Cradle 1.31

[hexagonian]

Originally posted by Bluevoss
Hex sez....
Boost the max city number in TYRANNY to 25. This is lower than the 30 for TRIBUNAL EMPIRE - the thinking here is that the human player is usually closer to the city cap of 30 than the AI, and if he is over the cap, there will be more difficulty in maintaining a stable happiness.

In thinking about this, it may be better to create a new government with the attributes of Tyranny, but also has the higher city cap. It's entirely possible that the AI, with the slighly improved build city priorities in Cradle 1.32, might actually get above the city cap at the beginning of the game, and then get in trouble when going to Dynasty/City State... with the reduced city caps. Plus the high city cap may make Tyranny a good long-term government in the early game - even though it has the lowest science/production/commerce numbers, the city cap may very well overcome those numbers. This is not my intent.

Originally posted by Bluevoss
Has there ever been any testing with the idea of reducing the empire sizes for ALL govt types? If the human is using this to pull ahead of the AI (especially in the latter stages of the game) and the AI isn't using it, then maybe the human empires should be reduced slightly. That would give empire builders something to consider besides who to kill and who to spare (namely, how to deal with all those rioting citys).

A lot depends on the map size - I'm finding that using the standard Gigantic maps with the current numbers makes for a good fit. Obviously, on smaller maps with the same numbers, global conquest is easier.

When setting up Cradle, I had dropped the city caps slightly from the default game. I also added the Nomad at the beginning of the game and added the Settler (which is a Nomad at double the cost, unlike the default Urban Planner which starts a size 3 city with some improvements) to the tech tree to slow down mid game expansionism.

Most importantly, I increased the unhappiness modifier to the city caps, so when you go beyond the cap, you will run into higher unhappiness. In my current game, I've been pushing the cap in all of the governments, and at the same time, I have to selectively choose what to conquer. With at least 150 enemy cities in the game, I cannot own the world and all of its cities.

But at the same time, there is the Raze City SLIC code in Cradle, so a lower cap will not stop a player from global conquest. But I generally do not play that way - I do not like razing cities.

The main problem has been that I feel the AI respects rival borders when it builds a city which reduces the amount of land where the AI will build a city - the human does not. Cradle 1.32 attempts to alter this - so far I have not gotten any reports on this area of the game though.

Originally posted by Bluevoss
Further, has anyone else looked into the advantages of drasitcally shortening the "stance upgrade" times. Taking 15 turns to go from peace to alert (and then another to go to war) is way too long. And this is shown by how few people EVER use it. I experimented with reducing the times to no more than 4 turns per level (I don't remember how I factorded the breakdown, but once I reinstall 1.32, I'm going to repeat it). It really adds an element into the game, namely that it can occasinally push fence-sitting AIs to attack you (which livens up the peace and gives the player excuse for some "clobberin' time".

I will probably adjust this in the next update, given that the AI is doing well in maintaining good production. The AI will not power down though...

Originally posted by Bluevoss
Sorry I've been away from the base for a while. Been reading the messages but haven't had a chance to participate. I got plaster all over a train layout and a book half-written. Busy, busy, busy...

Good to see you back! I'd like to read that book when published.

[Lou Wigman]

I have noticed the change to the AI's behaviour in 1.32 It is now putting cities either right next to my border or inside my border if at war. It is certainly more desparate to expand. I do have a concern about the AI's placement of its cities. Generally they are much too close together. It may be my imagination but they seem to be closer in the last few games than before. Could be wrong on that one though. Too hard to tell. I have tried fiddling with some parameters in const.txt but this doesn't seem to be effective.

As for the war/peace transition this has no part to play in CRADLE. Not to have your army on a permanent war footing is suicidal !

I have also noticed other changes in the AI behaviour which are less satisfactory. Is it true that the wick has been turned up on goal_attack/siege? If so this has been at the expense of city defense. I am finding many more cities with weak defense or indeed with no garrison at all. I am particularly concerned by the fact that the AI will frequently move its army out of a city that I am approaching leaving it empty. It always did this to some extent but it seems to be much, much more frequent now.

I know there is a dichotomy here. If the AI defends its cities furiously it will have huge garrisons everywhere. If not it does other inappropriate things. Don't know that there is a solution to this.

[hexagonian]

Some questions...

Originally posted by Lou Wigman
I have noticed the change to the AI's behaviour in 1.32 It is now putting cities either right next to my border or inside my border if at war. It is certainly more desparate to expand.

Do you see this as helping the AI? The reason why is that I had noted that in the course of my game, I would ofterntime see a nomad either sitting in a city or fortified (sometimes with a stack) with no place to go. I removed the flag for civs that are at war to respect borders in terms of settling, figuring that I would put an extra thorn in the side of the human player, if not only does he need to deal with the AI armies, but also the AI settler policies. The potential downside is that the human may get a city, or get the PW/Gold bonus when taking that city.

This change is made in the CRA_goals.txt file.

Originally posted by Lou Wigman
I do have a concern about the AI's placement of its cities. Generally they are much too close together. It may be my imagination but they seem to be closer in the last few games than before. Could be wrong on that one though. Too hard to tell. I have tried fiddling with some parameters in const.txt but this doesn't seem to be effective.

This is also an attempt to get the AI to expand more. I dropped the settle radius from 4 to 3. I'd like to see either a gamefile or a screenshot of this too.

This change is made in the CRA_strategies.txt file.

Originally posted by Lou Wigman
As for the war/peace transition this has no part to play in CRADLE. Not to have your army on a permanent war footing is suicidal !

Agreed, but allowing a player to walk on the edge is not a bad thing. When playing CTP1, I always used the transition slider, but haven't in CTP2 Cradle because there was no incentive to do so, due to the long time it took to get back up to speed and the fact that I was always at war anyhow.. So I may reduce the time (given the fact that I had already hamstrung the global sliders)

Originally posted by Lou Wigman
I have also noticed other changes in the AI behaviour which are less satisfactory. Is it true that the wick has been turned up on goal_attack/siege? If so this has been at the expense of city defense. I am finding many more cities with weak defense or indeed with no garrison at all. I am particularly concerned by the fact that the AI will frequently move its army out of a city that I am approaching leaving it empty. It always did this to some extent but it seems to be much, much more frequent now.

Question - what size map/number of civs and what turn are you on when you started noticing this?

I haven't adjusted the attack/seige priorities for quite some time now. It's possible that in increasing the priorities to build transports/settlers (this was because of a discussion with Dale and the success he was having with WAW in this area) that there is now less military in play. I did not use his numbers, as I thought they were too extreme, but I have really wanted the AI to build more cities.

This change is made in the CRA_strategies.txt file.

I had also added Martin's AI unit pathing SLIC, but I cannot see how this would cause this effect.

The two files, (CRA_goals.txt and CRA_strategies.txt) can be picked up from Update 1.3 and dropped into your 1.32 setup without affecting any of the other changes made in the updates. I may move back to those numbers.

I'm currently working on adding the extra Tyranny government for the Dark Ages for 1.33.

[Lou Wigman]

Well. I'm not sure that it is helping the AI at all. It is putting cities in pretty crazy paces. From the human players point of view this means that the 'settler' becomes a potent weapon. You could savely ignore a settler before but now you can't since it could build a city right in the middle of your empire. As for giving the human an extra city I doubt this very much. As I said these 'extras' are usually in the most inappropriate spots and are candidates for disband/bloodbath.

The problem with the close arrangement of enemy cities is that when I capture one of these I just want to get rid of it to allow the surrounding ones to grow. In some games I have put up the 'disband' size to allow this. In others I starve them to death or sometimes I use 'bloodbath'. Problem with bloodbath is that you must make up your mind immediately whereas the 'rationalisation' process is usually done much later. I would prefer it if cities were not crowded too much in the first place.

I am playing 10 civs on a gigantic map with 40% land and 70% continents. I really start to notice the the AI abandoning its cities circa 1000AD when there are many huge armies milling about. For some reason an army of 12 will leave a city when my army(s) is next to or near it. The moment it does this it is obvious the city is now empty and I just walk in. It's not as though the AI is short of troops. As I said, there are usually a number of others in the near vicinity. At this stage I normally have 2 or three armies close to other so they can support each other, a tactic the AI doesn't cope with very well since I can generally pick off one of theirs without to much trouble.

[Lou Wigman]

Tile improvements

Strategy games are about making choices. The harder the choice, the better the game. There are three variables as far as tile improvements are concerned. Food, Production and Commerce. The player should be forced to choose between these.

The problem here is Latifundia. It is way, way too powerful. It makes all tile improvements except the production ones obsolete. I am a slow learner and it took me a while to work this out. Why would you build anything other than a latifundia?

At the risk of repeating myself this throws the baby out with bath water. Where is the choice between food and gold when you can have both?

I suggest that Latifundia become a pure commerce improvement.

[Lou Wigman]

I just started another game and played through to about 0 AD. Whatever changes were made to the AI goals & strategies aren't working. While the AI field armies are a tough nut to crack, the cities are not being defended at all. Most AI cities appear to have a garrison of 2 units. This is despite the fact that these are on the frontier. I'll play some more but I'm getting worried.

[hexagonian]

Originally posted by Lou Wigman
Strategy games are about making choices. The harder the choice, the better the game. There are three variables as far as tile improvements are concerned. Food, Production and Commerce. The player should be forced to choose between these.

The problem here is Latifundia. It is way, way too powerful. It makes all tile improvements except the production ones obsolete. I am a slow learner and it took me a while to work this out. Why would you build anything other than a latifundia?

At the risk of repeating myself this throws the baby out with bath water. Where is the choice between food and gold when you can have both?

I suggest that Latifundia become a pure commerce improvement.

The reason why food is included in the Latifundia improvement is because the AI will not build it, or build very few of them without that food bonus added as a benefit. The AI is hardcoded to put the greatest priority on building Food Improvements. So, given the choice of building a Farm/Pasture or a Latifundia without the food, it will build the Farm or Pasture. Adding the benefit helps out the AI. I'm not sure if the AI would build them even if the food benefit was dropped to 5 on Plains and Grassland, because it would then default to the Farm or Pasture with its greater food benefit.

I agree that when you discover Latifundia, Farms and Pastures becomes a lot less meaningful to the point of becoming obsolete. But look at it from this standpoint. You will spend a good part of the early game building nothing but Farms and Pastures until you get to the point of having the ability to build Latifundias. Latifundias add 10 food on Plains/Grasslands - the next Food Improvement (Advanced Farms) adds 20 Food on that terrain. So from a purely growth standpoint, the Advanced Farms would be the way to go.

If anything, I would increase the benefit (and cost) of an Advanced Farm before I would drop the food benefit from the Latifundia.

RE: AI vacating cities
I suspect that this is due to Frenzy. I have posted in the past that one of the main things I wanted to do in future updates was get an alternative for Frenzy because of this one issue. In this matter, I'm at the mercy of the SLIC writers, because I am not proficient at SLIC. Dale was working on something along these lines for his WAW Mod and was going to adapt what he does to Cradle, but until that happens, this probably cannot be adjusted adequately.

RE: City placement
Go into CRA_strategies.txt and look for the entries that appear like this

// minimum distance between settled cities
// (eg. new cities must be founded atleast 2 cells from nearest cities collection border)
MinSettleDistance 3 //from 4 april 2002


Change it to this

// minimum distance between settled cities
// (eg. new cities must be founded atleast 2 cells from nearest cities collection border)
MinSettleDistance 4 //from 4 april 2002

This will allow for more room for AI cities.

I can't say that the issue concerning AI cities settling within Human borders bother me though, from a gameplay standpoint. What I would like to see is heavily armed settler stacks moving into hostile territory and building cities, and this was my intent in changing the settler priority for where it can settle.

[Lou Wigman]

Herein lies the rub. My cities already grow like fury. Advanced farms are also made redundant by Latifundia. It simply doesn't make sense to build these either.

The fact that the AI won't build them is probably no big deal. The reason I want the extra commerce is to get my gold/science up. The AI already has an enormous advantage here.

[Immortal Wombat]

Does Martin's commerce improvement SLIC not solve the problem of the AI not building them?

[hexagonian]

Martin's code was a more recent addition to Cradle, and this may (emphasis on may...) allow the AI to overcome the earlier limitations within the hardcoding.

The main goal with the current settings for a Latifundia was to make sure that the AI was creating enough gold through tile improvements, because this raw gold will be translated into science (as opposed to the gold created by trade, which goes into the RB fund). At least in Cradle 1.3-1.32, I am seeing a very robust AI in terms of infrastructure and scientific research through the Medieval Age, and I think in great part because the AI now puts a priority on building Latifundias and is raking in that gold.

The best solution may be to reduce the food growth element to 5 on Plains/Grasslands for a Latifundia, but this may ultimately hurt the AI more than the human because the AI may still default to building Farms/Pastures in those areas - the bottom line is that the AI will generate a lot less gold, allowing the human player to cut the tech gap quicker.

This is one of those areas that I feel should be playtested, and I am asking if you feel strongly about this issue, to go into the files and make whatever changes you want (in the various CRA..._improvement.txt files), test it out and report back. Chances are, I will go ahead and make the change myself, but I currently do not have the time to test it over the long term due to limited personal time. And I still want to finish my current game, and have time to play other games too.

Question: How has the effect of Plagues been in 1.32? This may be making the game easier too.

Part of my thinking about Latifundias/Advanced Farms was also tied into the effects of Plagues. I know in my current Cradle 1.3 game, I got nailed with several Plagues that devastated my population base - one occured at about turn 600, and I would of liked to have had Advanced Farms at that time (my research was focused on another branch of the tree) just to get my population up again.

However, I had also lessened the effects of the Plagues in 1.32, making it easier to maintain your population and growth so you may be seeing more rapid growth in 1.32 due to a larger pop. base.

Somewhat related to all this is that most of the reports also have said that it is at the Medieval Age where the human catches up and surpasses the AI and this is the case in my current game (although I am lagging behind the leaders, I am quickly cutting the gap).

This goes back to the earlier statement that I have made that, given enough time, the human can catch up to the AI. Making the game longer WILL work in the favor of the human player. The original design of Cradle was to end at the Age of Reason too. I am interested in maintaining the gap between the human and the AI in the game for as long as possible, but I am wondering just how to effectively accomplish this.

One of the reasons I created the various victory condition files was to give players a defined endgame, and a defined timeframe to finish the game. Think the game is too easy??? Play with a 600-turn limit, as it is set up in the 'Rise of a Nation' option - all of the other options have turn limits too. Play without disbanding any captured cities - this is the way I play and it does make the game a lot harder, because I am constantly bumping up against the empire size caps. Larger maps allow time for the AI to get established. And playing with less civs makes the AI civs more powerful, because there is less competition for land. Having the MaxPlayer setting on a big number does allow for more civs to enter the game too.

[Lou Wigman]

Here are the results of some play testing

Commerce only Latifundia (+10 commerce)

The AI builds these in Forest, Jungle, Desert and Sand Dunes but not anywhere else. In view of your post perhaps 'Advanced Farm' should have a minor commerce impact. Further to my earlier post 'Supermarket' is probably best reworked as 'Market Garden' using the old civ2 name.


Obsolete Governments

Making certain governments obsolete after the Dark Ages causes these to be removed from the government pick list. However it allows the player to remain in an out of date government. Thus I was running Tribunal Empire and continued in this even though this government was now no longer available. Some SLIC code may be required to eject players from these governments. Like yourself I am a novice at SLIC and this is better done by someone who knows what they are doing.

As for the plagues in 1.32 these are having only a minor impact. The population lost is soon regained. Can be nasty if you get a couple in quick succession.

I agree entirely that the catching up with the AI should be delayed as long as possible to keep the game interesting. But simply making the game harder has other undesirable consequences. As it is now I frequently find myself facing legions with hoplites. The only reason I survive this is because the military AI is not up to par. Any human player would finish me off quick smart. A better solution would be to give the AI a boost in mid game just as the human player starts to pull away. Merging civs is one such solution. There may be others.

[hexagonian]

Lou,
Thank you for running the tests

Originally posted by Lou Wigman
Here are the results of some play testing
Commerce only Latifundia (+10 commerce)
The AI builds these in Forest, Jungle, Desert and Sand Dunes but not anywhere else. In view of your post perhaps 'Advanced Farm' should have a minor commerce impact.

This may work - I will most likely have to rework all of the other later game food tile improvements to follow suit.

One thing that I did on pricing the improvements was (for the most part) for every (+5) bonus that an improvement gave, the cost of that improvement went up 250 PW. Improvements that offered dual benefits went up 300 PW per (+5) bonus.

Originally posted by Lou Wigman
Further to my earlier post 'Supermarket' is probably best reworked as 'Market Garden' using the old civ2 name.

An aesthetic change - one that can go either way for me.

Originally posted by Lou Wigman
Making certain governments obsolete after the Dark Ages causes these to be removed from the government pick list. However it allows the player to remain in an out of date government. Thus I was running Tribunal Empire and continued in this even though this government was now no longer available. Some SLIC code may be required to eject players from these governments. Like yourself I am a novice at SLIC and this is better done by someone who knows what they are doing.

I would really like to get this feature in the game now. It seems to be an basic SLIC file too.

Is there a SLICer who wants to create this? This is how it should work.

When a player researches Dark Ages, he will have access to a new type of government (Dark Ages Tyranny) which will parallel Tyranny in terms of science, food, gold and so forth. Basically, it will be a large step down from the other Ancient governments.

If a player gets Dark Ages, the following governments are no longer valid

GOVERNMENT_TYRANNY
GOVERNMENT_DYNASTY
GOVERNMENT_CITY_STATE
GOVERNMENT_OLIGARCHY
GOVERNMENT_REPUBLIC
GOVERNMENT_DICTATORSHIP
GOVERNMENT_TRIBUNAL_EMPIRE

and a player would be bumped from them into (ideally)
GOVERNMENT_ANARCHY
or
GOVERNMENT_TYRANNYDA (my coding designation for Dark Ages Tyranny)

At that point in time, the next government available on the tree will be Theocracy, which will bring a player up to a much better government form.

Originally posted by Lou Wigman
As for the plagues in 1.32 these are having only a minor impact. The population lost is soon regained. Can be nasty if you get a couple in quick succession.

I have mixed feeling about the Plagues. I do like them, and they actually serve the purpose of keeping the human player in check in a realistic manner (after all, plagues were common and oftentimes devastating because they could not be controlled.) However, some players have a valid point that they can be frustrating occurances. So striking the balance has been tricky. I don't want the effect to be so weak, that it is merely an annoyance, but I do not want them to be game crushers either.

Originally posted by Lou Wigman
I agree entirely that the catching up with the AI should be delayed as long as possible to keep the game interesting. But simply making the game harder has other undesirable consequences. As it is now I frequently find myself facing legions with hoplites. The only reason I survive this is because the military AI is not up to par. Any human player would finish me off quick smart.

The importance of getting your tech level to the level of the AI is one of the main challenges in Cradle. I guess I do not see the aspect of being outclassed in what unit types you can build vs the AI as a bad thing. Based on my current game - I can survive and actually grow in that situation, but I now have to pick my battles carefully. There is a lot of cat and mouse that goes on now.

Yes, it takes a long time to get to that point, but I am no longer hearing that players can beat the game after 200-300 turns either. (it may very well be possible on small maps though)

Being outclassed in unit type and numbers is the only means to keep the AI competitive in any computer game I have seen. I cannot see any way to get an AI to use clever tactics that truly simulate human thought. An example of this is the Frenzy code - it does a good job of moving and assembling stacks, but there are limitations within the code that can be exploited.

Originally posted by Lou Wigman
A better solution would be to give the AI a boost in mid game just as the human player starts to pull away. Merging civs is one such solution. There may be others.

There are ways to give the AI a late game boost - the problem is that there are many levels of player out there - so a late game boost may do the very thing you wanted to avoid in your last point above. Take a look at the file CRA_diffDB.txt to see how you can change the later game bonuses. The entries will look similar to this (there are food, commerce, science entries), and there are sets of entries for each of the difficulty levels

# % amount to multiply production cost by per age for ai
AI_MIN_AHEAD_PRODUCTION_COST_ADJUSTMENT 0.7 0.7 0.6 0.6 0.6
AI_MAX_AHEAD_PRODUCTION_COST_ADJUSTMENT 0.7 0.7 0.7 0.7 0.7


You can change it to
# % amount to multiply production cost by per age for ai
AI_MIN_AHEAD_PRODUCTION_COST_ADJUSTMENT 0.7 0.6 0.5 0.4 0.4
AI_MAX_AHEAD_PRODUCTION_COST_ADJUSTMENT 0.7 0.7 0.6 0.6 0.5

This will give later game bonuses to the AI

And yes, I would like to see a mergeciv SLIC...

[Immortal Wombat]

So would I...

re: plagues, maybe keep their effects, but make the rarer?

re: governments SLIC, the only way to enforce a governmental change is via a messagebox button. To avoid it, players would be able to close the messagbox normally, and carry on in the old government. Also, I don't know how this works for the AI. Do they use buttons? I doubt it. But this is less of a problem, as they need the help, right? Or is the AI good enough not to need handicapping now?

[Peter Triggs]

Re: Weak AI Defending

If it's being caused by Frenzy, there's not a lot I can do about it. OTOH, I recently saw something similar to what Lou reported in one of my games and I play without Frenzy. So another possibility is that it's being caused by the fact that we've set the base priorities for GOAL_SEIGE and GOAL_ATTACK to be higher than that of GOAL_DEFEND. In which case the following may help.

First add the following new 'strategy' to the bottom of CRA_strategies.txt:

STRATEGY_DEFEND_AFTER_CAPTURE {

GoalElement { Goal GOAL_DEFEND Priority 860000 MaxEval 2 MaxExec 1 PerCity }

}

The base priority here is set to trump the 850k one for GOAL_ATTACK in STRATEGY_ATTACK, just in case the AIciv is using that. It's the highest of all the GOAL_ATTACK and GOAL_SEIGE base priorities.

Next, add the following to the bottom of any handy CRA_XXX.slc file (e.g., CRA_updater.slc):

int_t DEFEND_AFTER_CAPTURE_COUNTER[];

HandleEvent(BeginTurn) 'InitDefendAfterCaptureCounter' pre {

int_t i;
int_t maxP;

    if (preference("MaxPlayers") < preference("NumPlayers")){
         maxP=preference("NumPlayers");
    }
    else{
         maxP=preference("MaxPlayers");
    }
    for (i=0;i < maxP;i=i+1){
         DEFEND_AFTER_CAPTURE_COUNTER[i]=0; 
    }
    DisableTrigger('InitDefendAfterCaptureCounter');
}


HandleEvent(CaptureCity) 'PreDefendAfterCaptureCounter' pre {

    //whenever an AIciv is about to capture a Human city
    if (!IsHumanPlayer(player[0]) && IsHumanPlayer(city[0].owner )){
         EnableTrigger('SetDefendAfterCaptureCounter');    
    }
}

HandleEvent(CaptureCity) 'SetDefendAfterCaptureCounter' post {

    if (CityIsValid(city[0])){//if it still exists after the capture
         DEFEND_AFTER_CAPTURE_COUNTER[player[0].owner]=7; //how long to prioritize defence
    }
    DisableTrigger('SetDefendAfterCaptureCounter');
}

HandleEvent(NextStrategicState) 'UseDefendAfterCaptureStrategy' pre {

    if (!IsHumanPlayer(player[0]) && DEFEND_AFTER_CAPTURE_COUNTER[player[0].owner]){
         ConsiderStrategicState(player[0],2000,StrategyDB(STRATEGY_DEFEND_AFTER_CAPTURE));
	    DEFEND_AFTER_CAPTURE_COUNTER[player[0].owner]=DEFEND_AFTER_CAPTURE_COUNTER[player[0].owner]-1;
    }
}

The number "7" above is the number of turns after capturing the city that the AIciv will prioritize GOAL_DEFEND. You guys might want to adjust this.


Plagues

I have mixed feeling about the Plagues. I do like them, and they actually serve the purpose of keeping the human player in check in a realistic manner (after all, plagues were common and oftentimes devastating because they could not be controlled.) However, some players have a valid point that they can be frustrating occurances. So striking the balance has been tricky. I don't want the effect to be so weak, that it is merely an annoyance, but I do not want them to be game crushers either.

I like plagues too, but agree they shouldn't be game crushers. One thing that occured to me is that the message you get when struck by a plague might be re-worded. Instead of "... There is nothing to do except hope that God will relent.", I think players might find it preferable to be told that we need to put more resources into medical research or something.

Another idea might be to build on this by adding feats that are connected with the improvements that alleviate plagues. So the first player to build 10 apothecaries gets a +1 happiness bonus for 10 or 15 turns. Likewise for physicians, but maybe a +2 bonus. This should help players feel that they're making progress and gaining control over the situation.

And even if you don't accomplish the feat, it might be an idea to send out a message when you've gained the enabling advance about how our overworked medical staff are grateful, or something like that. Just rewarding the player for making a good decision.


Governments

When a player researches Dark Ages, he will have access to a new type of government (Dark Ages Tyranny) which will parallel Tyranny in terms of science, food, gold and so forth. Basically, it will be a large step down from the other Ancient governments.

If a player gets Dark Ages, the following governments are no longer valid

GOVERNMENT_TYRANNY
GOVERNMENT_DYNASTY
GOVERNMENT_CITY_STATE
GOVERNMENT_OLIGARCHY
GOVERNMENT_REPUBLIC
GOVERNMENT_DICTATORSHIP
GOVERNMENT_TRIBUNAL_EMPIRE

and a player would be bumped from them into (ideally)
GOVERNMENT_ANARCHY
or
GOVERNMENT_TYRANNYDA (my coding designation for Dark Ages Tyranny)

At that point in time, the next government available on the tree will be Theocracy, which will bring a player up to a much better government form.

Aside from what IW mentioned, the only other information I can recall about changing governments is:

Not all parameters can be changed using the ConsiderStrategicState function. The following structure elements are only changed by a SetStrategicState function call: PopAssignmentElement elements, the order of Government elements and SliderElement elements.

So a possibilty might be to define a set of new strategies like:

STRATEGY_MILITARIST_MODERN_GOV {

//
// GOVERNMENT SETTING (MILITARIST)
//

// good for war
Government GOVERNMENT_TECHNOCRACY
Government GOVERNMENT_CORPORATE_REPUBLIC
Government GOVERNMENT_FASCISM
Government GOVERNMENT_COMMUNISM
Government GOVERNMENT_MONARCHY
Government GOVERNMENT_TYRANNYDA
Government GOVERNMENT_ANARCHY
}

and then arrange it so that when the player researches Dark Ages, we set up a call to the SetStrategicState function that loads these strategies. The STRATEGY_DEFAULT_MODERN_GOV would (hopefully) handle the Human. But I've never tried anything like this and for all I know the game might just crash.

[Lou Wigman]

Just a quick one. Putting 'Latifundia' at +10 commerce makes it identical to 'Trading Post' in the unmodded game. Oh well, what comes around goes around!

As for plagues, I've left these turned on for now but I find them a nuisance. They could, however, be used with devastating effect to pull back the human player once they are winning.

The first problem is how do you determine whether in fact the human player is winning by a significant margin. Having resolved that a plague of monumental proportions could be inflicted. Half the population, say. Some cities could even be removed altogether.

If this were to happen then the 'trigger' would have to turn itself off because a second occurance of this magnitude would simply not be cricket.

Peter - Updating units.
There are all sorts of problems with this in 1.32 AI spearman militia persist right through the game. That's not the reason for this post though.

At the beginning of some turns the player gets a message saying that such and such a civ has finished/ is working on a particular wonder. If this message coincides with a new advance then the updater code fails to fire. Hope this helps.

Hex
Instead of calling your dark age government 'Tyranny' why not give it a proper name such as 'Feudalism'.