The "Merged non-working" map gets "working"

[Antonov]

Salve omnibus!

I tried tackling the problem with Florin80's enlarged map today. Initially I tried running the game with it, got a CTD as was expected.

I looked at the TGAs dimensions and checked descr_terrain.txt. Everything seemed ok. Then I suspected the map size is going over some unknown until now limits, and returned the "original" ExRM map. Tried enlarging it by 1px just to see what was going to happen. It was working. I was like "hmm...".

I began returning over Florin80s TGAs to try something else (I don't remember what it was going to be) and while looking at Explorer's confirmation window when overwriting files, I noticed that Florn80's "map_features.tga" was of a lot smaller filesize than the file I was replacing it with (the ExRM map_features.tga).

I thought some of Florin80s files might be somehow corrupted. Opened up the original ExRM TGAs and Florin80s TGAs parallel in two instances of ACDSee. I compared the files' properties and saw that all of Florin80s files had a significant difference between "size" and "uncompressed size". With two exceptions -- "map_roughness.tga" and "map_trade_routes.tga". These two files had equal "size" and "uncompressed size" fields, meaning, simply, that they were not compressed when saved.

Next I opened all Florin80s TGAs in Photoshop and saved them over the ones in the "base" dir, without compressing them. Ran the game, everything was running normally.

I'm too lazy to test how/wether it wil run with all files compressed instead of uncompressed. My guess is that either 1) the game crashes because some TGAs are compressed, some not, or 2) the way in which new Photoshop versions compresses TGAs has changed and they are not compatible with the game any more. If anyone's feeling like it, check out which of the two it is.

While I was exiting the game though, I got an error about an Artavazd character that was in an invalid tile: x:348 y:71 I opened descr_strat and played around with his position, looking at other armenian characters' positions for reference.

A question: these x y coords, what are they relative to? Where is the 0 0 point?

Anyway, I tried a few variations and it seemed to work with x: 348 y:70. Then I tried it again with the values that were supposed to give me an error, and this time didn't get an error (?):hmmm: So, I left it the way I found it.

So, someone give me some instructions how and where to upload the files and I'll return you the .zip in working condition.

After I saw the enlarged map I've got to say I didn't really like it The added 12 pixels of height look too artificial and don't blend well with the existing terrain, neither when you look at the TGAs, nor ingame. You can see Florin80 has made efforts to make it fit but it needs more work IMO.

[Quinn Inuit]

Hmmm, the compression might've been my fault when I merged them with mine/PatricianS's. I didn't realize that would be an issue, since I saved the maps in the same way I've always done. Some of the blending is also my fault, as I had to choose a point to merge the two. SE Arabia looked much better in his originals.

The really weird thing here is that the roughness and trade routes maps don't actually do anything. They just need to be present. But I'm glad you got it working!

I recommend using either rapidshare.com or uploaded.to for uploading the files.

The 0,0 point is at the bottom left of the map. For some reason, though, Paintshop inverts the Y axis and adds 1.

[Antonov]

Hmmm, the compression might've been my fault when I merged them with mine/PatricianS's. I didn't realize that would be an issue, since I saved the maps in the same way I've always done. Some of the blending is also my fault, as I had to choose a point to merge the two. SE Arabia looked much better in his originals.

So, I gather that this map is made by you not by Florin80?

I meant the coords which I write in descr_strat.txt, not Psp. I've tried Psp but I use Photoshop regularly. Learned it before I had seen Psp and I was accustomed to it

As for the roughness and trade routes TGAs, I doubt they're even used. Someone from CA should confirm it but my guess is that these files were needed&used in previous (MTW and Shogun) games and in RTW they aren't needed but the .exe still checks if they are there.

The zip: http://rapidshare.com/files/20686697...rged_files.zip

Has anyone here ever tried working with GMT? You can find them over here: http://gmt.soest.hawaii.edu/
It's a bunch of command-line programs that are used to generate raster or vector images out of Gridsets. Gridsets are huge databases containing geographical information, available for free download off the sites of some US geography organizations and NASA. For example this map was generated with GMT and then probably slightly edited in Photoshop to add the colouring (but that's not needed for our purposes).

[Quinn Inuit]

I need to leave in a couple of minutes, but one quick note about the map.

Florin80 did an expanded map, but PatricianS did a redone map. So what I did was work on PatricianS's copy, then try to merge it with Florin's map by copy-pasting. So the expanded files are technically his, but most of the map isn't except for the bottom 17 pixels or so.

[Caligula Caesar]

By the way, I am not sure about the map_trade_routes (I expect that may affect where a lot of trade flows) but map_roughness certaily does do something. The east of the map is rougher that the west. That is why the mountains in the east are a different texture to those in the west. I have tested this; if you reduce the roughness in the east, the mountains turn dark grey, like the alps are. Also, if the roughness is too high, characters can't go through the terrain, no matter which ground_type is used.

[Antonov]

I stare at map_trade_routes and I jolly don't see any trade routes on it :hmmm:

As for map_roughness - I fail to understand what is this "roughness" parameter. We've got a heightmap, so what more does the game need?

[florin80]

Haven`t had the time to read all of the thread. Two things:
1. the map extension was made by Uranos. He deserves the credit. The only thing I did was to update the tga files to take into consideration the changes of ExRM(the map extension not having been made for ExRM originally) up to that date. Qi forgot about this and 6 month or so later when he remembered, some other changes had been made in the meantime to the ExRM files, changes which weren`t taken into consideration when the first update was made by me.
2. secondly, I have not compressed any of the files. I`m careful with that thing. With Gimp though, if you choose save it will compress them. So you need to choose save as and add a number to the name so it will be saved as a different file and not overwrite the original. Still, I remember compressed tga files being used with uncompressed tga files in the past(for rtrpe). A few times they were suspected for one thing or another, but the reality is that the game does load with them, even when some are compressed and others not. I know this through experience.

[Caligula Caesar]

I don't think the roughness is very important. It just affects the visual (maybe the battlemap as well). Perhaps it determines how hilly the terrain is on the battlemap. Anyway, the whiter the terrain is on map_roughness, the rougher it is.

[Antonov]

Hmm, I wonder what made it run then? Could it be that the files are saved slightly differently by Paint Shop Pro, Photoshop and Gimp? Because all I did to make it work was open all the TGAs, then overwrite them by the Save As command from Photoshop (ver. CS3)?

[pannonian]

What is this map extension?

[Antonov]

Imagine the map that comes with ExRM. This one continues south with 12 more pixels of height. You get part of North Sahara and the western tip of the Arabian Peninsula.

[florin80]

The main idea is actually that you get the whole of ptolemaic Egypt as far south as Syene(Aswan). So you could have some nubian/meroitic rebels spawning in the border region and even the native revolt that resulted in the Pharaonic state of Thebes which managed, quite impressively, to hold on from 206 BC to 186 BC. Theoretically this also gives the ptolemaics extra strength since they should receive at least one extra region and have more of a backyard than they have with the vanilla RTR 6 map.

EDIT: btw, the original map loaded fine with my game so it is fully functional. And I can`t think of anything I could have screwed when I ported it to ExRM. It`s just a matter of simple copy pasting.

[Quinn Inuit]

By the way, I am not sure about the map_trade_routes (I expect that may affect where a lot of trade flows) but map_roughness certaily does do something. The east of the map is rougher that the west. That is why the mountains in the east are a different texture to those in the west. I have tested this; if you reduce the roughness in the east, the mountains turn dark grey, like the alps are. Also, if the roughness is too high, characters can't go through the terrain, no matter which ground_type is used.

That's very interesting. I didn't know that.

I stare at map_trade_routes and I jolly don't see any trade routes on it :hmmm:

As for map_roughness - I fail to understand what is this "roughness" parameter. We've got a heightmap, so what more does the game need?

Darned if I know. They may have gotten paid by the file or something.

Haven`t had the time to read all of the thread. Two things:
1. the map extension was made by Uranos. He deserves the credit. The only thing I did was to update the tga files to take into consideration the changes of ExRM(the map extension not having been made for ExRM originally) up to that date. Qi forgot about this and 6 month or so later when he remembered, some other changes had been made in the meantime to the ExRM files, changes which weren`t taken into consideration when the first update was made by me.
2. secondly, I have not compressed any of the files. I`m careful with that thing. With Gimp though, if you choose save it will compress them. So you need to choose save as and add a number to the name so it will be saved as a different file and not overwrite the original. Still, I remember compressed tga files being used with uncompressed tga files in the past(for rtrpe). A few times they were suspected for one thing or another, but the reality is that the game does load with them, even when some are compressed and others not. I know this through experience.

I didn't forget. PatricianS was working on the other map simultaneously, and I was waiting for him to finish up before merging. The only thing I forgot was where I put the files when I was finally ready to merge the two sets. And I freely admit that it was working when you gave it to me and that I broke it somehow.

Hmm, I wonder what made it run then? Could it be that the files are saved slightly differently by Paint Shop Pro, Photoshop and Gimp? Because all I did to make it work was open all the TGAs, then overwrite them by the Save As command from Photoshop (ver. CS3)?

I suppose that's possible. I was editing them with an elderly copy of PSP4 that I picked up in undergrad, but I tried them uncompressed and saved w/the GIMP, too.

The main idea is actually that you get the whole of ptolemaic Egypt as far south as Syene(Aswan). So you could have some nubian/meroitic rebels spawning in the border region and even the native revolt that resulted in the Pharaonic state of Thebes which managed, quite impressively, to hold on from 206 BC to 186 BC. Theoretically this also gives the ptolemaics extra strength since they should receive at least one extra region and have more of a backyard than they have with the vanilla RTR 6 map.

EDIT: btw, the original map loaded fine with my game so it is fully functional. And I can`t think of anything I could have screwed when I ported it to ExRM. It`s just a matter of simple copy pasting.

I think it'll be awesome, and the rebellious south will keep the Ptolies a little off balance now.

[Antonov]

I may be getting predictable with this, but it has always been my main rant against RTW and MTW2: The AI sucks and we have to make it suck at least a little less.

Even if we divide all the provinces historically, if we load the factions with their historical units, even if we knew their order of battle at gamestart -- it doesn't bring much as effect if there's nobody to use these things in historical manner, besides, hopefully, the human player

[Quinn Inuit]

Well, we've made some great strides in army composition balancing. The AI now builds much better armies overall. The strategic AI could certainly use some help, though.

[Antonov]

Well, we've made some great strides in army composition balancing. The AI now builds much better armies overall. The strategic AI could certainly use some help, though.

Agreed. I'll devote some time to experimentation and testing Strategic AI behaviour this weekend. I've got a few ideas to try out.

By the way, nobody answered anything about the Generic Map Tools? Nobody's heared of them? I wonder if we can use these to generate very accurate versions of map_ground_types.tga and map_heights.tga, respectively also map_roughness.tga. I'm looking at this opportunity because the lowest 12 px of the merged map seem too inconsistent with the style of the rest of the map.

My only concern is that GMT would generate maps with the contemporary positioning of the landmasses, which, as far as I am informed, differs from the one that existed about 2300 years ago. But then again, I doubt CA took this into account when making the vanilla strategic map...?

[despot_of_rhodes]

Map balancing/creating etc. isn´t really my stuff, but i´ve got a question:
Is it possible to simulate the blockaded mountain passes during the winter somehow? ( iirc, romans had often to take the second, longer route into Germany, since the Alpinian passes were not really accessable during the winter? ( or was it even half of the year? ).

I allready suppose, that such things would be hardcoded, but i thought, why not ask?

[Antonov]

We can't simulate this by making changes to the strategic map every half year. The strategic map gets assembled from the files it consists of at gamestart.

I wonder though, if we can script the appearance of some object or another -- like an invisible and uncontrollable rebel "unit" that has zero movement points perhaps -- that could stay at the mountain passes. At the next half year the unit will be scripted to disappear and this could provide some solution. I hope I made it comprehensible?

I think that's a good idea, blocking the passes, worth a try.

[Quinn Inuit]

Agreed. I'll devote some time to experimentation and testing Strategic AI behaviour this weekend. I've got a few ideas to try out.

By the way, nobody answered anything about the Generic Map Tools? Nobody's heared of them? I wonder if we can use these to generate very accurate versions of map_ground_types.tga and map_heights.tga, respectively also map_roughness.tga. I'm looking at this opportunity because the lowest 12 px of the merged map seem too inconsistent with the style of the rest of the map.

My only concern is that GMT would generate maps with the contemporary positioning of the landmasses, which, as far as I am informed, differs from the one that existed about 2300 years ago. But then again, I doubt CA took this into account when making the vanilla strategic map...?

I've heard of those, but haven't had time to learn them and redo the whole map. I just worked everything into a rough approximation of reality and called it a day.

We can't simulate this by making changes to the strategic map every half year. The strategic map gets assembled from the files it consists of at gamestart.

I wonder though, if we can script the appearance of some object or another -- like an invisible and uncontrollable rebel "unit" that has zero movement points perhaps -- that could stay at the mountain passes. At the next half year the unit will be scripted to disappear and this could provide some solution. I hope I made it comprehensible?

I think that's a good idea, blocking the passes, worth a try.

I looked and I don't think it's possible to create a 1 movement unit that's not an immovable general. Invisible would be impossible. The ideal might be making a fort appear and then disappear, but I don't think there's a script command for forts.

[Antonov]

I looked and I don't think it's possible to create a 1 movement unit that's not an immovable general. Invisible would be impossible. The ideal might be making a fort appear and then disappear, but I don't think there's a script command for forts.

A little after I posted I wondered about how would we hide the unit banner for the invisible unit, so I'm not that sure either. But what other reason could be? We define a unit in the txt files, give it a fully transparent model and give it traits that prevent it from moving. Then we just add and delete it when and where it's needed. Or there's something I don't take into account?

I've heard of those, but haven't had time to learn them and redo the whole map. I just worked everything into a rough approximation of reality and called it a day.

I'll try to generate a map and then rework it to get the "physical_features" color scheme. The main thing I'm unsure about is, wether the units will be able to walk the map normally if its 3d model becomes too "bumpy" due to the higher detail the GMT map will have...