Converting mods from 1.2 to 1.5 - what needs to be done?

[Epistolary Richard]

I know that with the release of 1.5, a lot of modders (and even some mod-users) are trying to update their mods so they can be used with the latest patch.

What would be great is if those of us who have managed it can share their experience with everyone, to help tide us over the transition.

To start things off, descr_sm_factions

1.5 dsf adds in defining culture, movies and a couple of other criteria that are going to have to be added to any 1.2 mod.

[wlesmana]

Possible re-modding of all new custom skeletons. 1.3 and up introduced swimming in the skeleton's animation repertoir. Although it is disabled in non-BI versions so it MIGHT be possible to just use your old skeletons. HOWEVER, you will have to edit the new berserker skeleton or any new unit you're giving the berserker trait to. The reason is that there's a new attack at the last line in the list of animations for berserkers, used for their new "spin" attack.

The spin attack of the berserker is done by rotating and moving forward the model's root position. This is done by the engine and thus cannot be changed. If you used an old skeleton, your berserker will just spin around without attacking (the attack still registers though, so it just look like he's doing the waltz while the opponents got thrown away).

[Seasoned Alcoholic]

The spin attack of the berserker is done by rotating and moving forward the model's root position. This is done by the engine and thus cannot be changed. If you used an old skeleton, your berserker will just spin around without attacking (the attack still registers though, so it just look like he's doing the waltz while the opponents got thrown away).

I've changed the default skeleton for berserkers to fs_slow_2handed, and what you get is some very nice stabbing, cutting and pulling motions with their weapons.

[EDIT:]Most probably know this, but you need to change ct_carthage (from v1.2) in export_descr_buildings (EDB) to either carthaginian / carthage / numidia, dependent upon the unit / building in question. Same applies to export_descr_unit (EDU).

Also you may need to check the patch you have downloaded. Once I'd downloaded v1.3, I found that certain models (velites, velite gladiators) were corrupt, and had to partially recode them. This was probably because I didn't download RTW v1.3 from TOTAL WAR.COM

[EDIT2:]AFAIK (through testing), you can delete parthia from selected entries in EDB. By selected entries I mean logical ones, EG if eastern culture is already present, you don't need an entry for parthia as well. The game won't crash without this entry - I actually found some units were duplicated in the building browser list because of this, but this is probably due to modding and not present in a vanilla version.

[khelvan]

Is there a problem with "new buildings" in 1.5? If this problem exists for mods porting to 1.5, could someone please explain it to us in depth?

If EB were to move to 1.5, how would this affect us?

[Epistolary Richard]

There was a problem with placing new buildings in descr_strat in 1.4 only - something about the map.rwm file not regenerating properly. AFAIK it didn't effect 1.3 and shouldn't effect 1.5 - though I know that Adherbal was keeping a much closer eye on it for Chivalry.

[Teleklos Archelaou]

If he's still in EB maybe he could tell us.

[Monkwarrior]

I know that with the release of 1.5, a lot of modders (and even some mod-users) are trying to update their mods so they can be used with the latest patch.

What would be great is if those of us who have managed it can share their experience with everyone, to help tide us over the transition.

I've already successfully transformed ITW from 1.2 to 1.5 after a quite long and painful work.
I'll try to get into the process in detail in a sort of tutorial or guide.

For the moment, I think it is important to have useful tools for this type of work. Probably you already know, but just in case...

I'm using synchroX to compare directories.
http://www.clickteam.com/English/synchronX.php
It is very useful to see which files have been changed from one version to another. It is very easy to use and tell you the files present in only one of the two directories and the files more recent in one of the two directories.

The other tool is ExamDiff:
http://www.prestosoft.com/ps.asp?page=edp_examdiff
It allows to compare txt files and find the differences. It is very useful for files with not too much changes in them. You can see which lines have been deleted, which ones have been added or in which lines there are changes.

Cheers.

[PatWest]

There was a problem with placing new buildings in descr_strat in 1.4 only - something about the map.rwm file not regenerating properly. AFAIK it didn't effect 1.3 and shouldn't effect 1.5 - though I know that Adherbal was keeping a much closer eye on it for Chivalry.

I haven't had any problems with map.rwm regenerating in 1.5 after many changes to descr strat and descr regions.

[Lentonius]

I would work through it logically, firstly adding the units into the unit files, then the campaign, then the text files, then testing and tweaking until all 'oddities' are sorted out...

[Seasoned Alcoholic]

This is again obvious, but most units in EDU have had their charge bonus reduced considerably. For example, IIRC the charge bonus for merc sarmatian cavalry in v1.2 was around 17 / 18. This has been cut-down (no pun intended ) considerably in the latest version of RTW to 10, which is a significant drop.

Perhaps someone else can confirm this, but I think units' defence skill has been tweaked so that they are more capable of defending themselves in melee in v1.5 from what they were in v1.2. I haven't got a vanilla version of v1.2 to hand, so cannot confirm this.

[Epistolary Richard]

BTW according to this, as vanilla 1.5 already has a patch_1.pak, we will either need to:

1) Entirely replace the patch_1.pak, so people will have to download vanilla info as well.

2) Your installer will have to unpak the files, replace them and then repak it.

These are both possible, but it's just something to keep in your head.

[irishron2004]

In the descr_culture.txt file found 1.3 and newer, I have noticed while trying few favorite quirks that the Carthaginian culture and Eastern culture are identical line for line. Does this mean we really only have five cultures? PS Don't delete it, game will ctd.

[LestaT]

I am in the process of porting RTR 6.0 gold into RTW 1.5 and so far have finished most 'basic' files. Any more tips ? I didn't quite understand the descr_sm_faction file. I don't see much difference between RTR 6.0 file with the 1.5 file except those added movie things and such in every entry.

Does that mean I can use the 1.5 file instead of porting the RTR file ?

[Seasoned Alcoholic]

I am in the process of porting RTR 6.0 gold into RTW 1.5 and so far have finished most 'basic' files. Any more tips ? I didn't quite understand the descr_sm_faction file. I don't see much difference between RTR 6.0 file with the 1.5 file except those added movie things and such in every entry.

Does that mean I can use the 1.5 file instead of porting the RTR file ?

There have been a couple of changes in this file since the release of RTW v1.0 through to v1.5. Think these included turning on prefers_naval_invasions and such like. Regarding the movie files, well I'd noticed that in v1.2 (I think) that Numidia was using the Barbarian intro movie, but have since changed this so that they now make use of the Carthage one.

[Seasoned Alcoholic]

Another attribute in EDU that has changed is the mount bonus a unit receives when fighting against various mounted units.

This is the type of entry used in EDU in v1.2:

mount_effect     horse +4, chariot +4, camel +4

And here is an entry from v1.5:

stat_pri_attr    light_spear, spear_bonus_4

As you can see, the line reference has changed from mount_effect to stat_pri_attr. Actually, mount_effect has been removed by the looks of things. I haven't tested whether battle results are similar in both versions when each attribute has been applied seperately.

[Teleklos Archelaou]

It would appear as though there are some changes in descr_banners.txt

If you have your own banners, you'll need to remove (in 1.5) part of the line that directs the program to where your banners are. In 1.2, a sample line would read:

standard_texture data/models/textures/standard_julii.tga

But that same line in 1.5 reads:

standard_texture models/textures/standard_julii.tga


So you need to remove "data/" from them. This is **just** for the textures, not for the models. I had all of my banners show up as white till I figured this out.

Also, the textures are grouped by faction, and the roman factions just said "julii", "scipii", etc. for 1.2. In 1.5 they are called "romans_julii", "romans_scipii", etc. It's not a big deal to fix if you compare the two files, an old vanilla one vs. a 1.5 one, in a program like winmerge.

[nikolai1962]

Just a few things that could could give you a problem especially if porting an existing mod to 1.5.

1) descr_ui_buildings.txt

In 1.2 there were two entries for carthage,

carthaginian eastern
carthaginian egyptian

Worked fine in 1.2 for both vanilla and new buildings but 1.5 doesn't like any multiple lookups here if the culture involved is trying to find a *new* building. This may include vanilla buildings that culture had no access to in the vanilla game. Can create same error for other factions if just add a second lookup for that culture.

This error can take three forms:

1) If no map.rwn file then you will get a KTM while trying to start a campaign.
2) If map.rwn exists then you either
a) crash on game start.
b) game loads but all regions have zero base fertility.

The presence or absence of certain graphic files in the carthaginian ui folder switched the a) or b) for me but i never worked out why.


2) descr_strat

1.5 gives a show_err over the win_conditions lines at the top of the 1.2 start file:

win_conditions
land_owner 20 100
land_owner 20 100
land_owner 20 100
land_owner 20 100
land_owner 20 100
land_owner 20 100
land_owner 20 100
land_owner 20 100
end_win_conditions


Only an issue if porting over an existing 1.2 strat file. Easy problem in a way as the show_err tells you the problem sort of but if you've messed with the 1.2 strat files a lot you might be so used to seeing those win conditions lines there that is takes ages to make the connection. (I ending up deleting almost the entire strat file in chunks before it clicked.)



3) name.txt/descr_names (found by teleklos)

In 1.2 the faction headers factions with two word names are in the form "romans julii", "greek cities" etc. In 1.5 they have an underscore instead of a space between the words. This causes a crash on load. Only likely to have this if porting an existing 1.2 modded names file.

[Ciaran]

Another attribute in EDU that has changed is the mount bonus a unit receives when fighting against various mounted units.

This is the type of entry used in EDU in v1.2:

mount_effect     horse +4, chariot +4, camel +4

And here is an entry from v1.5:

stat_pri_attr    light_spear, spear_bonus_4

As you can see, the line reference has changed from mount_effect to stat_pri_attr. Actually, mount_effect has been removed by the looks of things. I haven't tested whether battle results are similar in both versions when each attribute has been applied seperately.

I took a look yesterday, the Mount_effect line still exists, just it´s not used with spear units any more. That´s because the old "spear" attribute of 1.2 was supposedly not working as intended. But take a look at, for example, Peltasts, Elephants, camels etc...
All mounted units still have them (usually negative, giving a malus against camels and elephants, or, if they are those, giving a bonus versus horses and chariots), but there are still some infantry units who retain them, namely javelin-armed ones or the Roman gladiator units.

[LestaT]

I hit the woodwork again. Now it CTD with any error written. All the name files are okay I think.

Is there any other way to search the bug without -show_err ?

This is my project here if anyone interested to help.

[nikolai1962]

One thing with changing name files is you have to make sure any rebel generals with sub-faction have the right names.

e.g if you made the vanilla thrace faction be the arverni i.e another gaulish tribe and gave them the gaul names, then as well as changing the starting thrace generals to have gaul names you need to change the names of any rebel generals that are sub-faction thrace.

[Cheexsta]

Alrighty, I've been helping LestaT with the port for a while now, and the error I'm getting is this:



Now, I've tried all manner of things to fix it, from changing Quintus' name (yes I've checked that the name files are good, including the romans_julii thing) to straight-out deleting all Roman characters except for one so I don't have to worry about family trees. But the game continues to question the man's gender.

One theory that I've been told is that maybe it's a map issue: the show_err report points to the middle of the character's y co-ordinate, so it may be having trouble placing him.

So is there anything that has changed in 1.5 in regards to maps that we need to be aware of?

Edit: I found the problem. descr_characters.txt didn't have the entry for "named character" so it didn't recognise Quintus as being of that character type.

[LestaT]

Ok now we have made a bit progress in our conversion project. Custom battle already works, but loading campaign still CTD. Any ideas which files we should look at ??

[Monkwarrior]

Although it is a lot of work, what I did was to make two different "half-campaigns". With JSGME it is simple to change from one sub-mod to the other one.

In one campaign I used the original RTW maps, but I put there the new buildings, the new units, the new ancillaries and traits.

In the other one, I did in the other way. I used the new maps, but I put the original RTW units, buildings and so on.

In this way the effects of maps and the rest of factors were put apart, and I was able to solve the problems in each of the two half-campaigns.

As units are functional, the problems in the first half-campaign must come from buildings and/or ancillaries-traits-etc.

The problems in the second half-campaign must come from maps or the associated text files (descr_regions, etc).

I didn't find any better method.

[nikolai1962]

I didn't find any better method.

Good idea imo given how there are so many possible ctd's in so many files.

[nikolai1962]

Ok now we have made a bit progress in our conversion project. Custom battle already works, but loading campaign still CTD. Any ideas which files we should look at ??

edit: problem was rtr gold overwriting the 1.5 patch_0.pak with the 1.2 (i guess) patch_0.pak

[LestaT]

Thanks to everyone here, we're basically have a working beta version of the conversion project to port RTR 6.0 Gold to RTW 1.5 .

Here's the link to our forum , http://forums.rometotalrealism.org/i...?showforum=153 and if anyone else have some projects of porting thier mods to RTR 1.5 we'll be happy to help in anyway we can.

Thanks. Cheers.

[nikolai1962]

Miscellaneous 1.5 stuff.

1) You have to increase the number beside Pirates and Brigands in the strat file to reduce them (doh). Zero gets you the most, 50 gets very few.

edit: varies by difficulty level

2) In some cases if a faction starts the game with no navy then it will never build ships.

3) Core-relationship values don't stop the AI factions attacking the rebels first. E.g if you set all core-relationships to 200 (including to slaves) and then set each faction to have one other AI faction that it is 300 or 400 to, then they still go for available rebel settlements first. I guess they calculate each rebel settlement as a mini-faction for military strength or something.

4) Provinces set in the victory_conditions don't automatically influence the direction the AI goes. They *seem* to effect the AI a bit at the start *maybe* but once they are at war with a faction they tend to follow the direction of that war. So if dalmatia is on rome's victory list it seems to make the romans go for it quite quickly. However if gaul gets there first and rome gets into war with gaul then they tend to expand west following the gauls while if macedon gets it first and rome gets into war with macedon then rome expands east.

Giving Rome, Sardinia, as a victory_condition target, setting Rome's core-attitude to carthage more hostile than any other faction and setting prefers_naval_invasions to yes didn't make Rome attack Sardinia throughout an entire game.

5) Map glitchs seem like the biggest problem with the AI. If you see situations where a ton of AI units are all standing round the same area in stacks of one or two units and they stay there for turn after turn (unless round capital) then that area might be due a map tweak. Messana on sicily seems to cause problems. The area around maeotis and the crimea the same. Mountains, river crossings etc any kind of impassable terrain. Gonna try tweaking the map of sardinia/sicily to see if that effects naval landings.

edit: big issue for AI. seems factions often have a primary target and if for some reason the ai can't work out how to get there then it will sit and do nothing for ages.

6) Can use map glitchs deliberately to influence AI. E.g if two adjoining regions have settlements very far apart then the AI won't attack even if they share that border the ghole game. Can make sahara touch the sea seperating lepcis magna and cyrene/libya so the only border region is sahara/libya. Their respective settlements are so far away carthage/egypt don't fight.

7) There's a theory that it was the save/load bug that made the AI get weird protectorates in 1.2. Still sames to happen in 1.5.

edit: weird ones seem to be when one faction has for some reason got a crazy amount of money.

[Lusted]

Just though i'd share some tips i found in relation to 1.5 modding. I have on several occasions now had files that look perfect, and should work, but have not or have caused CTD. The solution to this is remarkable simple. Whilst it may not look like much has change in the layout of the text file, so you think you can use the old one in 1.5, this can often result in problems. The solution is simple, simply copy the contents of the text file into a clean 1.5 version of the file and save it. Its that simple, this has solved many problems for me, as it looks like the text files in 1.5 hld more infomration or something. Hope this helps you guys.