually by human or internally by "governor", this last one leads to flickering predictions)

I like the way you get quickly the best 2nd ring tiles, albeit in a 'suspicious' order ... ... and can immediately buy important 3rd ring tiles, because waiting for them would take 'forever'.

 

[Uberfrog]

1) 2nd ring tiles available? After them: 3rd ring
2) special resource available? (luxury/strategic > bonus?)
3) number of yield icons, if tie: higher weight on science, culture & faith than food & production (gold?)
4) still tie: current priority towards different yields (set manually by human or internally by "governor", this last one leads to flickering predictions)

I like the way you get quickly the best 2nd ring tiles, albeit in a 'suspicious' order ... ... and can immediately buy important 3rd ring tiles, because waiting for them would take 'forever'.

If this is the logic, I'd say that it seems reasonable. Tile selection should definitely be automatic. Gold purchasing tiles to get them sooner is an important mechanic, and beelining resources in the third ring should definitely cost you.

Sure, going for the bananas before Ha Long Bay is clearly not optimal, but in other situations going for resources sooner would be preferred. It's easier and more understandable to have a hard and fast rule than a more involved calculation that works out your actual benefit. You've got to leave something for the player to do...

 

[Leyrann]

I'm probably late because I don't feel like reading through all the thread and someone probably mentioned it already, but I'm doing it anyways.

First off, most importantly, it always fills the second ring before starting on the third ring.
Then, it prioritizes resources. I don't know wheter there is any prioritization among them.
Then, I'm not sure what it prioritizes, but it may be yields or it may be open land.
Then, it prioritizes mountains.
Then, finally (I've mostly been playing Indonesia recently...) it goes for water tiles before moving on to the next ring.

 

[cvb]

If this is the logic, I'd say that it seems reasonable. Tile selection should definitely be automatic. Gold purchasing tiles to get them sooner is an important mechanic, and beelining resources in the third ring should definitely cost you.

I'm sure some small details are still missing or wrong. Anyway it is very difficult to say what the best tiles are - depends a lot on individual playing style.

On this occasion I like to point at TheGameMechanic's recent Let's Play Civilization 6 - Victoria part 1 . The gameplay starts @ 2:33!!!

London @ 7:38 shows the question of the thread (nearly) in a traditional fat-X style: gets one first Salt, Gypsum, Crabs or Rice ... 1) 10:57 2) 21:23 ...
... overlaid by the big question how to grow the initial population really FAST.
(I'm a fan of growing the 1st city at maximum speed until unhappiness begins to stir. This is a bit tricky on this interesting map though. I.e. "immediately" buy the 2nd ring Rice tile and start with scout & Builder??!)

 

[Forster]

I may be wrong, but it seems that when I purchase the tile that is slotted for expansion, say in 4 turn, the city does not expand into another tile in 4 turns. There seems to be some delay involved. Has anyone else seen this?

 

[Victoria]

On this occasion I like to point at TheGameMechanic's recent Let's Play Civilization 6 - Victoria part 1 .

Lol, interesting game, brilliant starting position, certainly not playing the way I would play

 

[Mahi]

It needs to be a bit random. Wouldn't make much of a challenge if the city just picks the best tiles for you. That's why you can buy tiles. One thing that I really like with Civ 5 and 6 is the asymmetric border expansions. For me it makes more sense you can't control cultural progression, other than try to enhance it to speed it all up. That said, it also makes a bit sense that a population will seek to expand its cultural border with best output. But seeing gold is so easy to acquire it make sense you need to use some of it to get the best tiles faster. Would make even more sense, if a district pushed harder one the nearest border tiles.

 

[shaglio]

It needs to be a bit random. Wouldn't make much of a challenge if the city just picks the best tiles for you. That's why you can buy tiles. One thing that I really like with Civ 5 and 6 is the asymmetric border expansions. For me it makes more sense you can't control cultural progression, other than try to enhance it to speed it all up. That said, it also makes a bit sense that a population will seek to expand its cultural border with best output. But seeing gold is so easy to acquire it make sense you need to use some of it to get the best tiles faster. Would make even more sense, if a district pushed harder one the nearest border tiles.

On release, Civ 5's border expansion tended to stop at mountains and rivers to simulate realistic borders. But then the players got pissed when there was a Luxury on the other side that the borders refused to expand to. But early game Gold was a bit more scarce in Civ 5 than it is in Civ 6.

 

[WillowBrook]

I may be wrong, but it seems that when I purchase the tile that is slotted for expansion, say in 4 turn, the city does not expand into another tile in 4 turns. There seems to be some delay involved. Has anyone else seen this?

I have not seen that, and I can't see why that should be, unless you also do something to slow the culture accumulation of the city (like stop working a resource that gives culture). Have you tried to get a screen shot before purchasing, and then another after with a different number of turns?

 

[UWHabs]

It needs to be a bit random. Wouldn't make much of a challenge if the city just picks the best tiles for you. That's why you can buy tiles. One thing that I really like with Civ 5 and 6 is the asymmetric border expansions. For me it makes more sense you can't control cultural progression, other than try to enhance it to speed it all up. That said, it also makes a bit sense that a population will seek to expand its cultural border with best output. But seeing gold is so easy to acquire it make sense you need to use some of it to get the best tiles faster. Would make even more sense, if a district pushed harder one the nearest border tiles.

Yeah, I like the random asymmetric feel to it. The weird not wanting to creep across a river, filling up the whole east side of a city before going west, etc... I agree as well that it should try to expand next to district tiles when possible as well. And it totally does make sense to not stretch past a mountain unless if you buy tiles yourself. That's why we have the ability to buy tiles.

 

[Xeribulos]

I may be wrong, but it seems that when I purchase the tile that is slotted for expansion, say in 4 turn, the city does not expand into another tile in 4 turns. There seems to be some delay involved. Has anyone else seen this?

Each tile costs a bit more culture than the one before. If you buy a tile, the culture needed to expand to another one should always go up. I think this is definitely intentional.

 

[Victoria]

Each tile costs a bit more culture than the one before. If you buy a tile, the culture needed to expand to another one should always go up. I think this is definitely intentional.

I did not think the price increased for purchased tiles.

 

[nzcamel]

I may be wrong, but it seems that when I purchase the tile that is slotted for expansion, say in 4 turn, the city does not expand into another tile in 4 turns. There seems to be some delay involved. Has anyone else seen this?

I have not seen that, and I can't see why that should be, unless you also do something to slow the culture accumulation of the city (like stop working a resource that gives culture). Have you tried to get a screen shot before purchasing, and then another after with a different number of turns?

Each tile costs a bit more culture than the one before. If you buy a tile, the culture needed to expand to another one should always go up. I think this is definitely intentional.

I have noticed it, and I made the assumption (silly me) that if you purchased a tile that was highlighted for next expansion, that you lost that build up of culture, and it had to start again. Now that you put it that way, I must be wrong, and it must still give you another tile, but as Xeribulos says, the culture cost for that next tile is likely higher than for the last, so it takes a little longer. I'll try to pay closer attention.

 

[Xeribulos]

I did not think the price increased for purchased tiles.

You may be right. But it would explain these obervations. I just assumed they wouldn't bother to make a difference between tiles gained through culture and tiles gained through gold.
However, Russia's unique ability gives cities a bunch of tiles without increasing the cost for future tiles, so there is that.