Tectonics

I haven't looked at climate yet so tundra is nto my fault ;)
As for lowlands, I must tune again to try to get some more hills. What map size do you use? I know there were sometimes too many flatlands in the civ IV version on big maps so I guess the same problem can be present here too.
I'll may be tweaking the starting positions someday in the future but that's not a priority.

The lowlands scale with the mapize. Not a suprise.
The tundra starting is a problem that only affects playing with tectonics. In contintents maps i hardly ever start in tundra, Startingbias enabled or not. So i guess that your (great ;) ) mapscript somehow interfers with the starting positions?
 
Typo?
Or what is it supposed to mean?
Am I missing something?

I remember having a Würm in my Magic the Gathering forest deck...

Würm is the name of the period of prehistoric galaciation circa 30000-10000 BC...

So, your Würm thing is like having a character called Gilgamesh in Final Fantasy...

Though, that Gilgamesh is a legendary King of city-state of Uruk, in Sumeria...
 
I've noticed the tundra start location thing with the latest version as well.
 
Nah, default balancer this time is much more "agnostic" so to speak. It respects what map script has generated and fix things only to return the balance to game, usually by spamming around resources. So other civilizations should also start in similarly bad locations ;)
 
Würm is the name of the period of prehistoric galaciation circa 30000-10000 BC...

So, your Würm thing is like having a character called Gilgamesh in Final Fantasy...

Though, that Gilgamesh is a legendary King of city-state of Uruk, in Sumeria...

Good thing about civfanatics, you learn new stuff :)
Never heard the term before. Seems it is named after a german river. "Würm" derives from germanic and means worm, serpent or similar. So river was named a snakeish word and the glaciation term was taken from the river. Also means the "Magic" name of the big dragon-thingy wasn't completely off :)

Now I'll stop derailing this thread haha.
 
Good thing about civfanatics, you learn new stuff :)
Never heard the term before. Seems it is named after a german river. "Würm" derives from germanic and means worm, serpent or similar. So river was named a snakeish word and the glaciation term was taken from the river. Also means the "Magic" name of the big dragon-thingy wasn't completely off :)

Now I'll stop derailing this thread haha.

The Würm-Kaltzeit has really been named by Penck und Brückner after the river Würm. But its not a general name for glaciation, but for the glaciation of the Alps. In northern europe this period is called "Weichsel-Kaltzeit".

And where we are at Kaltzeiten already. The tundra problem is still standing in the room. I have tried earlier builds of your mapscript and havent got that tundra-start-problem. It appeared in the latest build. So you must have triggered some climatic changes there ;)
 
First of all, I want to say that this script is awesome, far better than all the defaults.

I've played through 5 or 6 games with it.

I have a few comments. I play on normal and small sized maps so that may colour some of my thoughts.. been playing with 60-70% water since that option was added.

  • Starting position always seems to be near the north or south pole, probably what results in the tundra starts people are complaining about.
  • I wonder if the larger size means it needs more luxury resources? Always seem to run short with this script. Maybe it's a distribution problem, also the tundra doesn't seem to have many luxuries. So that could be it.
  • Poles have almost always both been attached to the largest continent in my games. Not really a bad thing but it would be nicer if it was a 50/50 split. But maybe it's just me.
  • There are almost never any small (4-5 tile) islands.
  • EDIT: Almost forgot!, in my most recent save I noticed there was some marsh tiles that had the graphic of a mountain. Very confusing when I saw artillery firing from a mountaintop! Almost ruined my war :) However this could be from one of the other mods I'm playing, or just a Civ5 bug.
 
Would be interesting to see this cript combined with Terra Incognita somehow. I mean add the possibility of forcing a new world and choose wether everyone starts in the old world or anywhere, at least.
 
Starting position always seems to be near the north or south pole, probably what results in the tundra starts people are complaining about.
I didn't touch starting positions yet. I may do so but I must tune some geographic stuff before.
I wonder if the larger size means it needs more luxury resources? Always seem to run short with this script. Maybe it's a distribution problem, also the tundra doesn't seem to have many luxuries. So that could be it.
I can't see why this would have an effect. I use the default resources generation algorithms, so if results seem different, it means these algorithms are somehow faulty. I will probably not look at resources distribution. Not anytime soon anyway.
Poles have almost always both been attached to the largest continent in my games. Not really a bad thing but it would be nicer if it was a 50/50 split. But maybe it's just me.
I've tried to force land masses to be nearer one hemisphere, so one hemisphere would be more likely to have some waterway going all around on 60/70% water maps, but it needs some tweaking.
There are almost never any small (4-5 tile) islands.
Yes. The script can generate 1 tile islands, or with a lot of luck bigger ones, but it's not there yet. I'll try to add this when I manage to get less pangaeas and less continents spanning from north pole to south pole.
Almost forgot!, in my most recent save I noticed there was some marsh tiles that had the graphic of a mountain. Very confusing when I saw artillery firing from a mountaintop! Almost ruined my war :) However this could be from one of the other mods I'm playing, or just a Civ5 bug.
I don't change marsh or such features, so I don't think that's due to the map script. I know I once saw rivers on the strategic map that didn't show on the 3d view. You could check if the hex was correct in the strategic view and not the 3d. If that's the case, then it's definitely a Civ5 bug where they may pick the tile for the 3d before the map generation is finished (like they add rivers after adding starting locations or stuff like that).
 
Would be interesting to see this cript combined with Terra Incognita somehow. I mean add the possibility of forcing a new world and choose wether everyone starts in the old world or anywhere, at least.
Yes, I had a Terra option in Civ IV, I'm planning to code it back in Civ V.
 
I've been seeing tiles placed wrong as well. Fields that have hills when there is no hill, or mnts. when it's really a plain. Stuff like that. It causes the curser to slow down when you try to move units in the area of a bugged tile. I'm wondering if it could also be causing slow down with turns. I've checked theses areas in the strategic veiw and it shows what should really be there. Reloading the game will sometimes correct the problem, it can also revert it back to the wrong tile when you restart the game later.

I see a lot of these bad tills when playing on larger maps, huge and bigger.
I don't see this happening with the default generators.
I'm also getting a lot of maps that will have most all coasts boarderef with mnts. or entire areas inland that will be walled in with only one or two entry ways. These would be cool except that the AI has been confirmed to have trouble navigating these types of choke points.

Would it be possible to add an option to adjust tge amount of mnts. That appear in this way, and instead have it generate hills instead.

I'm thinking of the way tge planet generator for Civ IV gave options to adjust most aspects of how the world would be built. Amount of mountains, hills, water between continents, ets. The planet generator mod was nice because of those options, but your mod always made better land massed.

One last thought.
Would it be possible to have an option to select between the diffirent types of world. To be able to select continent or Pangea, as apposed to having it done randomly during creation?

Keep up the great work. I'm looking forward to the next update.



Posted over an iPhone. Sorry for the errors.
 
The amount of mountains is a parameter, but exposing all the parameters through the current user interface is a bit clunky. I'd have to find a way to enter figures instead of having to choose from a list. There's also a figure which carves some passes through mountain ranges (line 303: if Map.Rand(100, "mountain pass" ) > 90 then ) if you replace the 90 with something lower, there will be less 'walls' of mountains.
As for Pangaea vs. Continent, it's a bit hard to do. A Pangaea is easy, but making sure there are several continents is not. I intend to try to make it so that 60/70% water more consistently give continents and less pangaeas, and then add a pangaea option.
 
v5 has been uploaded and is attached in first post too.
It includes a tuning of plate numbers and sets water in the north and south poles.
This makes for less small lakes and more continents, much less likelihood of a continent spanning from one pole to the other.
 
As for Pangaea vs. Continent, it's a bit hard to do. A Pangaea is easy, but making sure there are several continents is not.

You can borrow code from Continents.lua that does exactly that:

Code:
	local done = false;
	while done == false do
		-- ...map generation code goes here...
		biggest_area = Map.FindBiggestArea(false);
		iNumBiggestAreaTiles = biggest_area:GetNumTiles();
		if iNumBiggestAreaTiles <= iNumTotalLandTiles * 0.58 then
			done = true;
			iBiggestID = biggest_area:GetID();
		end
	end
 
Tectonics was my default map script in civ IV so I am looking forward to seeing it in civ V.

I just installed it and I am getting strange results on a huge map. The river is showing in the strategic view but not in the 3d view.
 
Yes, I saw that and reported this bug on the 2k forum. Considering I don't mess with river generation, I consider it's a bug in Civ, not in the script, but it probably only happens under certain conditions that are not clear to me. I think the river's added late in the map generation process to "beautify" the starting location, and the result is bugged in the same way rivers added at that stage in Civ IV didn't get flood plains in the deserts.

You can borrow code from Continents.lua that does exactly that:
I won't use that. Out of principle, I think that's dead ugly as there is zero guarantee that this program will converge and give a result. Of course, it's very very unlikely that it won't converge in a decent time, but I hate to resort to a "let's roll again until we find a nice result" approach.
 
Ok. I'm reading Bob Thomas' comments in AssignStartingPlots.lua and want to stress this out:
As for start placement and resources, a big part of Jon's vision for Civ5 was to
bring back grandiose terrain, realistically large regions of Desert, Tundra,
Plains, and so on. But this type of map, combined with the old start generation
method, tended to force starts on grassland, and do other things counter to the
vision. So I designed the new system to more accurately divide the map, not by
strict tile count, but by relative worth, trying to give each civ as fair a patch
of land as possible. Where the terrain would be too harsh, we would support it
with Bonus resources, which could now be placed in any quantity needed, thanks to
being untied from the trade system. The regonal division divides the map, then
the classification system identifies each region's dominant terrain type and aims
to give the civ who starts there a flavored environment, complete with a start in
or near that type of terrain, enough Bonus to remove the worst cases of "bad luck",
and a cluster of luxury resources at hand that is appropriate to that region type.
So, basically, if you start out in tundra, it's because Jon Shafer wants you to. Of course, you'll have more resources to get more food, which means you'll have less room to build trading posts, and ultimately get a poor gaming experience.
Now since Sirian also wrote this:
The distribution is handled wholly within this file, here in Lua. Whether you
approve of this change is your prerogative, but it does come with benefits in
the form of greatly increased power over how the resources are placed.
I'll just say that the benefit is illusory. In Civ IV, I could also totally ignore the xml files for resources and hardcode it the way I wanted to. I think the changes are rather bad.
 
Uploaded version 6.
It prevents players from starting isolated on a continent or big island, because I finished a game this way and won despite the game being deity, my being last in absolutely every domain during all the game, but the ai's so bad it can't handle that kind of start at all.
It also allows resources as a custom map option as a side-effect.
 
Ok. I'm reading Bob Thomas' comments in AssignStartingPlots.lua and want to stress this out:

So, basically, if you start out in tundra, it's because Jon Shafer wants you to. Of course, you'll have more resources to get more food, which means you'll have less room to build trading posts, and ultimately get a poor gaming experience.
Now since Sirian also wrote this:

I'll just say that the benefit is illusory. In Civ IV, I could also totally ignore the xml files for resources and hardcode it the way I wanted to. I think the changes are rather bad.

I totally agree with that. He does not use XML, instead he defines ton of data structures and arrays inside lua file (which could have been placed inside XML). Not only that, he leaves no way to extend these structures from other mods in modular way, nor get information about them from map scripts which might want to use the same weights. This reeks of bad coding practice and general laziness.

What I am thinking about is this:
- Starting locations should be determined by map script (because it has to know the best places to start)
- Then, use Djikstra's shortest path algorithm to "grab" nearest tiles near every starting location. Do that for every player at the same time, one tile at a time, randomly. If other player "grabbed" the tile, that means it is not available. Do it until at least one player does not have available tiles.
- The result is this: the number of grabbed tiles is the same for every player, and represents number of tiles nearest to the player's starting location. More, "borders" between players can be easily seen despite mountains and different land shapes. And there will be areas that are not "grabbed" by players, suitable for resources that players can compete for.
- Better still, use this not only for resources and city states, but also for land features, to place yield and production tiles in a way that would not require this extensive "balancing" later on.

Well, at least this is my secret agenda. ;)
 
would be cool if you could get rainshadow effects in ie: Deserts tend to be on the opposite (and inland) side of mountains from grassland / forest. As it is the game only seems to sort terrain types by lattitute (ie: Ice, Snow, Tundra, Plains/Grassland, Desert, Plains/Grassland, Jungle and then in reverse order to the other pole) and doesn't understand the way rainshadow effects work in creating realistic terrain
 
Top Bottom