Guide to the ETW map

[the emerald ingot]

In the grid_data-0003, i altered the unknown2 values for the two halves of the slot for Munster, in the Rhineland region.
this is the new line.
<boundaries passable_part="134 (of 255)" unknown2="65568 (10020)" path_type="7 (slot)" path_id="57 (rhineland)" vertex_index="230875"...
<boundaries passable_part="13 (of 255)" unknown2="532480 (82000)" path_type="7 (slot)" path_id="57 (rhineland)" vertex_index="230861"...

hex values changed from 82100 to 82000 and 10d20 to 10020. of course, these hex values were changed into the decimal format as well.

I've attached screenshots of the battles.

As you can see, there is no noticeable difference in terrain, and the few buildings on the map are the same. The only real difference I noticed was the 1st change battle was foggier, but that could be attributed to randomized weather... except that the battle was tested in darthmod and land battle weather was turned off. So, maybe it was a slight difference, but I don't know. Further testing is required. Do you guys have any particular unknown2 variables you'd like me to test out? Otherwise I'll probably just try making each hex value '0' in sequence from left to right and see what that changes. Let me know if this was helpful at all...

[uanime5]

In the grid_data-0003, i altered the unknown2 values for the two halves of the slot for Munster, in the Rhineland region.
this is the new line.
<boundaries passable_part="134 (of 255)" unknown2="65568 (10020)" path_type="7 (slot)" path_id="57 (rhineland)" vertex_index="230875"...
<boundaries passable_part="13 (of 255)" unknown2="532480 (82000)" path_type="7 (slot)" path_id="57 (rhineland)" vertex_index="230861"...

hex values changed from 82100 to 82000 and 10d20 to 10020. of course, these hex values were changed into the decimal format as well.

I've attached screenshots of the battles.

As you can see, there is no noticeable difference in terrain, and the few buildings on the map are the same. The only real difference I noticed was the 1st change battle was foggier, but that could be attributed to randomized weather... except that the battle was tested in darthmod and land battle weather was turned off. So, maybe it was a slight difference, but I don't know. Further testing is required. Do you guys have any particular unknown2 variables you'd like me to test out? Otherwise I'll probably just try making each hex value '0' in sequence from left to right and see what that changes. Let me know if this was helpful at all...

This is probably the best way to figure out what the unknown2 is for. I'd recommend that you find entries with high hex values, such as 8 or higher, and reduce them to 0 to see what effect it has in game.

To ensure your testing is accurate I'd recommend you do the following:
1) Play a battle on the grid_cell you want to change several times.
2) Play a battle on the grid_cell you've modified several times.
3) Repeat several times with a different grid_cell you've modified the same way.

1 is the control test and is performed several times to prevent random events distorting the results. 2 is to ensure that the change you've noticed always occurs. 3 is to test whether the same change will occur in other places. So if Munster is not foggy before the edit, always foggy after the edit, and other towns become foggy when you edit them the same way then you have strong evidence that this unknown2 controls how foggy the battle is.

As we don't know what any of the unknown2 values are for you may as well test them all. Though the middle one is different to the other 4 because it goes from 0-e, rather than 0-c, so I'd recommend testing this one first.

[PietroMicca]

In the grid_data-0003, i altered the unknown2 values for the two halves of the slot for Munster, in the Rhineland region.
this is the new line.
<boundaries passable_part="134 (of 255)" unknown2="65568 (10020)" path_type="7 (slot)" path_id="57 (rhineland)" vertex_index="230875"...
<boundaries passable_part="13 (of 255)" unknown2="532480 (82000)" path_type="7 (slot)" path_id="57 (rhineland)" vertex_index="230861"...

hex values changed from 82100 to 82000 and 10d20 to 10020. of course, these hex values were changed into the decimal format as well.

I've attached screenshots of the battles.

As you can see, there is no noticeable difference in terrain, and the few buildings on the map are the same. The only real difference I noticed was the 1st change battle was foggier, but that could be attributed to randomized weather... except that the battle was tested in darthmod and land battle weather was turned off. So, maybe it was a slight difference, but I don't know. Further testing is required. Do you guys have any particular unknown2 variables you'd like me to test out? Otherwise I'll probably just try making each hex value '0' in sequence from left to right and see what that changes. Let me know if this was helpful at all...

This is finally a great help

I just suggest you to change a larger number of boundaries. Change just 2 or 3 entries could be about just a tiny area. Change all middle hex in 0 for an entire region (path_id) for instance. Then change in 1, 2 and so on.

[the emerald ingot]

Ok. When im off work for the day, I will first fight several battles around the Rhineland region as a control. Then, I will change all the middle hex digits in Rhineland to 0 and then fight those battles in the same places. I think I will fight 5 battles each in 5 different spots as a control, and then fight 5 more battles in each of the 5 spots with each change I make. This will give a solid control and solid data to compare it against. Then, I will post my findings here for you guys to go over. However, I think we need at least two other people performing the same tests, so with each pathfinding.esf I repack after making changes ( assuming it doesn't cause a ctd) I will upload here so that volunteers can test battles in the Rhineland, take screenshots, and post their findings here as well. That will help us get more data, and much faster than if I do it alone. I work 14 hrs/day 7 days/week, so my free time is limited. Hopefully, I will be able to get a baseline control run and the first altered pathfinding.esf done and uploaded tonight. Sorry for any mistakes on this post, typing on my phone is hard. TT_TT.

[PietroMicca]

I don't know if we will notice something different on the battle map

Maybe in Campaign Map. We should not only try battles, but try to move army throughout region in order to find difference in path, speed or things like that.

[the emerald ingot]

I don't know if will notice something different on the battle map

Maybe in Campaign Map. We should not only try battles, but try to move army throughout region in order to find difference in path, speed or things like that.

I agree. If passable_part is not the only thing that affects movement on the campaign map, unknown2 may have something to do with it. That's the problem, we just don't know exactly what it does. So, I agree that we need to try many different things, not just the battle maps.

Once I get the time to make the changes and test it out, I'll play around with my army, march them all over the Rhineland, raid all the towns, and take the regional capital. Vive la France! lol.

[T.C.]

try to move army throughout region in order to find difference in path, speed or things like that.

Don't expect to find a difference in speed. That's defined by a DB table, and I can guarantee this as I have edited campaign ground type move speeds to make "going off road" a very unwise decision in one of my mods.

[the emerald ingot]

Just uploading the pictures for a baseline measurement, or a control. This is with the original, unmodified pathfinding.esf.

Including the battle pictures I posted earlier, i've played 5 battles in this same spot. All 5 have the same weather, and are lacking the foggy look from the modified battle. Also, I included a picture of the area the game says I can move my army, as well as the farthest I was able to go in 1 turn. After I modify all the middle hex values for the Rhineland and pack them back up, I'm going to try some battles in the same spot, as well as seeing if the movement area and exactly how far I can go changes any. I'm working on that right now. Once I'm done changing things and I've gotten it packed back up into a .esf, I'll upload that file here so that others can try it out, perhaps correlate this research. Call it a peer review. ^_^

[PietroMicca]

Don't expect to find a difference in speed. That's defined by a DB table, and I can guarantee this as I have edited campaign ground type move speeds to make "going off road" a very unwise decision in one of my mods.

how did you mod ground type?

[the emerald ingot]

[QUOTE=The Hedge Knight;11477912]

Some things might be relatively easier even without understanding the whole map:

• making travel through vast wilderness faster

campaign_ground_types - version 1
Might have been thought of already but check the db under campaign_ground_types tables. That isn't a height map modifier, its a movement modifier.

probably through the method mentioned above.

[T.C.]

how did you mod ground type?

Campaign_ground_type_tables.
First column is the movement modifier, next 3 columns define ground abilities. Respectively: Can ambush, Causes heat attrition (NTW), Causes cold attrition (NTW).

[PietroMicca]

Campaign_ground_type_tables.
First column is the movement modifier, next 3 columns define ground abilities. Respectively: Can ambush, Causes heat attrition (NTW), Causes cold attrition (NTW).

in this table there's only a movement modifier, that is coefficient for maximum speed

there should be somewhere a coefficient that makes roads faster than other possible routes.

[T.C.]

in this table there's only a movement modifier, that is coefficient for maximum speed

there should be somewhere a coefficient that makes roads faster than other possible routes.

It's partially hardcoded. Look in campaign_variables_tables, you will see entries for "road_level_0_action_point_cost", "road_level_1_action_point_cost" etc

[taw]

It's partially hardcoded. Look in campaign_variables_tables, you will see entries for "road_level_0_action_point_cost", "road_level_1_action_point_cost" etc

And in case anybody's wondering - these definitely work as expected.

[T.C.]

And in case anybody's wondering - these definitely work as expected.

As said earlier, I can also back that up. I've done extensive work with these modifiers.

[the emerald ingot]

Ok, guys, I've run the first experiment. Here's what I've got.

In grid_data-0003, I changed all the middle hex values to zero for the Rhineland region. I saved the .xml and packed it all back up into pathfinding.esf. Loaded up Empire and messed around for a while. No change in the ability to move in the region, and all but the first battle I played had foggy weather. I even assaulted the regional capital, and it loaded into a fort battle, which darthmod (supposedly) gets rid of. So... not exactly sure what happened there, but maybe the reason why the first battle didn't have fog was because darthmod was preventing weather? I dunno for certain.

I've attached the pictures below, as well as the new pathfinding.esf. If anyone's got some extra time, replace your regular pathfinding.esf with this file and try fighting some battles in Rhineland to see if you get the same foggy effects on all the battles there, like I did. I hope that this wasn't just a fluke on my computer.

By the way, I figured it out; if you place the pathfinding.esf in your empire total war\data\campaign_maps\global_map folder first, and then start up the DMUC launcher, apparently there's something about when it runs scripts that causes the effects of my modified pathfinding.esf to not work. So, if you use DMUC, open it up and select the campaign like you usually do, but before you actually launch ETW, replace the pathfinding.esf with my modified one.

[PietroMicca]

It's partially hardcoded. Look in campaign_variables_tables, you will see entries for "road_level_0_action_point_cost", "road_level_1_action_point_cost" etc

there are other interesting variables

minimum_population
pop_growth_for_spawn
recruitment_population_cost (unused)
....

[taw]

A lot of population-related stuff is hardcoded and the only way to fix it is by rescaling populations with rescale_populations script in esfxml.

[T.C.]

Ok, guys, I've run the first experiment. Here's what I've got.

In grid_data-0003, I changed all the middle hex values to zero for the Rhineland region. I saved the .xml and packed it all back up into pathfinding.esf. Loaded up Empire and messed around for a while. No change in the ability to move in the region, and all but the first battle I played had foggy weather. I even assaulted the regional capital, and it loaded into a fort battle, which darthmod (supposedly) gets rid of. So... not exactly sure what happened there, but maybe the reason why the first battle didn't have fog was because darthmod was preventing weather? I dunno for certain.

I've attached the pictures below, as well as the new pathfinding.esf. If anyone's got some extra time, replace your regular pathfinding.esf with this file and try fighting some battles in Rhineland to see if you get the same foggy effects on all the battles there, like I did. I hope that this wasn't just a fluke on my computer.

By the way, I figured it out; if you place the pathfinding.esf in your empire total war\data\campaign_maps\global_map folder first, and then start up the DMUC launcher, apparently there's something about when it runs scripts that causes the effects of my modified pathfinding.esf to not work. So, if you use DMUC, open it up and select the campaign like you usually do, but before you actually launch ETW, replace the pathfinding.esf with my modified one.

I'd imagine in this context it would make a lot of sense to disable all mods. Your results can't be verified by anyone else, given the different configuration enabled by DM. And all of your tests could be proving pointless due to the contents of your mods.

Good work anyway

[PietroMicca]

A lot of population-related stuff is hardcoded and the only way to fix it is by rescaling populations with rescale_populations script in esfxml.

I'm interested in how towns spawn. Hardcoded?