Yup, and marshes are fairly rare, which is why I never suggested such an idea before. I just brought it up because you asked about marshes.
The Netherlands
- Thread starter Funak
- Start date
Yup, and marshes are fairly rare, which is why I never suggested such an idea before. I just brought it up because you asked about marshes.
PurpleMentat
Videographer
I think the whole thing is academic anyway. Gazebo said adding features is not supported in the DLL.
The big G says a lot of things, and I don't understand half of it so let's keep going
Llednar Twem
Prince
- Joined
- Aug 13, 2014
- Messages
- 420
Well, if reforestation is already a thing that exists, i feel like it would be entirely possible to use that as a basis for implementing this. If it has to be done through a worker action then have it be done through the "build polder" command rather than having the polder change the underlying tile when finished. That seems like it should work anyway...maybe.
Ehm, we're not talking about the same thing anymore are we? By that I mean having the pre-req for building polders being fresh-water or a marsh.
Llednar Twem
Prince
- Joined
- Aug 13, 2014
- Messages
- 420
Oh, no, I was talking about the ability to make grassland into marsh and desert into flood plains.
This bit here. It's a rather important aspect of the idea.
Adding features to the map is a no-go in terms of AI evaluation and efficiency. Mods like the reforestation mod are aesthetically pleasing and interesting, but the AI has no idea what it is doing. Plus, features are largely 'static' entities in the DLL - making them dynamic and addable (from an AI perspective) adds a heavy layer of thought-processing (and thus slowdown).
Plot elements are the single most expensive DLL element aside from pathfinding (combined, they make up over 75% of civ's CPU need). Less is more, and our memory cap is very close.
G
Okay, but just having them buildable on any tile with freshwater would be fine? With that in mind the yields could be nerfed and an adjacency bonus would be added. That would solve all problems mentioned, right?
What's wrong with polders as they are now, again? As far as I can tell, most Dutch players I see are able to get a few per city on average.
G
Well, I can't get a break with mapgeneration, so I can't build any polders at all, but that was not the point. The point was that someone brought it up and I gave a solution.
PurpleMentat
Videographer
The complaint, really, is that Farms are SO good. Breaking up a Farm adjacency chain to put a Polder on a Flood Plains can cost you a lot of food.
That could be said for any improvement, though. And farm spam is okay it looks far more 'natural' than trading post spam used to look (and let's be honest, we all play civ for the aesthetics of the map, right?).
PurpleMentat
Videographer
I agree. I also think that, if you are playing a civ with a Unique Improvement, it should never be a tough decision whether you build the UI or a farm for adjacency. You're giving up a UB to get that UI, and I haven't played a civ in the mod yet that doesn't want their UB in every city. The Polder is in an especially tough place, as it's competing with prime Farm territory.
Atlas627
Deity
- Joined
- Aug 25, 2011
- Messages
- 3,123
I think the Polder is fine by itself, but since the Chateau got buffed it overshadows the Polder AND the rest of France is straight-up better too.
Is there a way to detect where rivers meet the coast, and let you build Polders there no matter the terrain? Or perhaps give the Polder a defense bonus (or ZoC?) to represent the Dutch tendency to break the levies that made Polders to block out attackers?
Is there a way to detect where rivers meet the coast, and let you build Polders there no matter the terrain? Or perhaps give the Polder a defense bonus (or ZoC?) to represent the Dutch tendency to break the levies that made Polders to block out attackers?
Ah yes, the ol' Led Zepplin maneuver.
It's a seesaw of balance with UIs- formerly the chateau was widely considered weakest, now the polder, while the chateau only got a single-yield bump.
I really don't think the polder is in that bad of a spot. BUT I do see that a lot of the frustration around UIs is that they are either so strong that they're always better than improvements (and should be spammed), or they're 'just good enough' to be built, but 'not good enough' to compensate for lost yields, especially when related to improvements with adjacency bonuses.
So, what I'm going to do next is this I'm going to see about adding in a plot element (as minimally as possible) that allows improvements buff specific other improvements by a set yield.
So for example, we could make it so that Terrace Farms give adjacent farms +1 production or +1 food, or Polders could give adjacent farms +1 culture, etc.
I'm going to experiment with this before I confirm that it is an incoming feature. Cool? Cool.
G
Llednar Twem
Prince
- Joined
- Aug 13, 2014
- Messages
- 420
To clarify, my specific complaint about polders is that they counter-intuitively make your surrounding farmland worse despite being conceptually based around improved farming methods. Plus their terrain requirements mean they have to go right where you would probably want to put your farms.
This would be fantastic if it ends up working! Here's hoping it does.
PurpleMentat
Videographer
Sounds awesome, hope it works!
Success!
Before
After
I set it to give +1 culture to adjacent farms. Works. I'm going to keep fiddling with parameters to make it as efficient as possible, but this will work.
The function can accept any pair of improvements, and any terrain-based yield.
So, now you need to decide what bonuses to give the polder (and terrace farm).
G
Before
After
I set it to give +1 culture to adjacent farms. Works. I'm going to keep fiddling with parameters to make it as efficient as possible, but this will work.
The function can accept any pair of improvements, and any terrain-based yield.
So, now you need to decide what bonuses to give the polder (and terrace farm).
G
