Civilisations
How to add civs
How to make leaderheads
Who's going to be in the scenario? And which of them will be playable? You need to think carefully about the focus of the scenario. What's it about? What sort of things will the player be doing? Could they do this only as one or two civs within the game? Or should all civs be playable?
Any civ that isn't playable is there purely to enhance the experience of the player. That means that if it's not needed you shouldn't put it in. Each civ takes up memory and slows the game down.
The classic example of a scenario where you can play only one civ is
The Rise and Fall of the Roman Empire, where you can play only Rome, and the other civs are there purely as your foils (and they are all cunningly engineered by the scenario setup to behave in certain ways). A good example of one where you can play more or less anyone might be the Middle Ages Conquest, where you can be European, Scandinavian, or Middle Eastern. Obviously each has its own advantages and disadvantages. A game with few playable civs is one where the tech tree, units etc can be tailored around that civ. Moreover, it's much easier to test, as you don't have to play innumerable games as all the different civs. But it can have much less replay value.
AI civs can do various things. They can be there as military opponents, as trading partners, or simply to fill up space on the map. You can have them as rivals for land-grabbing, for resource claiming, or for cultural dominance. The good thing about an AI civ, preplaced on the map, is that you can set things up so it heads a certain way. So in The Rood, Ireland is set up so that it is bound to get a cultural win if the player doesn't do anything about it. It's got cultural buildings preplaced in its cities. The Vikings, meanwhile, have a big island to expand into, but they cannot sail from it until later in the tech tree. The result is a big, powerful civ with masses of very cheap and powerful units, just itching to burst out and attack people in the late game.
It's not always obvious how to specify which civs are to be playable and which are AI-only. Here's how to do it. Open the Scenario Properties window and ensure that in the list of civs, each one is highlighted (meaning that they are in the game). Now go to the Players tab. You see that here you can specify things for "Player 1", "Player 2" and so on. There should be the same number of "Players" here as there are civs, and the idea is that you assign a civ to each Player (with the "Civilization" box that is there). So first you need to scroll through the Players and make sure that each one is playing one of the civs that are in the game. Try to avoid assigning the same civ to different Players as this has strange effects. Now, as you can see, you can specify various things for each Player, one of which is whether they are playable. Tick "Human Player" in the top right if you want it to be playable, or leave unticked if this civ is AI-only. "Starts with Embassies" is basically there to specify whether the civ begins knowing all the other civs.
Important note: when fiddling about with the "Players" and related things in Scenario Properties, it is very easy accidentally to remove a civ entirely from the scenario. Its details are still there under "Civilizations" in the Rules window, but any preplaced cities, units etc are removed from the map. This happens if you un-highlight the civ's name from the list under the "Scenario" tab, or de-associate the civ from a Player number. So when you've made your changes in this window, check the map very carefully and ensure that everyone is where they should be before saving.
I think it's nice to use custom leaderheads for civs, so have a look through the
graphics libraries. There are so many custom LHs now that you are bound to find something you need.
Bear in mind that, even if the scenario covers several eras, you may want to have single-era LHs only. It certainly makes the download smaller. In this case, you might like to make your own pcx files for the LHs, if you are using the "wrong" era, as it were. For example, The Rood uses Bismarck, but in the Ancient Age. Of course, the pcx images for Bismarck show him in the Industrial Age. R8XFT kindly made a new set of pcx images showing him as Ancient. It's actually not hard to do this , just use
Civ3FlcEdit to turn the flc into a storyboard, and cut and paste away.
Now, if you're going for several different playable civs, it's worth thinking a bit about how different they are going to be. To make the game interesting, it would be good if they offer different sorts of challenges (and have different difficulty levels). At its most basic, this is a matter of making one civ more powerful to start with (more cities, for example). No doubt in your period some civs are more powerful than others. You can also assign different starting techs and different traits. It's also good to have different UUs, if possible.
Getting more advanced, you may give different techs to different civs.
Here's a good tutorial on how to do it. You can do this to different degrees.
Anno Domini and
Rhye's of Civilization both have basically the same tech tree for all civs, but there are a few techs that are "flavoured" so only certain civs can have them. This way you can assign different religions to different civs, by making religions techs (Civ IV? Pah! Who needs it?). Or you can do it more thoroughly, as I did in The Desert, where the civs are divided into five groups, each of which has a completely different tech tree. A bit of a gimmick, perhaps, but it adds interest.
The use of such civ-specific techs is one way that you can determine how the AI will act. An AI that has access to some special tech that lets it build an uber-unit will research that tech and build the unit, and become aggressive. But there are a number of other tricks to controlling the AI, some of which we've already hinted at.
The more overpopulated the map, the more aggressive the AI will be. Moreover, the AI will be better disposed to members of its own culture grouping than to others.
The more powerful the AI thinks it is, the more aggressive it will be. You can create super-powerful units that can't move, and preplace them in the AI's core cities, and this will make it more ready to attack other people, even though those units can't be actually mobilised. Watch out, though: if you preplace too many of these, the AI will not build normal units here, even to accompany Settlers, and it will sit there building Settlers that don't go anywhere (because it thinks the immobile super-units should accompany them). Daft AI!
As this suggests, units are a good way to fine-tune the AI. Remember that you can make any given unit unique to a single civ or to a group of them. So you can give one civ an advantage (and make it more aggressive) by specifying that it can build more powerful or cheaper units. You can give an AI more powerful units at one point in the tech tree and weaker ones at another, influencing its power at those times. Here, once again, we see the importance of the tech tree in timing things. Of course, even if its units have the same stats as those of other civs (ie, "flavour" units that don't alter gameplay but are just for looks), I think it's a nice touch.
One important way you can control the AI is with "flavours", which are much misunderstood. There is a whole tab devoted to these in the Rules window. They are solely there to influence AI behaviour, and don't alter any other rules: they have no effect whatsoever on human players.
Flavours can be given to improvements and tech advances. You do this on the windows for those things, and it's fairly self-explanatory. A single improvement or tech can have multiple flavours. You can also assign flavours to civs. You do this on the civ window, and again it's pretty simple.
Now you control it from the "Flavours" tab. The idea is that a civ with a given flavour will be predisposed to go for improvements/techs with that flavour, but you can fine-tune it. For example, say you've given America the "Blue" flavour. And say you've given "Blue" to some improvements and techs, and you also have a "Red" flavour that you've given to others. You want America to build or research only the Blue things and not the Red things. First, highlight "Blue" on the left and "Blue" on the right. In the middle box, tell it that the relationship is 100%. Then highlight "Red" on the left and "Blue" on the right. Make this relationship 0%. This means that any AI civ with the Blue flavour will, when possible, always build or research Blue things, and it will never go for the Red ones (again, this has no effect on human players, who obviously can do whatever they like provided the option is open at all).
The clever thing is that you can fine-tune this. You could make it so that the relationship between Blue civs and Blue techs/improvements is only 75%, and the relationship between Blue civs and Red techs/improvements is 25%. Then, America will generally go for the Blue things, but it might sometimes go for the Red ones.
This system helps to ensure that the AI will behave in ways you want it to. In the Rood, I used this to ensure that the Vikings made a beeline for the tech that allowed them to build Longships. That meant that they always took roughly the same time to get to that tech (without being distracted by alternatives), so their invasions could be fairly well timed.
Governments
Government types are cunning things whose potential is not always recognised either. You will no doubt know that the government used by a civ determines a number of things: how many units can be supported by each city, what the cost is for each unit, whether you get extra commerce on trading tiles, how much war weariness there is, and how much corruption there is. But here are a few things that you might not have thought of.
First, you can use these basic variables in new ways. For example, the Feudalism government that came with Conquests gives you
more unit support for
smaller cities, rather than the other way around as with all other governments. This forces a quite new playing style on the player.
Then there are less frequently used effects. "Xenophobic", if ticked, will mean that cities whose populations are 50% or more foreign will not generate culture. That can be a big problem if you're capturing other people's cities. Then there's "Forced Resettlement", which means that when you switch to this government the population of your cities will drop.
"Rate Cap" can be good too. This specifies (in increments of 10%) the maximum % of income that can be allocated to science. So if you set it to 5, the player cannot set the science slider to higher than 50%. This is another great way of limiting governments.
One thing to bear in mind: the "OFF" button for corruption is broken. A government with this option actually has terrible corruption, not none at all. The best you can get is the "Minimal" option - a totally corruption-free government is not possible (unlike Civ 1!).
You can use governments to flavour the civs a bit more, too. Here's Rambuchan's take on the subject:
Governments can be a very good way of not only distinguishing between the different civs' economies but it's an important and easy way of controlling what they can build also. You can create real distinction through this technique. My experience relates of course to the Mughals scenario and here's what we did.
4 forms of government: Monarchy, Feudalism, Mansab System and Colonialism.
Colonialism is the important one for this example.
We made many improvements in the scenario require this form of government in order to build. This of course meant that we had a whole line of very powerful improvements which only the European Trading Companies could benefit from (auto-producing ships, musketmen, offering propaganda resistance against the Indians with higher cultural ratings etc). Furthermore, if you set these to be auto-producing improvements and the units they auto-prod to be 'unavailable' to anyone then you've got real control and distinction.
Technique is simple:
~ Step One: a) Go to 'Edit Rules > Governments'. Name your government and give it whatever qualities you see fit. b) Make sure that you set its 'Prerequisite' to an unresearchable tech. Very important this last bit.
~ Step Two: Go to 'Scenario Properties > Players'. Select the civs you want to distinguish. Typically they are in a culture group together but necessarily. Then select their 'Government' to the new government you just created.
These are the fundamental steps to take which then allows you freedom to give that culture group whatever special improvements you have in mind.
~ Step Three: Go to 'Edit Rules > Improvements and Wonders'. Select the improvement(s) you want and Select the 'Required Government' to the one you created in step one. This prevents anyone else from building these.
Optional for further distinction and control:
~ Step Four: Go to 'Edit Rules > Units > Available To'. Then make sure that no civ is able to build these units. (You don't have to do this, you can just make the unit available to the civs you want.)
Then you can really tailor what units come out of a culture group operating this form of government. What this allowed us, in practical and in-game terms, was to empower the Euro culture group, despite the fact they started with very few cities and poor land around them.
Text
How to edit Diplomacy
Possibly the most under-used element of modding? No, that would be sounds, but still. I think the pedia is pretty important, if only to add a bit of atmosphere. You can call a civ "Nobatia" if you want, but if the pedia entry doesn't tell you who the hell Nobatia was and what happened to it, it won't mean much to most players.
I think that as a bare minimum, the pedia should tell you what each unit, improvement, and Wonder does (that's just to make the game playable). As an almost bear minimum, it ought to include a little bit of information on what these things actually are or were in real life. Otherwise, why bother to make a mod about a particular period in history at all? One of the best pedias is Rambuchan's for
The Rise and Fall of the Mughals, which includes both historical info and gameplay advice.
Quite apart from the pedia, there's also the dreaded Diplomacy file, not to mention script.txt and labels.txt, which contain most of the text that actually appears in the game. You can edit all this stuff really easily (although the Diplomacy file will take ages if you do it thoroughly). Want to change the text about nuclear meltdowns to make it just a normal fire? Simple , search the script.txt and labels.txt for the standard text, and rewrite it.
The best use of text to make the game flow that I've seen is by Quasidemo's
LOTR: War of the Ring. Here, the pedia, script strings, and other textual elements are all cunningly rewritten and linked to make it really work. He even put quotes from the films in just the right parts. This scenario should be studied by all those whose approach to modding is to bung in masses of graphics and forget the text , this shows that a great and atmospheric scenario can be made with text alone and no custom graphics at all! (Although I'm not really convinced by Smokejaguar as Treebeard, but still, that was a bit of a no-win situation anyway.)
Putting it all together
A good scenario isn't just a map with some civs stuck on it and a tech tree and some units , it should form an organic whole. Now that's not very easy to quantify, but ideally, each element should feed into the whole. Pacing is everything. It should feel that the tech advancement is at about the right speed, that you get upgraded units at approximately the right time, that you always have roughly the right amount of things to choose to build in your cities, and so on. Ideally, players should rarely be forced to build only military units (they should have some kind of improvements left to build). There shouldn't be some unit that is in play for three quarters of the entire game, never getting upgraded, either because the unit upgrade line is sparse or because it's a killer unit. There should be a sense of purpose. Think about the Age of Discovery Conquest. You have to explore the New World and send your settlers out there. You have to research the technologies required to build the improvements that spawn Treasure units. You have to find the resources, link them to your colonies, and then build the improvements. Then you have to carry the units back to your capital. So there you have exploration, settling, building, and research all working together for the same goal. Achieving one element kicks off the next stage. They form an organic whole.
Of course you can have different, unrelated goals for the player. In particular, you can present the player with a choice: go for this way of winning or go for this other way. And there can be sort of "side quests" as well, naturally. But I think the experience is enhanced if you try to think holistically. Think about what the scenario is about, what the "feel" should be, what goals you want the player to be aiming for. And think about how to make the tech tree, improvements, and the rest of it all feed into these goals.
Testing
This is a terribly important part of the process, one that begins more or less as soon as you start modding at all. It goes without saying that every time you add a unit or something in the game you will need to play it and test. You'll be in Debug mode, of course, which you get by going to the Scenario Properties page and ticking the relevant box. This lets you "build" new units and things in the middle of the game, and also watch what the AI is doing at all times (put the Caps Lock key on to speed up its unit movements). But equally importantly, you must test the game balance. If there is particular behaviour you want the AI to engage in, you will have to play over and over again to ensure that it is doing it reliably. You need to see which bits of the game seem to drag and which ones work. You will be doing a lot of tinkering with tech research times, improvement costs, unit stats , all the little details whose cumulative effects are unpredictable. If you can get people to help you beta-test, that is very helpful, but you will need to do lots of testing yourself before it's even at a beta stage.
If you find yourself playing a game in order to check whether something works, and you get your answer but continue playing anyway, this is a very good sign.