Guide to the ETW map

[taw]

It's probably regions.esf from some tutorial campaign, I use it for testing because it was smaller.

[PietroMicca]

It's probably regions.esf from some tutorial campaign, I use it for testing because it was smaller.

anyway meaning of those ids is not far from our intuition

[PietroMicca]

take a look here

[Taraphir]

There must be some people with expertise on this forum who can join PietroMicca, taw and uanime5 and help on campaign map research.

[PietroMicca]

What about these ids?

france-0000.xml
8| <u2>387</u2><!-- unknown1 id -->
6787| <u2>32</u2><!-- unknown2 id -->
6788| <u2>34</u2><!-- island id -->

france-0001.xml
8| <u2>388</u2><!-- unknown1 id -->
192| <u2>32</u2><!-- unknown2 id -->
193| <u2>34</u2><!-- island id -->

Now we have a complete explanation for these IDs

for of all two definitions

unexplorable is a land region you can't walk through
impassable is a sea region you can't sail through

Generally it is believed to be the contrary. I will use those definitions inside this post. We will find an agreement later.

now a meaningful example Egypt campaign from NTW

this is the map

Spoiler Alert, click show to read: 

mountains in brown
boundaries in black
rivers and lakes in blue and light blue

in the following map areas using the same ID1 are highlighted with the same color

Spoiler Alert, click show to read: 

So areas with same ID1 are explorable or passable adjoining areas. Each other kind of areas has a different ID1.

Now ID2

Spoiler Alert, click show to read: 

Areas with the same ID2 are adjoining unexplorable areas. 65535 is for explorable areas.

Finally ID3

Spoiler Alert, click show to read: 

ID3 is the same in adjoining impassable areas. 65535 is for passable areas.

[hellfire15]

wow.

[ROYAL SIPHAI]

Haven not been in these forums for ages.

Hows progress going with this is it possible to mod the Campaign Map and add new Regions.

[Taraphir]

Well Gentlemen more than 80000 views of this thead already. There is certainly existing much interest in matters regarding campaign map research. A well deserved compliment to the hard working researchers. Perhaps even some people from Creative Assembly visited this thread in secret. Distressing and also disgraceful that they never even gave the slightest help to the community on Campaign map modifications.

[PietroMicca]

There are 3 IDs for each area, but there are also 3 booleans.

The first is true for land bridges, the third for passable areas. The second is true only for 2 areas of English Channel.



Green and orange areas are part of sea region English Channel. What these 2 areas have in common?

Certainly the 3 IDs: 30 286 65535. So those areas are navigable. But they have also 286 as second ID. This id is typical of blocked mountain areas. Maybe this means that in these parts of french and english coast we can't land?

PS: ID1 is normal. 30 is ID for adjoining passable sea areas in Europe.

[hellfire15]

for fear of sounding like a fool, I will ask the following question: How does the recent knowledge affect what you already know, and how far are you from a fully moddable campaign map?

I feel like such a noob, only cheering you on but doing nothing

[PietroMicca]

research situation

regions - 99,9 % (unknown wind parameter, minor irrilevant doubts about areas IDs)
startpos - 33,5 % (a full mess, there are holes everywhere like in a Gruyère: CAI problems in general and identification problems of IDs here and there)
pathfinding type 1 - 62,5 % (the main problem are flags and their usage; there are also problems with boundaries IDs)
pathfinding type 2 - 0% found a new structure of pathfinding in Peninsular War

this structure uses a different type of long coordinates (no more short coord * 2^20), it has a different header and after a quick look it seems that it doesn't use flags. NTW uses both types of pathfinding, so there's a possibility that pathfinding type 2 is compatible also in ETW. If reading this structure is easier, we will try to use it in ETW.

sea_grids - 70% (main problem is sea_grid_connection)
trade_routes - 99,9% (some uncertainty about calculation of length of routes)
poi - 10% (just an almost complete explanation of code1)

PS: many of startpos problems are less important, because game recalculate each turn startpos data in saved game format. So hybrid modding we will solve the greatest part of our problems. We just need to know enough to make start a new campaign.

[hellfire15]

research situation

We just need to know enough to make start a new campaign.

And it seems you're pretty far along with that! GREAT!

[Taraphir]

Grand work PietroMicca, uanime5 and taw. You deserve the gratitude of the whole community.
Keep your chins up. I´m absolutely sure you will be successful eventually.

[Croshy]

Is there yet any discovery about the Musikk/Effect areas ? I would find it interesting if i could make a musik pack where there is culture specific musik and not theatre specific. For example british/irish/scottish musik on the british islands, special folk and classical music for the german, french, swedish and austrian areas etc.

[taw]

Is there yet any discovery about the Musikk/Effect areas ? I would find it interesting if i could make a musik pack where there is culture specific musik and not theatre specific. For example british/irish/scottish musik on the british islands, special folk and classical music for the german, french, swedish and austrian areas etc.

I tried to decode sound databank but failed in my attempt, and I'm not sure if that would make it possible to do what you want if they didn't implement such functionality.

[The Hedge Knight]

Crux3d made a sound bank decoder: http://www.twcenter.net/forums/showthread.php?t=478907

[Croshy]

but how does ETW know that there shall be another music in America then in Europe?

[General Brewster]

thats in the DB files.

[Croshy]

ah, ok nevermind öl

[PietroMicca]

There's now a new xml format for pathfinding

in particular there's a new format of boundaries

the double <u> tag is replaced by a new one

<ary type="boundaries">
    <boundaries unknown="2266038529 (87110101)" path_id="64" vertex_index="5"/>
    <boundaries unknown="2 (00000002)" path_id="65472" vertex_index="12"/>
</ary>

following a taw's suggestion, I have discovered that old format uses great unsigned integer just because we have never thought that those values are simply hex numbers. So we have split in 2 the second <u> and we have obtain something more meaningful. There aren't complicated formulas to calculate the second <U>. It simply contains information about path ID (same ID same path) and vertex index. vertex_index attribute indicates where the corresponding trait/segment is in pathfinding_areas <u4_ary>. In particular you will find that position indicates by vertex_index is a vertices count that defines how many vertices are present in a trait.

Now the problem is decoding unknown attribute. At this moment this attribute is expressed as a decimal and as hex and it replaces the old first <u> tag.