Changes to restrictinfra

[panch93]

This post is based on discussions with louis, corbeau, wieder and jwrober. I am just putting it all together here.

Why do we want to change it:
The whole point of restrict infra is to avoid massive invasions where you can lose entire empires in a single turn. There have been quite a bit of realism/gameplay discussions around this feature, but it still doesnt exactly solve the problem it is supposed to solve. Any talented player can quite easily figure out a way to attack quite deep using the fact that borders change immediately after capturing the city.

A couple of ideas have been proposed regarding this.
1) Borders change only at TC
2) There will be n turns of restrict infra on newly conquered cities.
3) *The Winning idea IMO* Generalize restrict infra to be based on units too along with borders.
4) Reduce working area of city for n turns after conquering

I will explain 3 in more detail from this point on.
Realistically, it is not the borders which make units move slower in enemy terrain but the presence of enemy units. So, we dont just let borders decide if there is restrict infra or not but the presence of enemy units too. Here is now it works, each unit has a strength for restrict infra. Say, immediate circle 0.5, one ring higher 0.25 etc.. and borders also have a strength (say, 0.5). In each tile, all the strengths are added up and then if the total strength is > 1, restrictinfra is activated.

In this model, our current behavior can be recreated if we set border strength to 1 and units to zero.

One of the issues is info leak from units inside cities preventing bluffing. We can probably get around that by giving occupied cities a fixed strength irrespective of units inside.

This can also probably added to effects so that we can have buildings/wonders/tech which change these factors.

[Corbeau]

An additional idea regarding info leak.

Disable GoTo for tiles where there is ANY enemy influence, dominant or not. So there you have to move tile by tile, "by hand", using keys, without knowing the effect (MPs spent) until you move there. In real life, this is called reconnaissance and should be put to use in Longturn.

Yes, that way you also get some info about enemy nearby, but at a cost, and need to plan for it.

Also, I'd give borders without units the lowest strength factor, not 0.5.

[Hans_Lemurson]

I think approach #1 would be the simplest, and would do a lot to prevent deep penetration.

#2 seems a bit harsh, and would be weird to have restrict-infra be applied within your own borders. Enemy borders at least give a visual indication of where you can freely move.

Option #3 looks pretty complicated to me. We already have the "zone of control" preventing movement of units past other units. But this would instead have the presence of units invisibly make roads non-useable? I get that it's a way to dynamically generate the restrict-infra rule based on the local conditions that produce it, but I don't like how it works invisibly. (Also, it would do nothing to solve the "Lose an empire in a single turn" phenomenon, since you'd be able to travel freely through your conquests since all the enemy troops are probably dead.)

My Ideas:
Slow movement through enemy territory has historically been due to having to proceed cautiously due to risk of enemy ambush and from sabotage of infrastructure by enemy forces. However, the way that the game works, small harassing forces that slow an enemy's advance are unfeasible. Solitary units are vulnerable to bribery, and larger groups would just leave the city's defenses weakened. A unit behind walls is worth three in the field.

Restrict-infra can thus be thought of as a way of abstracting this harassment (and other logistical difficulties armies suffer in hostile land) which can't be adequately modelled by the game's existing combat system. I don't have any real gripes with it (other than maybe it penalizes movement over hills and forests too much). The main problem seems to be that it doesn't work well enough.

The problem as I see it is not with Restrict-Infra but rather with Borders themselves. This is why I'm in favor of option #1. Borders, in my opinion, are the best way to represent the territory that you have "control" over. Not just the obvious fact of "that's literally what they represent", but rather that they are visual! Clear lines on the map that clearly say what land you have freedom and control in, and what land is hostile to you.

The problem with being able to penetrate much-too-deeply into enemy territory is that the borders themselves are failing. This is because in the existing system, control of land is based almost completely on the age of the city. I have participated in invasion-plans that meticulously mapped out cities based on their founding dates (seen by city ID) to chart a course with "favorable borders" where the conquered cities would gain you land right up to the walls of the next city on your list. This largely bypassed the infrastructure restrictions that the rule was supposed to create, because the borders were behaving badly.

If borders were only changed at the end of the turn (easy), or could be altered by some reasonable dynamic process during a turn (harder, but more interesting), then I think the system would largely work as intended.

But how should borders work? The current system of "land is owned by the oldest city" would still be broken, even if we go with Option #1 and delay border recalculation until the TC.

[Corbeau]

Borders, in my opinion, are the best way to represent the territory that you have "control" over. Not just the obvious fact of "that's literally what they represent",

Actually, this is very wrong. You can have a border 3 tiles away from the city, the city can be under siege with space between enemy and the border completely under enemy control, and the enemy would get restricted movement, not you.

Borders are an agreed construct. They lie where the two parties agree in peace. In war, they are meaningless.

but rather that they are visual! Clear lines on the map that clearly say what land you have freedom and control in, and what land is hostile to you.

This is also very problematic. The only way to know if you control an area is having someone there. Hell, the only way to CONTROL an area is to have someone there. Nobody there - no control. You can say that Civ is an approximation and that borders actually represent something. I'd say that they SHOULD represent something, but they do it very poorly under current mechanic.

[Hans_Lemurson]

Borders, in my opinion, are the best way to represent the territory that you have "control" over. Not just the obvious fact of "that's literally what they represent",

Actually, this is very wrong. You can have a border 3 tiles away from the city, the city can be under siege with space between enemy and the border completely under enemy control, and the enemy would get restricted movement, not you.

Borders are an agreed construct. They lie where the two parties agree in peace. In war, they are meaningless.

but rather that they are visual! Clear lines on the map that clearly say what land you have freedom and control in, and what land is hostile to you.

This is also very problematic. The only way to know if you control an area is having someone there. Hell, the only way to CONTROL an area is to have someone there. Nobody there - no control. You can say that Civ is an approximation and that borders actually represent something. I'd say that they SHOULD represent something, but they do it very poorly under current mechanic.

You're just repeating the point we all agree on: Borders as they currently exist are a poor way of representing things.

But "Automatically demarcated areas of the map that represent the land that a nation has control over" is a very useful concept for deciding freedom of movement and also access to resources.

"Drawing borders on the map" is an existing mechanic that already works with the restrict-infra rule. So if we can simply change HOW the borders are drawn, then borders WILL be useful.

So now the question is, as it was before: How should borders work?
I think some system that dynamically calculates it based on the presence of troops in nearby tiles or cities would be a good idea. Of course, the devil is in the details. Let's discuss some details! Anybody have any ideas of how they'd like to see it work?

[Corbeau]

The whole point I'm trying to make: in reality, "borders" and "control" are two completely different things. "Borders cause control" only during peacetime because all parties agree on it, while in war borders effectively don't exist.

I don't think a "clear depiction of control in the form of borders" is something we should aim for. Official borders are a consequence of agreements. Control is a matter of the situation in the field. If you want to do a war right, there should be fuzzyness and unclarity (if that's a word).

[Hans_Lemurson]

I think that any system that has effects on the map should be visible on the map. Borders are an existing mechanic that draws out regions of control and defines one tile as being controlled by one nation or another. Since the slowness of militaries in enemy territory is due to some notion of the enemy still having "control" over that area, then it seems like it should apply just fine. Visible zones of military dominance = good in my book.

What we need is a system where we can decide which nation "owns" what tiles based on the presence of troops.