• 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 trying to run a conversion (with the same game as earlier, later date)... and I ran into the error "too many CKII nations."

I take it this can arise when you've got some nasty civil wars going on in some part of the world?

Yeah, we don't properly merge them back with their liege. Also, all those mercenary groups and holy orders are considered for conversion. The latter will be fixed next release. In the meanwhile, if you open configuration.txt and change the line
Code:
HRETitle = "e_hre"
to
Code:
HRETitle = ""
you should be able to get around that error (by merging the HRE).
 
Not really my area of expertise, but if I recall correctly (i.e. the wiki is correct) there are three conditions that must all be fulfilled:
- not on same continent
- distance from capital equal to or greater than 250
- no land-connection to capital

The first one can be checked using the province id:s from the continent.txt file in the map folder.

I found another file, positions.txt, in the same folder, that could possibly be used to check the distance. However, I'm not sure if it's got anything to do with "ingame location" or if it's only concerned with how the graphichal representations of things are placed on the map.

For the last one, I've got no idea.

Well, I imagine the second would by pretty easy to test against. I can just assume that's true and see how accurate we are.

The third is kind of tricky. If we had a list of which provinces are adjacent to others, we could search through it. But the only file I can see that would be relevant only lists the available fords. I'm guessing the data is extracted directly from the map. And I'm not really good at algorithms that deal with that. Edit: I take it back. I see how I could extract the data from that file. It's inefficient, but computers are fast, and I really only have to do it once. Edit2: Or I can use adjacencies.bin in the map cache folder. Just gotta decipher their format.
 
Last edited:
Yeah, we don't properly merge them back with their liege. Also, all those mercenary groups and holy orders are considered for conversion. The latter will be fixed next release. In the meanwhile, if you open configuration.txt and change the line
Code:
HRETitle = "e_hre"
to
Code:
HRETitle = ""
you should be able to get around that error (by merging the HRE).

Thanks. I'll just use other games to test conversions until then... the idea of the HRE starting merged for any reason really irks me. Also, my thoughts on the HRE: they should never start merged. If, say, the requirements were Absolute Authority, than ~90% of all converted games would have em' already as a giant über-blob.

Also... how does the converter mod work, exactly? I didn't see any differences, so I'm confident I'm doing something wrong.
 
Thanks. I'll just use other games to test conversions until then... the idea of the HRE starting merged for any reason really irks me. Also, my thoughts on the HRE: they should never start merged. If, say, the requirements were Absolute Authority, than ~90% of all converted games would have em' already as a giant über-blob.

Also... how does the converter mod work, exactly? I didn't see any differences, so I'm confident I'm doing something wrong.

Everything in the mod files directory should go in your mod folder for EU3.
Everything in the converter files directory should replace the equivalent files in the converter directory.

Right now, it just adds some missing cultures and religions to EU3. You'd most likely see it if pagans have survived, or perhaps (CK2) Saxons.
 
Adjacencies.bin Format

I'm posting this in case someone else might find it useful, and it'd be nice to be able to just search for the information rather than figure it out. I'll update it as more bits and pieces become known.

The basic unit of adjacencies.bin is a simple structure I'll call an adjacency, defined in C as
Code:
typedef struct {
	int type;
	int to;
	int via;
	int unknown1;
	int pathX;
	int pathY;
} adjacency;

The structure is 24 bytes in size. For each province there is an integer (4 bytes) that defines how many of these structures follow, then the structures immediately follow. There are no delimiters between provinces. The file begins with province 0, which having no existence has no adjacencies (that is, it starts with four bytes of zeroes). If there is more data after the provinces, I do not yet know (I'm sure I'll find out once I've written code to read this).

More on the structure:
type - the type of connection between the provinces. Possible values are
  • 0 - Normal Connection
  • 1 - Ford
  • 2 - River Crossing
to - the province adjacent to the current province. That province will have an equivalent structure pointing back.
via - for fords, this is the sea province between the connected provinces.
unknown1 - Seems to be proportional to the time to move units between the provinces, but at least one value is way off.
pathX, pathY - if moving units' path is not drawn with a straight line, this is the midpoint the path crosses through. If -1, path is a straight line.

In any case, this is easy enough to parse for our purposes, though it'd be nice to settle the unknowns, as they may contain useful data.


Additional relevant posts:
1 2
 
Last edited:
I'm posting this in case someone else might find it useful, and it'd be nice to be able to just search for the information rather than figure it out. I'll update it as more bits and pieces become known.

That's genuinely useful, for all kinds of things. I never got that far (looked into it briefly for front creation and headquarters positioning in Vic2ToHoI3). I assume the format is similar or identical for all Clauswitz games?

Does it include both land and sea provinces? If so, are adjacencies between land and sea marked? I suppose they could be derived, but if they're directly in adjacencies.bin, that would give us a reliable way to determine port provinces, which I would LOVE.

I would suggest that at least one of the additional values is probably "time to traverse."
 
That's genuinely useful, for all kinds of things. I never got that far (looked into it briefly for front creation and headquarters positioning in Vic2ToHoI3). I assume the format is similar or identical for all Clauswitz games?
I haven't looked at others, but I assume so. Edit: a quick look shows that the V2 structure is 28 bytes in size, and that CK2 doesn't have such a file (or at least not stored in the map folder). HOI3 is hard to tell, because it doesn't have those -1s everywhere making things obvious (-1 in hex [in 2s-compliment notation, which is standard] is 0xFFFFFFFF). Possibly a 20-byte structure, if 13197 is a valid province number (that happens to be adjacent to province 1).

Does it include both land and sea provinces? If so, are adjacencies between land and sea marked? I suppose they could be derived, but if they're directly in adjacencies.bin, that would give us a reliable way to determine port provinces, which I would LOVE.

Both are included, but I don't see any obvious way of telling the difference.

I would suggest that at least one of the additional values is probably "time to traverse."
That was my thought, and it'd likely be unknown1, since that's always present. But it's not the same for both directions, which is not what you'd expect from that value.
 
Last edited:
I didn't mean for my HRE / Vassal post to come across as demanding, I just tend to lay ideas out in a very frank manner as there is less chance of a misunderstanding then. Please do not take these as requirements for the mod, they are just suggestions. I have some experience in game design myself, and I tend to put out a "dream version" of a product, then try and whittle down what is actually possible / feasable through discussion after that. I absolutely love what you have done with the EU3 -> V2 Converter, and am really enthusiastic about the possibilities of this one, and would like to contribute in whatever ways I can, which at this point would primarily be idea development and analysis.

Also I see that you replied to some of this stuff two days ago, but for some reason my browser didn't refresh so I've not been able to get back to you until now :S Great news on the test game though, I am pleased that most of the things listed are working as intended (particularly the Papal issue!)

While I'm inclined to agree, you should provide justification for why the HRE gets the HRE mechanics (I agree that probably no-one else should). CK2 mechanics tend to drive the CK2 HRE in a different direction than the RL one, and there's really nothing that is similar to the Golden Bull in the mechanics.

Personal preference more than anything concrete. I just like the fact that EU3 already has mechanics to handle the Holy Roman Empire, and as a result it makes it a more interesting institution. Additionally the HRE is the only empire which really has the size to make these mechanics worthwhile (at least in your average game). It might be possible to give players the option to define a different empire to use the HRE mechanics in the configuration, this would definitely create alternative scenarios in EU3.

I would contend that CK2 HRE territory should always be EU3 HRE territory. The HRE mechanics in a 1399 start are pretty similar to what I always picture autonomous vassals are intended to be in CK2.

I admit to being a little unsure of how to handle the autonomous vassals level, particularly with regards in how to differentiate between the different Crown Authority levels. Another method might be to have all HRE vassals begin as HRE territories, but start them off with a relations penalty (and thus less likely to help each other out). I think the primary problem is that the "lowest" level of HRE unity in EU3, doesn't seem to scale very well with the lowest CK2 one. At least that is my impression, I always think that the HRE in 1399 is more like Low CA, because there is a centralised power base, a centralised military (the Emperor gets free troop allowance from the other princes, its close enough!) and joint diplomatic status (in the form of a massive defensive military alliance), though they do have the freedom to act outside of it. Another potential solution is to start the HRE with some Imperial Territory (without cores), though this could cause it to implode VERY quickly due to the badboy hits, and with an automatic converter could be a nightmare to balance. If however we can pull that off, then this could certainly be a mechanic to use in one of the CA steps.

I'm of the opinion that vassals should always be vassals. Take France as an example. At the latest point you can start in CK2, it has autonomous vassals (which I understand covers the RL history pretty well), and at the start of EU3, everything in its cored territory is vassals (or held by another kingdom).

Again, im afraid I disagree with the definitions again, though this maybe because of my lack of knowledge of early renaissance France. Purely mechanics-wise, having a Vassal is a big step towards diplo-annexing them, and having them cored just makes this step even easier (especially due to the Reconquest CB). My problem with them all beginning as a vassal, is that nations will begin to cluster far too quickly in my opinion, though I will grant you that all the stages could do with being upped by one (ie drop the current Autonomous stage, and move the others down a step). With Autonomous / Low CA, my impression is those states are more like a confederation rather than actually owing their liege much of anything. I think we should make the road to unification for low CA nations significantly harder, and stripping them of vassal relations is certainly one way of doing this. Remember, to diplo-vassalise someone, you need high relations, a royal marriage and an alliance, so most of these nations will already be 2/3 of the way to that.

PUs should be one of the ways we handle multiple same-level titles held by the same character (the other way being a merging of the titles).

Yup, this works. Nice one!

I wouldn't give any non-CK2 Empire title the EU3 Empire government. They can take the decision on their own if they like, or else get their hands on a CK2 empire. But the other government ideas are interesting. I'll revisit them once I've implemented a few more things and seen how the current government methods are working. It really could be the distinction we need.

Fair enough. I did actually have it that way at first, but though some nations might be large enough to warrant Empire status, but then if that were true then they would likely have adopted an imperial title in CK2, so its really a moot point.

Hmm…possibly. Though if you check the wiki page on governments, it's not really a matter of stronger benefits, but different benefits.

You misunderstand me here, this was primarily to be applied to Counts and Dukes. I don't believe there should be much difference between the two with the exception of size, so either give them the same government, or introduce a new government type which is a clone of the Ducal one, but with lesser bonuses / restrictions to account for the smaller governmental ability. That said, having read how you are handling advisors, this shouldn't really be neccesary, as Dukes will have a pool of better advisors and therefore have the advantage I was trying to create here anyway.


Completely agree with your comments on this.

HRE

The HRE is tricky. Historically, it remained highly decentralized, but in CK2 it tends to evolve to Absolute CA pretty frequently. As well, there's no mechanic that's equivalent to the Golden Bull. So I think HRE conversion should always be somewhat configurable, simply because CK2 doesn't model it very well.

My primary concern with making the HRE too unified / strong at the start, is that it will give HRE players very little to actually do in EU3 except curb-stomp everyone. My personal opinion on this is, that it would make a very boring game indeed. If however this is what people want then that's fine. We also need to consider however, that the AI will often be the HRE player, and if they get absolute CA (with Primogeniture as some have suggested) as often as people claim, then we will be faced with massive Germanic blob most of the time, and this sounds unbalanced to me.

A related comment on Vassals and unification here, there is a reason that the 2nd to last HRE decision is to make all the member states a vassal, because it is a massive step towards total blobbing unity. I strongly feel this is a state of affairs we should hold off as long as possible, or else make it possible only in very rare / exceptional circumstances.

Your points about what constitutes HRE territory vs HRE members is good, likewise how to determine the emperor!

One last thing to consider for the HRE... this is the one model which lets us play with three tiers of control. Member states can have their own vassals. Therefore we could have an Austrian Emperor, with a Brandenburg member state, and Brandenburg's vassal Thuringia. We might be able to play with this if we wanted to add extra depth to one particular player / empire / group / whatever.


I still disagree with you about the vassals under an autonomous state for the reasons given above :p We could certainly bump the vassal status up a CA tier or two, and offset the power of this with the decentralisation slider, this is a good idea. I would still urge that we do not make Autonomous CA have vassals though, but rather start this process with Low or better yet, Medium CA.

In EU3, you can diplo-annex vassals, which could be considered the equivalent of using a combination of these techniques. If we convert CK2 vassals as EU3 nations that are not vassals, then you can only use war or add the step of diplo-vassalization first. And the CK2 mechanic of asking an independent (lower-titled) ruler to become your vassal is fairly equivalent to diplo-vassalization.

This is a good example of why I disagree with you. In CK2, your vassals tend to be Counts / Dukes, or Kingdoms only if you are REALLY powerful. However, in EU those vassal states are usually going to be entire nations (ie Kingdoms). I guess it really comes down to how quickly you want large nations to form in EU3, if they start out with vassals then this process will be quick, I estimate 50 years maximum. If you start them with good relations, in an alliance, with a royal marriage, then it will be significantly longer (roughly 100 years probably?). A good intermediary step is to have them in the Sphere of Influence, from observation AI nations tend to try to diplo-vassalise SoI nations more quickly, probably not least because of the relations bonus this grants.

My proposal as such would be;

Autonomous - Alliance / SoI
Low - Alliance / SoI / Cored
Medium - Vassal
High - Vassal / Cored
Absolute - Unified (with cores remaining)

Or;

Autonomous - Alliance / SoI / Cored
Low - Vassal / Decentralised
Medium - Vassal / Cored / Moderately Decentralised
High - Vassal / Cored
Absolute - Unified (with cores remaining)

Again, I really do love what you are doing with the converter, these are just meant as suggestions and just one player's impression of their "dream" version, though ultimately the content of such is determined by you, and I have absolute faith that it will be great either way :)
 
Don't worry, I didn't interpret your ideas as pushy or demanding. If my tone implied I did, well, I'm not always the best at managing the tone in my writing. I love that you're contributing ideas and discussion, that many people do so is what helps develop a strong design for these converters. And that you've got some game design experience is even more helpful (have you worked on anything I might have heard of?). Anyways, on to further discussion!

I agree with you about the HRE, it should have a very hard time of converting united. I know DasGuntLord01 disagrees, I'd tried to incorporate what I understand of his views into my own.

Regarding vassals, what do you think of my proposal to use more than CA to determine the level of independence for vassals? Beyond that I just like the idea, I suspect whatever system we decide on, we'll be forced to be more restrictive than expected due to limits on the number of tags available (the mod would make this easier, but I don't want it ever to be required). I fear that if we only use CA, we won't have enough flexibility , and we'll end up mostly blobified in the end again. CA would be the most significant factor, of course.

You've gone and convinced me about CK2 vassals not necessarily being EU3 vassals. Alliance + SOI should be a good far end of the continuum. Maybe even just SOI.

However, I'd like cores to be based on de jure territory of the title in question (with over 50% incorporation as de jure also giving a core). They seem to be equivalent mechanics. So vassals as cored/non-cored shouldn't be a part of things.
 
It's unlikely that you've heard of us, we're a tiny indie studio. I did some modding work for Mount and Blade, and formed a studio with some of the team. We released our first title back in April but we're still doing ungodly amounts of marketing to get it noticed; http://zatobo.com/games/. We're now working on our next project though we've not announced details for that yet! Considering those roots, I try to help out those modding projects that I can :)

You've gone and convinced me about CK2 vassals not necessarily being EU3 vassals. Alliance + SOI should be a good far end of the continuum. Maybe even just SOI.

I think you're taking it too far now, heh. Before vassalisation was being used too much, but it definitely has a place in the system, particularly as we do have so few mechanics to play with to make the starting relationships meaningful.

If we have a continuum, we also need to know where to place each vassal on it. I propose the following items to consider: CA, relative hierarchy of titles (a duke will be more independent than a count), similarity of cultures, similarity of religions, relative prestige, liege piety, vassal piety, if the vassal is de jure or just de facto, relative military strength, relative number of holdings.

Ah nice, I must have missed that last paragraph of your vassal concept. Yes, splitting vassals down further within a realm would certainly be interesting. It depends on how complex you want to make the rules here (bearing in mind the more complex, the less predictable the behaviour is likely to be). I've come up with a couple of designs and thrown them all away (must have rewritten this post 5 times now!)

I think the strongest idea however, tied CA to the continuum stages like I outlined in my above post (with a few changes, to be listed), while the additional factors (Hierarchy, Culture, Religion, Military, Holdings, together labelled as "Vassal Loyalty", just so we have a term for it) making each vassal deviate from the continuum by one stage on either side. My reasoning for doing it this way, is that Crown Authority remains the most important factor, and it lends a certain level of "development" towards ultimate unity. Should a nation have poor Vassal Loyalty, but good CA, then they will be slightly behind the curve, while one with poor CA but good relations would be slightly ahead of it, compared to where they should be comparatively.

The downside to this system, is because we are essentially treating each vassal on an individual basis, it gets far more difficult to assign global modifiers (sliders most particularly). This will become especially painful with larger nations as they would take penalties more often than smaller ones. To overcome this, the global mods should be averaged out rather than applied each time.

So, the different levels of the Continuum;

Autonomous Disloyal - Guarantee^ - Autonomous CA AND the vassals hate the liege.
Autonomous - SoI / Alliance* / 2 Decentralisation# - Autonomous CA or the vassals hate the liege
Low. - SoI / Alliance / 1 Decentralisation - Low CA or the vassals dislike the liege
Medium. - SoI / Alliance - Medium CA or vassals are indifferent to the liege
High. - Vassal / 2 Decentralisation - High CA or vassals like the liege
Absolute. - Vassal / 1 Decentralisation - Absolute CA or vassals love the liege
Absolute Loyal - Unity / 2 Decentralisation - Absolute CA AND the vassals love the liege

^ (Yes, going back to the Guarantee, this time however I propose that both the Liege AND the Vassal are Guaranteed. This will automatically call the other into wars, but allow them to refuse, plus giving the relationship bonus. This is purely defensive, Guarantees are not called into offensive battles, unlike an Alliance, I think this perfectly reflects the state of affairs that both nations have a relation with the other, but neither want to commit too much to it.
* (I like the alliance being a part of it, as it gives more of an impression of the SoI being a part of the same entity, they may be an autonomous vassal, but they are a vassal non the less. SoI-ing nations later does have a somewhat different definition; namely that the Sphere is more about blocking other nations from influencing that province but it maintaining the majority of its autonomy.)
# (Decentralisation will only be applied to the Liege, this is to give vassals a little boost. It will be in added to any other factors we decide on to determine decentralisation, perhaps a blanket decentralisation depending on CA? -3 Autonomous, -2 Low, -1 Medium, 0 High, +1 Absolute, this will produce numbers closer to normal EU3 starting sliders. Therefore an Absolute independent with no vassals will start with +1 Centralisation, while one with a vassal will begin with 0.)

I've added a stage on either end of the scale to account for especially good / bad Vassal Loyalty at either end of the spectrum. Also, I have removed the core allocations, as I fully subscribe to your point about de-jure territories having core, regardless of continuum stage. A query about this though; will cores be granted for de-jure titles which no longer exist in CK2? (if for instance Lancaster were absorbed into York, would Lancaster still have cores in EU3?)

Of course we could make the system more complex, however the number of factors we can play with on the EU side limit what these stages actually mean.

You mentioned piety among the factors, but I think this has very little to do with the vassal loyalty / authority of the liege, instead I would use this to determine starting Papal opinion (the thing that determines chance of getting cardinals, I don't usually pay much mind to that aspect so the terminology escapes me). We would need to determine another use for for piety for the non-Catholic nations though.

Also prestige was one of the factors mentioned, however I took this out as I would use this primarily on the global scale (ie the factions starting prestige) rather than apply even more effects to it as that makes this stat more important than it really is.
 
Last edited:
Hey attempting to use this mod and I'm new to EU3 could anyone give a quick explanation on how to use the mod?
 
Hey attempting to use this mod and I'm new to EU3 could anyone give a quick explanation on how to use the mod?

Sure! If you go to your EU3 install directory, there's a folder called 'mod'. Put everything under the 'mod files' directory there. Put everything from the 'converter files' directory in the converter directory. Convert your game. Then, instead of placing the save in '\Europa Universalis III\save games\', place it in '\Europa Universalis III\mod\Converter\save games\'.

Finally, when you launch EU3, the launcher will have options in the drop down menu. Select 'CK2 to EU3 Converter' before starting the game.
 
unknown1 - Seems to be proportional to the time to move units between the provinces, but at least one value is way off.

Could this be a modifier for the movement time rather than the actual time?

Note; I say this without actually having looked at the files in question, it just seemed logical.
 
Well, here's the data I have:

Code:
From	To	Travel Time	Unknown1	Unknown1 / travel time
1	2	12		11616		968
1	4	16		15616		976
1	9	32		31766		992
1	27	18		18166		956
1	1253	12		15866		1322
1	1255	12		5333		444
2	1	12		11316		943
2	3	16		15833		989
2	4	17		16083		946
2	5	14		13150		1153
2	7	14		13500		964
2	1255	12		10000		833
 
Hmm, I see what you mean. Judging by the context, I think movement time remains the most likely (particularly if you look at Russia then some province movement times really vary from days to months).

Next most likely would be supply limits, though why they do not use rounded numbers for that I don't know.

Additionally they could be distance from the originator province (or another defined point, no idea how this is measured though), or the chance of things spreading to the adjacent province, like culture.

Rather less plausible are that the numbers actually refer to other things. I see for instance that "16" is repeated quite often, this could be a function called elsewhere in the code.

Afraid that's all I can come up with at the moment.
 
Oh, another thought that just occurred. Do these adjacent include sea zones as well as land provinces? If so, the larger numbers could be distance from a port to the adjacent sea zone. The distances / movement speeds involved there might be different to those on land.

Backing this theory up, is the fact that the province ID is massively higher than the others, suggesting it could be a different type of province.
 
Oh, another thought that just occurred. Do these adjacent include sea zones as well as land provinces? If so, the larger numbers could be distance from a port to the adjacent sea zone. The distances / movement speeds involved there might be different to those on land.

Backing this theory up, is the fact that the province ID is massively higher than the others, suggesting it could be a different type of province.

Yes, the ones in the 1200 range are sea provinces.

I was thinking maybe the numbers give the supply distance (times some multiplier), but as you can only see the supply distance on sea provinces, and can only see the total instead of the delta between provinces, it's hard to check.

And it would make sense (at least it does in my head) for supply distance and movement speed to be proportional. If you assume a 12-day minimum to board ships as well, then this hypothesis fits. It's just untestable for now.

Edit: While writing code to input data from adjacencies.txt, I also wrote some code to output that data in csv format. I uploaded the resulting file to the Sourceforge file area for those who want to try and figure it all out.
 
Last edited:
The code I just wrote to input the supply modifiers for the various trade goods has to be the ugliest morass of code I've written in awhile.

And sadly, soon to be superseded by the demand modifiers code…
 
But do they work? That has to be the important thing :)

(and its got to be better than any code i've ever written!)