Dale/Snoopy PatchMod (Bug-fixes and more)

[snoopy369]

5:1 should be competitive, actually, if you build right (and 150 should not outnumber you 5:1 I'd think ... that's only 30 units, or 1500 guns). 5:1 is about the limit of competitive, though, and requires good tactics to defeat. Obviously not appropriate for the lowest level, though.

Killing warships helps more than just killing the units - it makes the units take longer to come out also, doesn't it?

What level were you playing at?

[crantor]

These games were at the middle setting, Conquistador? I will try a few runs on easiest vs hardest to see the REF sizes. I never said I am a good player! But I have won vanilla with 1:3 about 1:4 ratios on easiest, so with some more strategy I agree that a 1:5 should certainly be doable now that the mechanics are slowly sinking in.

You are ofcourse correct about ships having an impact, I am just not sure it is the most cost-effective way. I'm thinking privateers might be more cost-effective, SoL are just too expensive for mass production. But I am waiting for some numbers to see which is best.

[snoopy369]

Hmm, Conquistador is hard enough I think 150 is acceptable.

Ships ... cost-effective maybe, I guess. I think in large numbers defending an empire with a small coastal footprint they can be useful, but there's a minimum to be useful; having 5 or 6 SoLs isn't enough, and if you have a large coastal footprint that's too much area to defend.

[AgentTBC]

As I've stated, I think Col2 has the potential to be a really good game but the gigantic incentives towards maintaining as small a colony (in terms of # of cities) as possible combined with the massive incentive towards never generating liberty bells until you have 3 statesmen+newspaper in every city kind of ruins it for me.

I think I have a solution to the first part of the problem at least; greatly ramp up AI aggressiveness towards the human player if they outnumber you. If you've got 2 cities and Spain has 7, they should declare war on you even with no provocation. This would require the human player to maintain a colony size relative to the map size big enough to hang with the AI players.

Right now I've never had another european power declare war on me on any difficulty level, and only have natives declare war on me one time.

Or ramp up the aggressiveness with difficulty level. In any case, I shouldn't be able to go 200 years with two size 5 cities without Spain or France or whoever coming over and kicking my butt.

[snoopy369]

I think you would be surprised to find that a lot of people think there are substantial incentives to having many colonies. There are incentives to having fewer, but having more gives you significant bonuses as well - more diverse trading goods (thus less vulnerability to price shifts and boycotts), more guns produced (10 cities, 5 of them mining+making guns, can make a LOT of soldiers), more money, and more Founding Fathers. A large colony (10-15 cities) can handle any size REF fine (when played well, of course).

I do think increasing the base REF would change this equation somewhat, though, further towards larger/more colonies. I also think that Dale's AI changes will have some of the effect you want (encouraging AI aggressiveness).

[crantor]

A more decisive victory this time. Conquistador, REF size ~220, Rebels ~70. I gave the king quite a few bags of gold and it seemed to have a positive (for me) effect on the REF size.

Is there any disadvantage to squeezing troops from the king, besides the "becoming a drain on my army!" negative hit? (does negative attitudes affect anything?)

I fell behind in the early stages and the english focused on liberty bells. Could have been dangerous, I was stuck in tundra land (with alot of silver) and angry natives. Two dragoons wrecked their independce party and colonies though, but they did put up a defense, arming colonists and moving between colonies. So something is going on, at least they aren't completly apathic once they get their colonies up and running.

Defensive structures as coast is a complete waste of resources, or am I wrong as usual? Inland fortresses atleast keep them off for a turn or two longer.

[Dale]

Neg attitudes from the King have a small effect on tax hikes and the size of the REF. So whilst you can reject the King, it will have a minor effect on other things.

[Kriegsspieler]

Originally posted by AgentTBC
As I've stated, I think Col2 has the potential to be a really good game but the gigantic incentives towards maintaining as small a colony (in terms of # of cities) as possible combined with the massive incentive towards never generating liberty bells until you have 3 statesmen+newspaper in every city kind of ruins it for me.

I think I have a solution to the first part of the problem at least; greatly ramp up AI aggressiveness towards the human player if they outnumber you. If you've got 2 cities and Spain has 7, they should declare war on you even with no provocation. This would require the human player to maintain a colony size relative to the map size big enough to hang with the AI players.

It's kind of surprising this doesn't happen with the game at persent, given that in Civ4 the other powers would be all over you if there was a disparity of the kind you're describing!
It makes you wonder whether they just left parts of the game engine behind here . . . .

[joncnunn]

AI behavior is not part of a game engine.
The game engine is basically responsible for displaying the map and maping mouse clicks to tiles and units, etc.

[AgentTBC]

Originally posted by snoopy369
I think you would be surprised to find that a lot of people think there are substantial incentives to having many colonies. There are incentives to having fewer, but having more gives you significant bonuses as well - more diverse trading goods (thus less vulnerability to price shifts and boycotts), more guns produced (10 cities, 5 of them mining+making guns, can make a LOT of soldiers), more money, and more Founding Fathers. A large colony (10-15 cities) can handle any size REF fine (when played well, of course).

But the bonuses from having many colonies are only relevant if you HAVE many colonies. You don't need those bonuses if you have few colonies. If you have 2 cities, you don't need much money because all you want is enough to buy 10-12 cannon, one statesman, and maybe 750 guns and horses. That's not really a lot of money. So who cares about diverse trading goods since you'll never sell enough that price shifts matter much. Gun production doesn't matter since you'll buy it all from Europe. Producing guns requires a higher population which makes it harder to get enough liberty bells in a very short period of time, thus keeping the REF small.. And if you're getting founding fathers it means you're producing liberty bells, which you shouldn't be doing anyway unless you're ready for the revolution in which case the revolution will occur when the REF is like 14 soldiers, 12 dragoons, and 8 cannon.

So you're correct that there are advantages to a lot of colonies, but you only NEED those advantages if you have a lot of colonies in the first place. It's kind of a self fulfilling prophecy: Having many colonies makes it much harder to win, so you need advantages you wouldn't have with fewer colonies.

[snoopy369]

How is more money and more guns not relevant? Having more colonies = more of what you need to win = advantage; having more colonies = REF is larger = disadvantage. I could just as easily reverse the equation and say having fewer colonies=smaller REF = advantage, having fewer colonies = fewer guns = disadvantage.

I think it's a good thing that the game to some extent scales for colony size, in that it allows you to win without having a ton of colonies. That's one of the most significant criticisms of Civ 1-3, after all, that you had to have as many cities as possible to win (ICS) at any high difficulty level; it was the one-and-only strategy that worked well.

[AgentTBC]

I don't know how to explain it more clearly; more guns is irrelevant because you don't need more guns to win. If the REF only has like 8 cannon, 12 dragoons, and 14 soldiers it is trivial to defeat it with 10 cannon and 10 dragoons. So you don't need to produce ANY guns, just buy 750 guns + 750 horses from Europe which you can do very cheaply. Much easier to do that than to build up the industrial base to produce enough guns to equip people. That takes much longer, is more complicated, and thus is much more prone to problems.

[Kriegsspieler]

Originally posted by joncnunn
AI behavior is not part of a game engine.
The game engine is basically responsible for displaying the map and maping mouse clicks to tiles and units, etc.

Maybe so, but that doesn't make it any less odd that an a.i. behavior in Civ4 would be so conspicuously absent in this game.

Originally posted by AgentTBC
I don't know how to explain it more clearly; more guns is irrelevant because you don't need more guns to win. If the REF only has like 8 cannon, 12 dragoons, and 14 soldiers it is trivial to defeat it with 10 cannon and 10 dragoons. So you don't need to produce ANY guns, just buy 750 guns + 750 horses from Europe which you can do very cheaply. Much easier to do that than to build up the industrial base to produce enough guns to equip people. That takes much longer, is more complicated, and thus is much more prone to problems.

Forgive me, and maybe I'm missing something obvious. But I thought that one point of this mod was to make it so the home country ramps up its forces in anticipation of the revolution. So maybe the strategy you are describing works for the 1.00 version but won't fit a game with a better prepared a.i.

[AgentTBC]

Forgive me, and maybe I'm missing something obvious. But I thought that one point of this mod was to make it so the home country ramps up its forces in anticipation of the revolution. So maybe the strategy you are describing works for the 1.00 version but won't fit a game with a better prepared a.i.

This will require decoupling of the REF size from production of liberty bells. I hope we move in that direction. If so, then my point is indeed rendered moot.

Making the AI much more aggressive combined with significantly increasing the initial size of the REF (depending on map size and difficulty, of course) would certainly address most of my problems with the game. If the AI is more aggressive and the REF starts significantly larger, it will no longer be the ideal strategy to keep your number of colonies and population as low as possible because you won't be able to defend yourself or make enough money. As it is, though, you don't need to do either of those things so it is unnecessary to found more than 2-3 colonies, ever, no matter the difficulty or map size.

[Steve Clark]

I am playing my first game (with this mod loaded) and when I try to disembark any unit onto land on my second turn, the game crashes.

I started a new game without the mod and was able to disembark without crashing.

[Dale]

The latest v1.03? The initial 1.03 had a bug that caused a crash on starting the mod.

So redownload it and try again please, and let us know if it reoccurs.

[Steve Clark]

Will do.

[Dale]

Originally posted by AgentTBC


This will require decoupling of the REF size from production of liberty bells. I hope we move in that direction. If so, then my point is indeed rendered moot.

Making the AI much more aggressive combined with significantly increasing the initial size of the REF (depending on map size and difficulty, of course) would certainly address most of my problems with the game. If the AI is more aggressive and the REF starts significantly larger, it will no longer be the ideal strategy to keep your number of colonies and population as low as possible because you won't be able to defend yourself or make enough money. As it is, though, you don't need to do either of those things so it is unnecessary to found more than 2-3 colonies, ever, no matter the difficulty or map size.

Making changes to the AI to make it competitive is one of my goals. With v1.03 it is the first Col AI where it can actually survive a WoI or Indian War.

I'm sure everyone will agree that advances in the AI's ability to survive in this game will greatly improve the playability of the game.

[Steve Clark]

I think I downloaded yesterday's version of 1.03 but today's version worked without crashing at that point. Thank you.

[joncnunn]

I'm not so sure REF should be completely decoupled from Liberty Bells. I'd just reduce the weight of that and increase weight of other things to increase REF:

1. *** Refusal of Kings Demands *** (I'd make refusal of them the fastest way to increase REF)

2. Other REFs increasing. (King concerned about Balance of Power with the other Kings)

3. The player being involved in a war prior to DOI (against either Indians or other players).

4. The player losing tiles by any means prior to DOI. (King concerned "his" territory shrinking)

5. The player is at peace, and his army is way higher than what is needed for self defense (considering his population, cities, and rival players armies) AND is growing.

[AgentTBC]

I'd love to see refusing the King's demands be the fastest way to increase REF. It would really make it a more difficult decision. As it is, I always refuse to pay his danegeld unless it's for some petty little amount.

[Kriegsspieler]

Originally posted by AgentTBC
I'd love to see refusing the King's demands be the fastest way to increase REF. It would really make it a more difficult decision. As it is, I always refuse to pay his danegeld unless it's for some petty little amount.

Yes, I agree. Or it could be coupled to higher tax increases and refusing THEM leads to faster growth of the REF. Either way, you've got to be made to PAY for getting the King mad at you.

On another point: do you guys think anything easy can be done to make a.i. defenses in the New World a little more robust? I just attacked the Dutch and was surprised that in taking 3 of their cities they only had one armed minuteman unit.

Maybe I was just lucky.

[Durnil]

Dale/Snoopy - Are there any plans to improve the functionality of the Route List window? Right now it's basically horrid. I'd like to be able to sort by item, starting city, and ending city. I can write the code for everything but the sort_by buttons themselves (GUI programming is anything but my strong point). Please let me know if you can add a three-radio control to the route list somewhere, and I'll write up the code to handle it.

[Dale]

Originally posted by Kriegsspieler

Yes, I agree. Or it could be coupled to higher tax increases and refusing THEM leads to faster growth of the REF. Either way, you've got to be made to PAY for getting the King mad at you.

On another point: do you guys think anything easy can be done to make a.i. defenses in the New World a little more robust? I just attacked the Dutch and was surprised that in taking 3 of their cities they only had one armed minuteman unit.

Maybe I was just lucky.

What version of this patch, what turn (so what stage of the game) and what difficulty?

[Kriegsspieler]

Originally posted by Dale


What version of this patch, what turn (so what stage of the game) and what difficulty?

1.03, on conquistador.
I think I was playing around 1550 or so.

[snoopy369]

Originally posted by Durnil
Dale/Snoopy - Are there any plans to improve the functionality of the Route List window? Right now it's basically horrid. I'd like to be able to sort by item, starting city, and ending city. I can write the code for everything but the sort_by buttons themselves (GUI programming is anything but my strong point). Please let me know if you can add a three-radio control to the route list somewhere, and I'll write up the code to handle it.

Hmmm, python is a big weakness here as well but when I'm done with what I'm currently doing, i'll give it a look...

[MontyLaremane]

I've looked at the code for how the REF reduction in this mod (I'm nosey like that ) and I'm confused. I'm sure I've missed something, but perhaps someone could explain how it's working?

If my understanding of C++ integer division is correct (i.e. the result of the division of any two integers is truncated to an integer), it looks like the new KingUnitThresholdPercent handicap modifier (CvPlayer.cpp line 7904) won't work as intended. Instead the bellRateMultiplier will end up as 0 or a simple multiple (1x, 2x) of the vanilla threshold increase rather than a nice percentage of it (90%, 120%, 200%).

A similar thing might be happening in the gold related reduction at the end of CvPlayer too.

Have I missed something? (Am I making any sense?! )

Sorry to be a pain chaps! Keep up the excellent modding

[snoopy369]

Can you explain it a bit further (copy/paste what you're looking at and how you think it is wrong)? I tested the latter out fairly thoroughly (and rewrote it from the original code to be a bit more clear, the original one was a mess of math) and it seemed to behave as expected (though I can't actively debug it, so I had to manually do so by running through REF increases). Though Dale might have changed it from my coding actually when he added the gold increases, don't recall.

[Ellestar]

I guess MontyLaremane wanted to say that you divide an integer getKingUnitThresholdPercent() by 100, that rounds it down to a nearest whole number

bellRateMultiplier = ((GC.getHandicapInfo(getHandicapType()).getKingUnitThresholdPercent() / 100) * 
GC.getDefineINT("REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE"));

Now, my question is why do you bother with getRevolutionEuropeUnitThresholdMultiplier() at all? If i understand it correctly, it doesn't significantly change REF size, as it just defines when units will start to be added in batches of 2,3,4... units. Actually, with your approach you may as well increase REF, since REF increase triggers with a smaller getRevolutionEuropeUnitThresholdMultiplier() and a number of units added in that increase will happen with a bigger getRevolutionEuropeUnitThresholdMultiplier()

[MontyLaremane]

It's this particular line from CvPlayer.cpp (version 1.03), specifically the part in bold.

bellRateMultiplier = ((GC.getHandicapInfo(getHandicapType()).getKingUnitThresholdPercent() / 100) * GC.getDefineINT("REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE"));

My understanding is that the result of that division will always be truncated to an integer. So, if the handicap threshold is less than 100, the result of the division above will be 0, so bellRateMultiplier is also 0. If the handicap threshold is 100-199, the division result is 1. If the handicap is 200-299, the result is 2 etc.

Perhaps it should look like this so the multiplication happens first and the division last (I think that's why the maths looks screwy):

bellRateMultiplier = ( (GC.getHandicapInfo(getHandicapType()).getKingUnitThresholdPercent() * GC.getDefineINT("REVOLUTION_EUROPE_UNIT_THRESHOLD_INCREASE")) / 100 );

As I say, I could very well be wrong as my understanding of C++ is limited.