Discussion: how should native raids cause damage?

When natives raid a colony, they have a chance to cause damage even if the raid failed. We use the vanilla code, which sometimes can seem quite unfair and unbalanced, like the reported failed raid stealing 6000 gold.

If we are to redo this to make it fair, the question is how? My take on it is that we follow the same pattern as with goody huts (ruins) in the sense that we code a bunch of possible outcomes and then the difficulty level has a list of possible outcomes. This will allow raids to be more nasty at high difficulty levels.

Another interesting idea would be to add defense to the raid results. Certain raids can only happen with 0% defense and some can only happen if the defense is max the default defense (as in no stockade). To make this even more interesting, one of the raid results could be damage to the defense like the colony being bombarded.

The question is, what should the possible raid outcomes be? What kind of rules should we apply to a raid results? What would be fair losses in percentage and amount? (preferably we cap at both). When should each type of damage be triggered relative to defense?

 

First things that come to mind: I think they should take goods as they are available: guns first, then blades, then gold, then horses, then tools. Maybe steal a number of random settlers. As for proportion, perhaps the settlement defense could reduce the losses by the defense percentage?

 

Not proud to admit, but in my last game I committed genocide in a rage after Tupi braves destroyed my university construction in a fort city defended by a high level Hessian. I tolerated perhaps 50 raids up that point, lost at least 20-30k gold, numerous constructions- all failed raids and yes, I also had that failed raid stole 6k gold, though mostly smaller amounts like 1-2k. Patriot difficulty.

I like the ideas on github like death of colonists, which could happen even if raid is unsuccessful with a minor chance and believe natives are much likely to attack outlying infrastructure like farms and plantations, rather than fortified and garrisoned cities. I believe these should be in event format, as if they could pillage while not at war that would be incredibly annoying. Something like "if plot has x native population and has a mine/farm/plantation and x is furious", mean time to happen scaling up with negative relation. Destruction of goods even with unsuccessful raids is also a good opinion, food granaries most likely targets.

 

True, but that was implemented to prevent exploiting Native Raids purposely to train your Units.

This is not Vanilla Code. "Native Raids" does not exist in Civ4Col Vanilla. (It does however in Original of 1994)
It was introduced by RaR - I wrote the code - partially adapting code from another modder though.

The max. amount of gold stolen depends on how much gold you have acquired.
There is of course some random factor in there as well.

Your concept is definitely more flexible than my old code.
But it is also more effort.

Also Native Raids are already affected by difficulty levels.
They are much less likely to even start Native Raids.
(At least that is how I remember it - but I might be wrong.)

The outcomes (seriousness of consequences) themselves however are probably not affected.
(Would need to check my old code again however to be sure of that.)

In early versions of RaR we realized that if Native Raids become too unlikely to succeed or do not have any consequences they become an exploit.
Native Raids themselves can also be easily prevented by having good relations with Natives (Missioning, Trading and not expanding too fast).

Native Raids in RaR were supposed to have chances for bad consequences if you expand too heavily or treat your Native neighbours badly.

RaR purposely wanted to add a bit more challenge for experienced players.
And yes we knew that this made the game have a steeper learning curve.

The game concept "Naive Raids" (like most new game concepts we added) is also explained in Colopedia by the way.
Some new players have simply not adapted their game style to it or do not even yet know all the new game concepts.

Of course myself and all the old RaR team members have their own personal taste for gameplay.
Thus I of course coded features a specific way and I am open to any changes to them.

I am a bit careful about generally changing features just to make the game easier or more beginner friendly.
Adjusting features to difficulty settings where needed is of course fine.

------

Summary:

Yes, RaR and now its successor WTP play very differently compared to Vanilla or TAC.
But that is what the old team wanted. We wanted to make it more challenging and complex.

There should be the "beginner version": TAC - which already had great quality
But there should also be the "experienced player version": RaR (now WTP) - which added some more challenge to TAC

If you don't know the concepts and design decisions (e.g. not reading Forum or Colopedia) and thus don't adapt your play style you can easily get your but kicked in higher difficulties by e.g. Natives if you are a bit unlucky as well.

But of course:

• Not everything might be balanced perfectly.
• Some implementations can definitely be improved.

Edit:
By the way, did you know that the Improvement "Fort" was added to RaR to also help with Native Raids (or generally protecting your borders)?
Raiding Natives cannot simply walk by these Forts if they are in their path and thus will attack those instead of your cities.
In these Forts they cannot really do much damage compared to what they can do in Cities (like killing colonits, destroying buildings, stealing gold, ...)

 

I wasn't saying that nothing should ever happen regarding failed raids. I'm saying that it would make sense to restrict how bad it can get, particularly if there is a fully functioning stockade or fort.

Admittedly I haven't played vanilla in around a decade or so, but I do remember them being in the classic colonization. It just shows that vanilla didn't live up to the expectation of being the classic colonization implemented in a modern engine. Too much missing contents. To be honest I was rather disappointed when it came out.

I was thinking something along the lines of if it can't destroy a building, then it will be skipped in the list of what can happen, hence increasing the chance that they damage the defense instead. This means the next raid could get to the building due to the damaged defense. It would need mob raids in order to really work or the defense would just be repaired between raids.

Another thing the raid could do would be damage units working on plots (in the colony, not pioneers). If a unit is damaged, you need to get them out and heal them or if you just let them work, they might be killed during the next raid. It seems obvious raids would make such targets if the colony itself has a stockade, which prevents them from going inside.

You do have a point though. It shouldn't turn into a free combat experience setup, though usually either I have the military to defend myself or I get raided. I rarely get both at the same time.

It should be. This thread not about avoiding consequences. It's about avoiding unreasonable consequences, which can make people rage quit. In other words it's about figuring out a balanced system.

As the title indicates: it's a debate about what people think would work best and match expectations. If everything goes well, a whole lot of proposals will show up and a number of them will be discarded due to various counter arguments. As such no post or statement should be read as "we decided" or anything like that. Objections with proper arguments are just as valuable as proposals.

 

That is why most of us became Civ4Col modders.

Would be fully ok for me.
Also it does not sound like too much effort to implement.

Such an implementation is not undoable.
But I think we should try to keep the logic / implementation a bit more simple.

Native Raid logic does not consider your Military Strength that much if I remeber correctly.
(It might impact the Attitude of Native Nations though and thus affect Native Raids indirectly.)
They are mainly triggered by the Attitude of the Natives towards you - and the distance of your colonies to them.
(I would need to check my code again to be sure.)

I don't disagree at all.

But as I said, if you play carefully and understand the RaR / WTP concepts a bit Native Raids can usually be avoided completely.
But yes, there is also some bad luck involved like settling next to a Native Nation that is really easy to get angered.

Again I fully disagree.

I just wanted to explain some game feature dependencies and old design thoughts considering balancing that might help to understand.
Sometimes the problem is simply a completly different one.

 

This is by the way also true.
In RaR, Natives declare generally less wars if your game style is careful and you know how to use certain features.

In RaR we had heavily reworked the balancing of Native Leaders. (Compared to TAC their personal difference were leveled out a bit.)
Other balancing and game mechanic changes were added over the years to adjust to new features implemented.

Just to name a few that affect this current topic:

A) In TAC some of Native Leaders were extremely aggressive and some of them were way too peaceful. (Trying to model their specific characters.)
With some bad luck (settling near an aggressive Native Nation) you could hardly prevent early wars with them and just were extinguished.

--> This was not understandable for many of our players.
They simply felt they were treated unfair without having done anything wrong.

--> These Native Leaders are however still more likely to start "Native Raids".
It is just not as bad as having them directly declare war (which however also still do a bit more often).

B) Some adjustments were needed because of the huge cities in 2-Plot-City-Radius.
Your Cities sometimes just took away too much land from the Natives and thus also they declared war too often.

--> Although more or less understandable to most players many still did not like that.
They often felt that they were forced to play "War game style" too much.

C) Some adjustments were needed because in "Gigantic" Maps the Natives have much more settlements and more Units.
They became way too powerful to defend against them in early game when your army is still weak.

--> Thus we for example increased the min. distance between Native Settlements.
Leaving a bit more space for European settlements and thus less Wars with the Natives caused by expansion.

--> Missions and Trading have become more powerful considering impact on Native Player Attitiude.
Allowing to get better Attitude with your Native neighbours as well.

...

----------

This would have resulted in a situation were RaR was a little bit too much "easy expansion" without any consequences.
"Native Raids" was one of the ways too counter that. (And of course it was request because it existed in Original of 1994.)

Result now:
A) Natives don't directly declare War as easily as they did but instead do start Native Raids first.
This should be a clear sign to the player to start thinking about potential adjustment to game style.

B) You have an additional arsenal of game mechanics and features to counter even Native Raids completely.
These are your possibilities to adjust your game style without completely abadoning your strategy (e.g. expansion).

----------

Sorry for explaining / repeating all that old stuff again.
But maybe it will help our considerations.

Currently I feel that we generally agree on almost everything.

• There is still room for improvement and some negative effects of "Native Raids" can be changed or made a bit less grave.
• We need to do this carefully though because there are dependencies to be considered or else balancing might suffer.

----------

I also believe however that we have not really talked enough about some other aspect:

Many players seem to simply not know about many of the changes and new game mechanics in RaR which now became WTP.
(We tried hard to explain those in Colopedia though.)

• If we read posts like "I had 30 Native Raids in this one game." or "I don't yet understand how native relations in WTP work." we should also think about ways to explain our features and game concepts better.

For TAC e.g. there were several "Let's Play Videos" as a Tutorial.
Yes there are some old RaR "Let's Play Videos" as well but most of them are really old and were not a Tutorial.

There were also many long strategy threads for TAC and RaR posting advices for other players.

It would be awesome if somebody from community created a WTP "Let's Play" and explain some of our features.
It would also be cool if somebody from community created some WTP "Strategy thread" and post advices here in our forum.
Something like this would really be awesome.

 

I implemented the game option "reduced colony distance" where the distance is reduced by one if the player already owns the 9x9 plots the colony will take up at first. It is great at "filling out the map", without leaving a bunch of unused plots, particularly in 1 plot radius games. However it backfired regarding the number of native settlements and it actually makes the game much harder. I'm considering making it a map option instead and provide multiple options like off, human only, European only or all.

We had a bug at some point where the min colony distance was set to 1 by mistake. Using this meant the distance was set to 0 meaning you could place colonies anywhere, including next to other colonies. That was.... interesting, but not very playable when the AI had that ability.

That goes back to adding some tutorial system. We need to figure out what to do with this issue. Maybe we should use the idea of adding the settings xml I proposed at some point because that way the mod can remember settings between games. We can then add tutorials and add buttons "don't show again in this game" and "never show again", where the latter will save this in the xml settings file. This way if we add a new feature, then players will get a popup about it the first time they encounter the screen in question. They can then tell they will never see the message again, but they are notified something changed.

We also need an updated forum system, which prevents threads from going off topic compared to the original post

 

1. I like the concept of raiding. (defenses and units in cities should discourage by a percentage, since there should be some common sense involved)
2. I like the concept that certain tribes are more likely to raid, since that puts a strategic decision in the game whether I should tolerate or just wipe them out. And how much effort I should put into keeping them happy.
3. I like the concept of failing defenses increasing damage.
4. While percentage of what you have determines penalty, put a cap on it. No failed raid should (or even one that succeeds should ever get 6000 gold.
5. Guns and horses should always be taken first. Add wine or rum if present. Population also.
6. There have to be actions you can do to really discourage them so there are choices.

Going back to a CIV example. You start next to Monty, you know what's going to happen eventually unless you really expend a lot of resources into keeping him happy. If it's Asoka, you know you need minimal effort.

 

My only request is that however raids work, it be documented. I'm surprised, reading this thread, to learn the combat odds I create are not determining the raid outcomes I get. If any behavior is not intuitive to Civ4, it needs documentation.

One thing that should be visible to the player is how their actions raise or lower the risk of raids. I usually have no idea when to expect a raid, so I basically always expect one. And before long I get one every turn, without any idea how I'd reduce that frequency. Not to imply I'd find that desirable. I win the raids frequently enough that they just seem to be a way to increase my labor base with slaves. Worth the turn or two of construction I lose every 20th raid.

The original Coloniation showed !!! marks over villages you had pissed off. That gave enough feedback to the player whether a discrete action just caused a big increase in risk, or if it was just creeping up over time due to the population of nearby colonies. Comparing village anger on your border to villages farther inland helped the player to see their footprint. In WtP it's opaque to me. Maybe I need to learn the names of Indian leaders and match their names and colors to the ridiculous foreign advisor web to see how frowny they are, but I'd prefer to continue ignoring that. The difference between cautious, annoyed, enraged, or whatever, has never been clear to this player.

Spoiler :

 

Now I feel stupid for not thinking about this already. There are those colored smilies in the diplomacy system to indicate attitude. Adding those to the city billboards isn't hard. The only problem is that we only have them in low resolution, meaning they would be scaled up unless we can draw/find higher resolution ones.

I never use it either due to being.. well ridiculous, particularly after encountering many players. We can make a new one, but that brings up the question: what should it look like and what actions should the player be able to do there?

Really? They are just an indicator for the attitude based on past actions. If you take their land, the attitude drops. If you trade with them and give them gifts, the attitude increases. Low attitude makes them go on raids. The named stages is just an indicator of the internal attitude meaning you get an idea if they love or hate you.

 

How does the player know that the attitude drops if they take their land? It's logical, but where does he see it? By checking the Foreign Advisor? By initiating diplomacy? Looking at the scoreboard? I think it's one of those, and this player is not checking any of those every turn, and they do not update within a turn, I don't believe. Further, how much does attitude increase if I trade? The player should have some sense of magnitude. Does a trade of 1 good increase attitude the same as a trade of 100 goods? If I give them a good deal, based on European prices, does their attitude increase more than a competitive bargain? How much trading does a player need to do to offset the drop from taking 1 tile. 10 tiles? I don't think a player can say, were they interested.

 

Did you check "Colopedia"?
I am pretty sure that the "Native Raids" is documented in "Game Concepts".

But you are right, it could be documented and explained better.
Maybe we can also visualize it better ingame.

But it would also really help if community members (that know RaR / WTP) explain features or strategies to other community members.
(e.g. Wiki, Strategy Threads / Let's Play Tutorials)

I simply did not know how I could have visualized it, when implementing "Native Raids" in RaR.
But maybe we can come up with a good technical solution / implementation concept in the team.

 

It is absurd that the Indians can gain 6000 gold from a failed raid, this is far more damaging than the REF. Also the player cannot defend against this, as even if the attackers are killed they still steal money. When a player destroys a village they may gain a treasure, which needs to be protected etc. Whereas natives can have a raid fail and then thousands of gold are magically spirited away to their settlements. This is internally inconsistent to the rules of the game. Instead it would be good to see more nuance in the affects of raids, where not every raid has to end in the death of a unit and not have the natives just raid Cities.

It would be good to see more reasonable consequences, such as:
" A Native raid was successful, 100 sheep etc were stolen"
or "A Native raid was successful, they have destroyed a plantation"
or " A Native raid was unsuccessful, but a storehouse was destroyed, 80 Coats etc were lost"
or " A native raid was successful a farmer has been killed / kidnapped"
or " A native raid was unsuccessful, but 15 tools were stolen from a wagon"

Perhaps the underutilised happiness mechanic could also be used, with the towns populace or even the King being made unhappy by continual raids on settlements.

Original post here : https://github.com/We-the-People-civ4col-mod/Mod/issues/323

 

Here is a concrete proposal, pasted from my notes on balancing:


NATIVE GOLD RAIDS


Factors that should increase or reduce the amount of gold that can be stolen:

Higher pop : More gold to steal
Defenders and walls\defensive structures : Reduction of potential gold to steal

Proposed formula:
gold stolen = (Population / ( max(1, fortified defender count) / (1 + fortificationFactor)) * randomRange(10-20)


Examples

size 20 city with 2 defenders and +60% defense:

(20 / 2) / 1.6 = 6.25 base (gold stolen will be about 62-120)


size 40 city with 1 defender and +30% defense

40 / 1 / 1.3 = 30 base (gold stolen will be in the range 300-600)


size 5 city with 1 defender and no defense

5 / 1 / 1 = 5 base (gold stolen will be in the range 50-100)

This gives players more of a reason to build fortifications, particularly in inland cities
near natives. It also gives an incentive to guard cities that can be raided. We may also consider
a promotion that reduces the probability of raids being successful.

Feel free to suggest different parameters. As for the formula itself, I think that we should not
consider too many factors.

 

Sounds good.

Don't forget to consider the available / current gold of the Player.
(We should of course never end up with negative gold amounts.)

Otherwise I think the suggestion is really good.

Probalby not a good idea.
It would be really difficult to teach AI when it needs it or not.

Let us please leave that out.
(We should not make it overly complicated either.)

 

Many interesting ideas. Raid gold could be halved(or some other percentage) if no defender is killed. This would give incentives to build defenses and have stronger troops. Just an idea. But yeah, the 6000 gold example. (which i've experienced) is just plain stupid and really off putting.

 

Is this currently hardcoded to apply to native civs or UNIT_BRAVE, or could the system be adjusted to be more generalized/flexible? e.g. this could instead use Promotioninfos or Unitinfos tags enabling adjustment of various raid strengths on attacking a city, such as <iRaidGoldTheft>, <iRaidYieldTheft>, <iRaidUnitKilled>, <iRaidBuildingDestroyed>; you could then have buildinginfos.xml be able to grant defense strengths versus each effect. The raid effects and available building defenses could then be balanced/adjusted until people feel they are challenging but not frustrating.

 

• Native Raids is hardcoded for Units of Native Players only (thus indirectly Unit Native Brave), yes.
• The logic for success vs. failure or the different possible outcomes is also pretty much hardcoded. (I did not think anybody would like to balance that in XML anyways.)
  
• Also the UnitAI the logic will assigne is currently only valid for Native Braves. (XML config for assigning UnitAI, but UnitAI itself hardcoded.).
• There are XML settings for the Native Leaders as well (e.g. <NativeRaidAttitudeThreshold>) so some Native Leaders / Nations are more likely to raid, others less.
• And there are few XML settings for global balancing in GlobalAltDefines (for tuning some total feature thresholds and chances - but not for specific outcomes).

So while there is some XML balancing options in there, I only coded the ones I felt necessary.

Summary:
There is still some room for making most of my old features more XML configurable.
Doing so is however effort, adds more XML config parameteres and might be hard to understand without expansive documentation.
We need to find a good balance of having all necessary XML config parameteres vs. having too many XML config parameters.

 

I love the idea but it"s probably a little complicated.
What about simulating this by reducing the settlement's health ? This would have the added benefit of adding some depth to health management.