Topic: Trees and Terrains
king_of_nowhere
Joined: 2014-09-15, 17:35
Posts: 1668  One Elder of Players |
Posted at: 2015-07-30, 00:00
Considering that in late game you always have (or should have) excess of wood, and considering that in any map I've ever seen you always have better terrain for trees, then I still don't see it mattering. I'd rther dismantle a few farms and rebuild them on the barrent ground, while making foresters and woodcutters in their place, than make foresters for a 10% land.
In greenland and desert it's not a problem, it's fairly intuitive which terrains are good for trees. I mean, you can easily figure out that the green terrain is ok, while the yellow terrain is not. For winter and wasteland it is not easy, also because the terrains are not easy to tell apart - especially winter, there are a dozen terrains that are just slightly different variations in patterns of snow, even I am not able to telll them apart by sight. But I'd say, that's the kind of things that comes from experience. I would not want it as a tutorial: if we put it in a tutorial, new players may think they have to learn that in order to play, and that may put them down. on the other hand, mentioning it in passing during other tutorials (something like "not all terrains are equally good for trees; trees won't grow well on barren ground, and not at alll on mountains or deserts") could be good. If someone wants more information, it is available on the website. or should be available.
trees don't grow on the terrain called snow, and I suppose that's as intended; that's the barren terrain of that biome. On the various tundra terrains the trees grow well, and on taiga they grow somewhat (that's a mistake in the naming, by the way: tundra refers to the treeless artic plain, taiga to the boreal forest. but seeing there were half a dozen tundras and only one taiga, i assumed the taiga was the one supposed to be treeless). As for why trees don't grow in your map, the only thing I can think of is that you did not change the values for trees. since the currently coded winter terrains have temperatures around 270, while i put them around 50 to increase the difference and make sure that palms wouldn't grow in it, if you did not change also the preferred values for the trees nothign will grow there anymore. If you did change the values for the trees, then I have no idea why it's not working. Edited: 2015-07-30, 00:01
Top
Quote
|
|
einstein13
Joined: 2013-07-28, 23:01
Posts: 1118 One Elder of Players Location: Poland |
Posted at: 2015-07-30, 00:35
I am sorry to hear that. That is why I will be against changes that break current rules. Sorry. Edited: 2015-07-30, 00:43
einstein13
Top
Quote
|
kaputtnik
Topic Opener
Joined: 2013-02-18, 19:48
Posts: 2448 OS: Archlinux Version: current master One Elder of Players Location: Germany |
Posted at: 2015-07-30, 06:58
Could you please send me you edited file(s)? So i could provide it here and some more players could test this (also with older maps) Fight simulator for Widelands:
Top
Quote
|
|
king_of_nowhere
Joined: 2014-09-15, 17:35
Posts: 1668 One Elder of Players |
Posted at: 2015-07-30, 07:09
Is that only einstein's opinion, or the rest of the people here feel the same? Also, what would be, exactly, those "rules"? Something like, for every biome, main terrain sequence has a yield of at least 90%, and there is at least one buildable terrain lower than 5%? What else? If it's only that, I can make a full test once I have refined the values a bit more, and give one last adjustment. It would not take too much, since it would only involve a handful of terrains. But I would like to hear what other people think about it. As long as it's my opinion against einstein's, we're not going anywhere. I would also like to hear if there are other constrains to tree growth that I should keep in mind. Another point: I was starting to google tree names to see what are their favored conditions, but I realized it's the wrong approach. Most of those names are generic names for tree genus, which include dozens or hundreds of species that can groow anywhere from subpolar mountains to subtropical swamps. So there's not a "right" preferred terrain. But. There were predefined values before the terrains were merged, so every tree has a favored habitat. That is important for old maps, which include campaign maps. Let's suppose that tree A used to grow on meadow. But I made another tree to be fit for meadow, and I made A better for steppe. Someone plays an old map, which has tree A on meadow. IT is supposed to be a forest. Instead, with new values, the trees will die. So, I'd need a list that reported the favored terrain for every tree, so I can try to reproduce a similar behavior to what was before. I know einstein has a list of tree growth percentage on every terrain, but is that list referring to the "correct" values, at least for the intended biome? By that I mean, that list clearly showed palms growing on snow, which is clearly wrong, but are the values for palms growing on desert biome terrains correct? And where can I find that list? I assumed it was linked in einstein' signature, but I can't find it anymore. Anyway, when I have values for how the trees should grow, I can try to make my values compatible with older maps. Not exactly, mind. Calculating values of the parameters that would give exactly the same growth rate as the older builds, while also ensuring that trees don't grow in wildly wrong biomes, is beyond my skill. But I can guarantee that where there was a maple forest there will still be a maple forest. @ kaputtnik. I will send you a mail in a few minutes. Note that the values I used still need refining. For now, they are more a proof-of-cconcept: a human can manually set those values to have the desirred result, and can do it better than a computer. Also, since every tree has a different init file, it will be a bit boring to change every single one. EDIT: mail sent. Edited: 2015-07-30, 07:21
Top
Quote
|
|
einstein13
Joined: 2013-07-28, 23:01
Posts: 1118 One Elder of Players Location: Poland |
Posted at: 2015-07-30, 12:54
That's it. Only this. I don't have anything else what I need from terrains. I name it "full variety of terrain types". So you have some terrain types with 1-5% possibility to grow a tree (highest possibility!), also some in the middle (40-70%) and some very fertile (>90%). Old values were made with build-18 only for greenland (summer): http://student.agh.edu.pl/~rak/widelands/files/ForestersWoodcutters/Velocity2.pdf You can see that there is "mu factor" which is defined as: But those values aren't the untouchable. You can tweak them a bit, but not from 90% -> 1% or 1% -> 90%. I assume that you know what you are doing.
http://student.agh.edu.pl/~rak/widelands/files/TreesModel-explanation/data0.03.pdf But the list is for sure incorrect. We all agree that this behaviour is wrong. Especially for palms on ice And as I know if you make the trees in tutorial (or other maps) not growing in correct place, most of the forests will stay alive. For trees it is much easier to "just live as mature tree" than "grow from a sample". Most of maps include fully grown forests, so it will not be a huge problem. If you know that a forester can grow any other forest (for example spruce instead of maple) it will be ok for me. Mostly Hope I answered all your questions about me einstein13
Top
Quote
|
|
kaputtnik
Topic Opener
Joined: 2013-02-18, 19:48
Posts: 2448 OS: Archlinux Version: current master One Elder of Players Location: Germany |
Posted at: 2015-07-30, 18:27
Hmmm... an opinion... i am not sure. For me it would be fine if we have no dependency of Tree <-> Terrain, or in a very short range so that palm trees do not grow on snowy terrains or similar (a definition of biome<->tree ). Having one (or two) terrains per biome where no tree would grow and on other terrains they grow all equal except the resource Water exist. If water resource is given at a place on the map, the growing could be faster (except the terrains which are really treeless). Doing so would give the player a hint by searching for water by the geologist to find a good place for a forester. The automatic seeding of trees depends on the trees which already exist at this place, independent from type of terrain, except really treeless terrains and the definitions of biomes <->trees. Placing a forester on treeless terrain will let him seed trees, which are found on this map. So this would be a complete different implementation, but i think it's a bit easier to learn and also some kind of logical if we use water resource for speeding up the growing of trees. Also there is no need to learn which type of terrain is good for seeding trees (except the ones where trees never grow). And, something i would like to see: Trees will grow on mountains But we have now the current implementation, and thus we should give him a chance to show what it is good for Old values: SirVer has extracted the old values to tweak the new ones. I don't know how the old values are calculated ... as i understoud it right the old decision of which tree do grow on which terrain was just a comparison of some values. You could find a short explanation of the tweaking here (The whole thread for this ) The old values could be found here. I think the files "terrains.json" and "trees.json" are relevant. Fight simulator for Widelands:
Top
Quote
|
|
king_of_nowhere
Joined: 2014-09-15, 17:35
Posts: 1668 One Elder of Players |
Posted at: 2015-07-31, 05:48
then we share the same objective. Only difference is that i don't feel the need to be very accurate with the numbers, and if instead of 5% it is 7%, or even 10%, it's still acceptable to me. BY the way, I find quite curious that you would sternly disagree to my changes if they brough barren steppe up to 10%, while with the current situation wasteland has no terrain at all below 60% and desert none below 15%. I doubt my values could do worse than that. I wait to hear some feedback from kaputtnik and see what I should change.
Top
Quote
|
|
fk
Joined: 2013-07-30, 21:58
Posts: 151  At home in WL-forums |
Posted at: 2015-07-31, 05:53
kaputtnik wrote:
Currently there is no possibility for water resources on mountain terrain.
Top
Quote
|
|
kaputtnik
Topic Opener
Joined: 2013-02-18, 19:48
Posts: 2448 OS: Archlinux Version: current master One Elder of Players Location: Germany |
Posted at: 2015-07-31, 07:08
Could take some time. I think next week.... Here are the files:
Thanks for your work
Then trees will grow not as fast Fight simulator for Widelands:
Top
Quote
|
|
DragonAtma
Joined: 2014-09-14, 00:54
Posts: 351  Tribe Member |
Posted at: 2015-07-31, 10:56
Sometimes people make mountain rivers by placing a stream of regular water; logically, the edges between those and mountains should count as having water.
Edited: 2015-07-31, 10:59
Top
Quote
|
That is why you do the work (?).
