Purchase land

[borg83]

Hi

Why doesn't s.th. like AITile.PurchaseLand exist? The Wiki suggestions page simply tells "won't happen" without any reason.

Placing s.th. like depots on a tile where you would like to expand really looks ugly. But it seems to be the only way to prevent towns and other players from building on tiles where you would like to replace a small airport with a larger one (for example).

greetz, BørG

[CommanderZ]

Buying land by players is limited by the amount of clicking they are willing to do. Ai's wouldn't. Ai with its superior computing and clicking power could in mere seconds buy a thick circle of land around all industries - prohibiting other players from doing anything.

[borg83]

That's the reason for it

Okay, purchase land costs some money, but money is not a problem in this game

If I wanted to block other players I could do so by buying s.th. different than these landmarks. Actually, I am doing this with the AI I am playing with, but
a) it looks ugly and
b) the AI needs to find out wether the object is meant as a placeholder or is really needed.

[Yexo]

That's the reason for it

Yes, that's the reason for it. Actually when discussing the function we said.
1. The purchase function is flawed, because it's way too cheap to buy land.
2. It doesn't cost any upkeep, so nothing prevents you from buying the complete map.
3. Best option would be to remove the buy land option completely (also for human players) but that won't happen, so until 1) and 2) are solved, AIs won't be allowed to buy land.

So please come with suggestions so solve 1) and 2) if you want your AI to buy land.

Please note that this was discussed buy some people working on NoAI, not the complete dev team. Othere devs may have other opinions on it.

[borg83]

I see...

But now I am wondering why you do not give AIs this option even more. If it is s.th. you don't like in general, but do not have a suitable substitute for it (yet), so why are humans still allowed to use it but AIs are not?

When playing on an internet server and someone uses this "tactic" of massively buying land I am of course frustrated. But I had no chance to know he/she would use this tactic in advance. If, however, an AI would use this "tactic", it would always use it and I simply wouldn't play against this AI.

Anyway, I won't complain as long as I am just playing around with NoAI. I have found a solution that works for me, even though it is kind of a hack.

PS: In fact, I am planning to be able to ask my AI to remove some specific rail/road/whatever part by placing a special sign on a tile. But that's another topic

[Zuu]

For upgrading airports, I've figured that it would probably be just simplier for PAXLink (my most advanced AI) to build a new airport. Then the orders-code will pick the best airport. And later, the station code will remove all unserviced airports, if there is a better airport in the same town. This way there won't be any problem of what to do with the planes during the upgrade. A strive I have is to have simple processes that can be recovered if they are performed only half way.


But I should also say that in last PAXLink, I do reserve land using rail tracks, but that is limited to a row of land between airports and the access-road to the road stations that are planed but not yet allowed by some town. An issue is that the road pathfinder in some cases find it a good idea to use this space.

But if we now are talking about property maintenance, then it is quite silly that a bus stop cost the same as an airport annually.

[borg83]

This might work for PAXLink, but not for my AI

I am always trying to have many houses close to an airport (I, as a human and my AI). But the largest airport are only available later in the game, when cities already have grown a lot. If I already have a large airport I do only need to clear two 1x7 tiles rows to replace it with an international airport. Still this can be enough for the local authority to refuse to remove some houses or to build the better airport, so I/the AI needs to pause a while.

The more I think about it I come to the conclusion that I/my AI should use a PAXLink-like transfer, too ^^

A strive I have is to have simple processes that can be recovered if they are performed only half way.

A problem that probably every AI coder has had yet

But if we now are talking about property maintenance, then it is quite silly that a bus stop cost the same as an airport annually.

LOL?

[Zuu]

The more I think about it I come to the conclusion that I/my AI should use a PAXLink-like transfer, too ^^

It depends on why you do it. If you do it to not have to upgrade airports, then you don't reduce your complexity by having to manage the buses too. For me the feeder network is a core feature of the AI and thus not seen as an overhead. But if you just see it as an option compared to taking more care when upgrading airports then I don't think you gain that much by using feeders. Unless you like the reality/elegance aspect of using feeders.

A problem I realized is that towns have a limited number of airports allowed, and there can be cases when one need to remove the old airport in order to be allowed to build a new one. So the same care about what to do with the aircrafts still apply.

[Moriarty]

Observation - what's to stop the AI from "buying land" using something other than the buy land function? I.e. by placing some railroad there.
That's easily just as abuseable as the buying-land function (I used to use that method to square out my city airports in TTD given it didn't have a buy-land function).