• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
So I was planning to get heirs to titles worked out next. But to my dismay, it seems that data is not directly stored anywhere. I guess it's all derived data, which means I'll have to derive it. This is going to be a mess…
I took a really quick look at the save file. Heirs isn't something I've been looking at since it's of no interest to my current project, but could
title = "title_prince"
be of use? Looking at setup.log it doesn't seem like it. I guess you'll have a fun time trying to derive that. Looking at my own kingdom of Sicily, I get:
k_sicily=
{
holder=639477
succession=primogeniture
gender=cognatic
law="succ_primogeniture"
law="feudal_contract_2"
law="city_contract_1"
law="temple_contract_1"
law="temple_tax_1"
law="cognatic_succession"
law="feudal_tax_1"
law="investiture_law_1"
law="centralization_3"
law="city_tax_2"
I don't know what level of government you're working on here -- if you're working on kingdoms, it shouldn't be too har... actually... Yeah. Parse each kingdom/duchy/county/empire for their succession laws, identify holder, identify holder offspring. Not entirely sure how elective works, but I suspect that you'll find those entries in each vote holder's personal data... Sooo... for kingdoms I guess elective voters would be counts and dukes (I always switch to primogeniture so I wouldn't know, but I think it works like that). For Seniority you just have to find the oldest member of the dynasty. And for gavelkind you just have to find offspring and divide titles among them.

Actually, that looks really hard. I think I'll go back to my own parser and re-write the data collection methods instead.
 
For succession, I'm going to a series of quick releases, each implementing more successors. I'll need people to do conversions, and for all European nations that have an heir, confirm it is the same as the CK2 heir. And if not, try to figure what I did wrong.

For instance, 0.2A implements Primogeniture heirs, but has at least two errors.

Also, you can currently only see the heirs in the save loading screen. Just hover over the ruler name.
 
Don't forget that tech groups also determine the available unit types, with the slower groups usually having better starting units.

Actually, the muslim countries start with slightly better tech levels (5 versus europe's 3). Looking at the differences between western and eastern units at tech levels 1 to 5, they seem pretty comparable. :) So as long as tech level is converted properly, everything will be good.

And in the 1399 start, there's very little variation in tech level across the European region. I need to see what the tech situation is like in 1453, but I think we should also keep variation low (but not non-existent).

Tech levels from 0 to 5, in the scheme of things, isn't much variation at all, really. Remember, also, that nations get the "neighbor bonus" to tech gain, not to mention another bonus if they're "behind the times". So I don't think we need to worry about it.

The average tech level for an EU3 1453 start is around 8. I was going to recommend not worrying about it, but the truth is that the ROTW (since we are pulling their information for scenario files) might advance more than they should over that 50 odd years relative to Europe. Since relative scores are extremely important, we might be able to doctor something. For example, your tech level might be the average of your CK2 tech (in that category) plus X ammount for every Y years after 1399 rounded down. (Such that the largest tech value possible was 9 or 10).


Finally, only Muslim and up tech groups appear in the European region in 1399 (except for the Timurids), so we should probably only consider those groups.

Quite so. It just means using different cutoffs.
 
I'm gonna go ahead and test 0.2A with my ireland save. Then I might varnish my eyes and have another go at eastern europe mappings.

EDIT: Crashed on selecting save. Investigating.

EDIT: Narrowed it down to something in nation entries.

EDIT: Reading through the save file, I can confirm that I do not have the correct Heir (or dynasty!). My dynasty converted as "Yngling", but I am of the house "Ua Briain". (The Ynglings are those damned Scandiwegians!) My current monarch is quite correct (Oswiu Ua Briain), only my heir is inexplicably wrong.

I have no idea if this would cause a crash, though.
 
Last edited:
I'm gonna go ahead and test 0.2A with my ireland save. Then I might varnish my eyes and have another go at eastern europe mappings.

EDIT: Crashed on selecting save. Investigating.

EDIT: Narrowed it down to something in nation entries.

EDIT: Reading through the save file, I can confirm that I do not have the correct Heir (or dynasty!). My dynasty converted as "Yngling", but I am of the house "Ua Briain". (The Ynglings are those damned Scandiwegians!) My current monarch is quite correct (Oswiu Ua Briain), only my heir is inexplicably wrong.

I have no idea if this would cause a crash, though.

It shouldn't cause a crash, though I left a lot of things out that might make things unstable. I've commited a few changes since that release, including some bits that should allow more nations to have (the wrong) heirs since then, I should proabably do a new mini-release tonight.

Also, is it possible to trace a line of descent to that Yngling in your preconverted game? Through an eldest daughter or a bastard perhaps (the two things I know I don't account for).

Edit: Also, I think the next major release will be when I get all the character stuff done. So I need to finish up succession, place the correct stats, and then generate advisors.

Stats we've discussed before, though I don't think we figured out how to do stats for characters younger than 16 (10-16 when can probably extrapolate from existing ones, but younger than that might be tricky).

Advisors, I've no clue. We need to take a CK2 character and determine which type of advisor they should be, and which level.
 
I got this project building properly, and want to run the tests (test_all_saves.bat)

What should the environment variable IZARC_LOC point to?

The zip/unzip utility IZArc. It's the first one I found with a command line interface.
 
I've uploaded 0.2B. It goes up the family tree from the current ruler to find an heir, if necessary. It always goes to the father, which I'm not sure is correct. It also puts out more data for heirs, and a default government type so you can see heirs from within the game.
 
Also, is it possible to trace a line of descent to that Yngling in your preconverted game? Through an eldest daughter or a bastard perhaps (the two things I know I don't account for).

Bingo, it was the oldest son of my oldest daughter (married in to house Yngling). Well, that's a bit silly, it should just be my oldest son. :p

EDIT: Converted with 0.2B. Didn't crash! :D Well done, whatever it was you did. :p Still have the wrong heir though.

England converted a bit funny. Did we include the many-to-one conversion rules yet?


I converted with the government type "tribal despotism".

EDIT: In fact, everyone has tribal despotism.


EDIT EDIT: Let the game run for a bit: really wish I had a population :p. Double checked CK2: doesn't have a population mechanic! We'll, that's unfortunate... The closest thing is Levy size as one indicator of province population, but since levy size is determined by military structures, this doesn't seem like a good idea. Simply pulling the historical data isn't right, of course. I want to say that we should do a weighting based on total structures (or just economic structures), and use it to rearrange the historical old world population.

I'm also thinking about Legitimacy. The Legitimacy of your CK2 ruler is implicit in how he came to power, and how removed he is from the ruling family (and isn't stored in a single value). How do we derive it? Very difficult.

The only thing I can think of is that you can measure the distance from the "rightful ruler" to the current ruler (by traversing the tree) and the greater the distance the less legitimate you are (by, say, 10 per step), but then there is another difficulty in determining how "legitimate" the previous ruler was. You can, theoretically, apply this method to all rulers on the tree, and make it incremental, but then we are traversing one (potentially) very large tree for each of 200 nations. So... low priority on this one.



Use the parents stats (and/or guardian stats) to "skew" the child's stats. If the child is a newborn, then you might just make then a clone of their parents or their guardian or whatever. Or, make each stat something like Child's_Current_Stat + (age/16 * guardian stat) such that the younger the child is more like his guardian he is, and the older he gets the more like himself he is. I hope that's not too stupid.




Yeah, I have no idea. We might be able to use traits, if we can think up some rules for determining which traits map to which advisor types, but they have very little in common. So... low priority on this one. :p


EDIT 38: Also, we need to have ocean provinces explored eventually.

EDIT 72: Eastern Europe dirtymapped. All mappings dirtycompleted.



Uploaded with ImageShack.us

EDIT69: Whoops, forgot greece. Fixed!
 
Last edited:
Because it'll come up in my replies, here's my current plan for development:

Version 0.3 will have at least rough versions of character items complete: heirs, stats, advisors (even if they just take a default form, I want the basic code created).
Version 0.4 will focus on provinces: culture, population, base tax, manpower, etc.
Beyond that I'm not sure, but we'll have time to find an appropriate group of features to focus on.

Of course, discussion of conversion mechanics can happen at any time, I just need to keep track of things in the second post so we can find the discussions again.
 
Bingo, it was the oldest son of my oldest daughter (married in to house Yngling). Well, that's a bit silly, it should just be my oldest son. :p

EDIT: Converted with 0.2B. Didn't crash! :D Well done, whatever it was you did. :p Still have the wrong heir though.

Gender laws will be 0.2C or 0.2D. I'm itching to get elective successions in place.

Oh, and I'm curious: if succession passes to a cousin or uncle or something, is it always traced through the father of the current ruler? That's how I'm doing it currently.


England converted a bit funny. Did we include the many-to-one conversion rules yet?

Nope, those will probably wait until we revisit title-to-tag mapping.


I converted with the government type "tribal despotism".

EDIT: In fact, everyone has tribal despotism.
Yep, that's the default. It makes your heir appear in-game, and is probably what stopped the crashing (though I don't know why my game was more stable).
 
EDIT EDIT: Let the game run for a bit: really wish I had a population :p. Double checked CK2: doesn't have a population mechanic! We'll, that's unfortunate... The closest thing is Levy size as one indicator of province population, but since levy size is determined by military structures, this doesn't seem like a good idea. Simply pulling the historical data isn't right, of course. I want to say that we should do a weighting based on total structures (or just economic structures), and use it to rearrange the historical old world population.

I was also thinking something of that sort. I'd say that population isn't too important, because it's a minor EU3 mechanic, but for the sake of megacampaigns, we need to address it.
 
I'm also thinking about Legitimacy. The Legitimacy of your CK2 ruler is implicit in how he came to power, and how removed he is from the ruling family (and isn't stored in a single value). How do we derive it? Very difficult.

The only thing I can think of is that you can measure the distance from the "rightful ruler" to the current ruler (by traversing the tree) and the greater the distance the less legitimate you are (by, say, 10 per step), but then there is another difficulty in determining how "legitimate" the previous ruler was. You can, theoretically, apply this method to all rulers on the tree, and make it incremental, but then we are traversing one (potentially) very large tree for each of 200 nations. So... low priority on this one.

I had been thinking that we default to 100, and lower it for a few special cases (not a direct descendent of current ruler, legetimized bastard, etc).
 
Stats:

Use the parents stats (and/or guardian stats) to "skew" the child's stats. If the child is a newborn, then you might just make then a clone of their parents or their guardian or whatever. Or, make each stat something like Child's_Current_Stat + (age/16 * guardian stat) such that the younger the child is more like his guardian he is, and the older he gets the more like himself he is. I hope that's not too stupid.

I like that idea. I'll consider that the way we do it unless someone else jumps in with an objection or better idea.
 
Advisor conversion:

Yeah, I have no idea. We might be able to use traits, if we can think up some rules for determining which traits map to which advisor types, but they have very little in common. So... low priority on this one. :p

Well, I can at least put in some code that will turn everyone into some kind of default advisor. I want to at least have them in there, even if we ignore the details for now.
 
EDIT 38: Also, we need to have ocean provinces explored eventually.

Yep. It's disconcerting seeing only the land around Europe as explored.

EDIT 72: Eastern Europe dirtymapped. All mappings dirtycompleted.



Uploaded with ImageShack.us

EDIT69: Whoops, forgot greece. Fixed!


Hooray! Well done!
 
Advisor conversion:

Well, I can at least put in some code that will turn everyone into some kind of default advisor. I want to at least have them in there, even if we ignore the details for now.

The only thing I can think of is using your CK2 council. These 5 characters are the only characters in your realm that have definite identifiable roles: Chancelor (diplomacy), Martial (military), Steward (economy (taxes)), Spymaster (espionage), and Chaplain (culture/religion). There are potential EU3 advisors for each of these.

Chancellor => Diplomat
Martial => Army Reformer?
Steward => Sherrif? Treasurer?
Spymaster => Spymaster
Court Chaplain => Theologian

Their level will depend on their relevent CK2 skill. If your CK2 chancellor has 0 diplomacy, then he will be level 1. If however he has the greatest dimplomacy level in the world (either THE greatest diplomacy level in the world, or greater than some arbitrary cutoff) then he will be level 6.

As the rest of the characters in your court, well... we'll have to leave them alone, since we're still left deciding which advisor type to convert them to.


This method has a potential issue in that the world will be full of diplomats and army reformers, when a standard EU3 start semi-randomises a large swathe of different advisor types at the beginning of the game. This is probably not a huge problem though, as new advisor will be getting created straight away.
 
Does this work with ck 2 plus
 
The only thing I can think of is using your CK2 council. These 5 characters are the only characters in your realm that have definite identifiable roles: Chancelor (diplomacy), Martial (military), Steward (economy (taxes)), Spymaster (espionage), and Chaplain (culture/religion). There are potential EU3 advisors for each of these.

Chancellor => Diplomat
Martial => Army Reformer?
Steward => Sherrif? Treasurer?
Spymaster => Spymaster
Court Chaplain => Theologian

Or have it randomly decide within a certain "group" for advisors.
For example:

Chancellor => Diplomat/Statesman/(Diplomatic Skill Guy)
Martial => Army Reformer/Grand Captain/(Fort Defense Guy)/(Army Tradition Guy)/Naval Guys(Not for landlocked nations)
Steward => Sherrif/Treasurer/Master of Mint/Banker/Trader/etc.
Spymaster => Spymaster/(Something possibly?)
Court Chaplain => Theologian/Inquisitor/Natural Philosopher/Philosopher


This way it keeps the advisor types spread out.