Topic: Trees and Terrains

This is a photo of thick pyroclastic layers in the vicinity of Weibern, DE, in the Eiffel: https://upload.wikimedia.org/wikipedia/commons/thumb/7/7a/WeibernV.JPG/350px-WeibernV.JPG

The area must be very fertile because it is covered with trees.

Top

Quote

einstein13 wrote:

I don't like the situation where foresters can plant on every type of terrain and trees will grow there. Is there any type of terrain* where trees planted by forester have possibility to grow to the max size less than 2%? (So you need few hours to get some wood for building up forester and woodcutter with barbarians, for example)

*-buildable terrain! I don't need beaches, deserts and mountains. Also lava is excluded ;)

I did not try to calculate percentages. I can tell for sure trees have a low chance of groowing on barren steppe. I would guess for sure a planted tree has less than 10% chance of growing. Less than 2%? I have no idea. However, I made fairly small patches of ground to test - basically, i started with an headquarter in the middle of lava, four peninusular running from it, each one with a biome, and each one made of a mix of all relevant terrains for that biome. each one was fully covered by two foresters in the middle of it, so it wasn''t very big. the patch of barren steppe would have had maybe a dozen nodes in it, maybe less. and once there are two foresters planting only on it - because everywhere else was already full of trees - it is inevitable that eventually the place will have a tree cover. But as I said, maybe i could lower its values from 0.2 to 0.15.

We can argue if the ashes should be treefull or treeless. But the main reason why I started my words for that is that basicly old wasteland was as fertile as greenland. You can compare it with wasteland maps in Build-18. If it is as tree-like as it was before, I will not argue anymore. (But this don't mean that it has to be like before, we can change everything, but only a little!)

I have not played much on wasteland so I have no idea how fertile that was. But as much as i could determine, trees are fine on it. There is igneous rock that is maybe the equivalent of barren steppe, ashes is like mountain meadow or steppe, and on hard ground trees are growing fine. So wasteland is not at rick of being barren.

Also, I gave it even higher fertility values compared to greenland. the actual values are available in one of the files that was posted.

Top

Quote

My test was simply having that small map with small patches of terrain and starting treeless. Build foresters, look what happens. When situation is at equilibrium, stop foresters, look what happens. Nothing more. I can only tell if a certain kind of terrain can grow trees well (it sprouts a forest quickly under the forester's action) or somewhat (it takes more time for the forest to take hold) or very little (there is never a real forest, but there are sparse trees), or even not at all. And I can tell, once the foresters are stopped, if the forest is self-sustaining, or if it will eventually die off.

YEs, I could make elaborate tests like you do, but I feel it's a huge waste of time to get irrelevant information. As both player and mapmaker, the rough division good-mediocre-bad-barren is all I could ever want. I don't need any more information, because the exact rate at which a terrain can grow a tree is influenced by a lot of other factors, even something as minor as making a road and a few buildings in the middle of the placce will change that, and as long as I know that terrain is "good" for trees I have other strategic priorrities than to try too increase the number of trees thhat will grow by a few %. Also, I'm probably going to change some of those values, so spending days trying to judge the exact percentages of growth, only to then change all the values again, seems quite pointless. The test I did took less than one hour, included the mapmaking, and gave me all the information I needed. When you are a lazy person, you learn to get the maximum result with the minimum effort

Now, I can guarantee that unless you place plenty of foresters on it and no woodcutters, barren steppe will remain barren. The chance of a planted tree reaching maturity is certainly less than 20%, and almost certainly less than 10%. If you put enough effort with foresters, you can grow a forest there. After all, you can grow a forest on a barren steppe with enough effort. google "green wall of china" for an example. Also, if you don't keep up with the effort, the forest will eventually die. Again, green wall of china. I can tell you that if you place foresters and woodcutters on that ground, you will get very few trunks, but you will get some. I can tell you that if you use barren steppe for a port space, you won't find a foreste has grown onto it, although you may find a solitary tree if you are really, really unlucky. That's enough for me. You want a percentage? that, I cannot give.

Anyway, if you are looking for extremely barren, but buildable terrains, then the most barren ones are: Desert: desert (there are many terrains called desert, but one of them is buildable). That one is probably the most barren of all buildable terrains. Greenland: barren steppe. Wasteland: igneous rocks. That one is probably the most fertile of the four terrains here described, but still a terrain where a forest will spontaneously die. Winter: snow.

Top

Quote

i don't see what's insufficient in that information for a mapmaker or for the game. You have a barren terrain, you know trees have little channce of growing there. whhether it is 2, 5 or 10%, what does it matter? if you are making a map, you won't place trees there, not and expect them to survive. If you are playing, you will try to avoid foresters planting trees there. Do you really need to know if the percentage is 2% or 5% to decide it is not a good tactic to build your foresters there?

As for the most fertile grounds, in every biome type there is at least one tree that is specialized for the main ground type. generally there are four main ground types (meadow 1 to 4, hard ground 1 to 4, tundra 1 to 4; summer is exception); I assigned to those grounds values of ferility and humidity very close, with a mmaximum difference of 0.1 or 10 degrees (with arbitrary temperatures ranging from 180 to 50). The specialized tree has values exactly in the middle of it, so it is at most wrong by 0.05 and 5 degrees. even with a pickiness of 0.8, it will grow fine. I have every faith that they are above 100%, but even if for some reason it was 90%, would it really make a difference as long as more trees spawn and grow than die? If you are a mapmaker, you know you can plant a forest there, and it will survive and expand. If you are a player, you know it is a good place to make foresters. again, I don't see the purpose of knowing the exact number. I certainly never felt any need to know them when playing, and certainly not for lack of strategical planning.

No, the only information you really need, and only as a mapmaker, is which tree is suited to which terrain. while you can guess that deciduous trees grow in greenland terrain, you have no way of knowing that i picked aspen to be the best tree that grows on meadows, but it is too picky to grow well anywhere else, while birch will grow well on mountain meadow, and it is more adaptable while not being optimal in any terrain. notice that the forester test won't tell you that, since foresters plant a variety of trees. for that, there is nothing but putting a text file somewhere. It needs coontain just the indication I made for myself when planning the values to give the trees: aspen is meadow, specialized, borch is green/cold, adaptable... and I have to rework those after making some research on which are really those trees habitats.

Different trees like any condition of temperature and humidity, but no tree has a preferred fertility lower than 0.4, or a pickyness lower than 0.6. I specifically made no tree adaptable to really low fertility to be able to have barren ground. that gaussian diistribution dictates that on a fertility of 0.24 no tree will have a better than even chance of making it at any step of growth, without taking into account temperature and humidity. and that's multiplied by four stages of growth. however, i don't know how the chances of the tree dieing are calculated, so i can't say how that translates in terms of chances of succesful growth before death happen. anyway, on a fertiility much lower than 0.2 nothing grows. i checked and seen that booth snow and desert have 0.1, and they remain treeless even with foresters planting relentlessly. igneous rocks and barren steppe are a bit more suited for growth, but very difficult nonetheless.

EDIT: also, the test you suggest is flawed, because the forester will eventually run out of space, so you get different results with diferent times. of cours eyou could set a standardized time, but then it would be just a test of a very specific characteristic that has relatively little bearing on how actually fertile the ground is. At this point you are spending a huge amount of time to measure very exactly something whose application is subjected to error and statistical noise. It would be like you must calculate the area of a circle, and you try to use 20 digits of pi, while having a 10% inaccuracy on the radius. it doesn't matter how many digits of pi you get, your measurement won't get any more accurate for it.

Top

Quote

I understand both of your points.

There is one thing i am thinking of: You are both good players and have good experience about the trees <-> terrains. While gameplay you may notice that an area of a map is "good"/"not so good"/"bad" place for a forester. You could difference between the used terrains (soils) in that area because you know the terrains. What about not experienced players? In other words: Could we expect to new players that they have to learn the same level of experience than you have (knowing the growability of trees to each type of terrain)?

I believe we should have a look at the problem from a view of an unexperienced player. Otherwise we ran into something that is only good for really experienced players. [ironic]Or should there be a tutorial "Trees and Terrains"?[/ironic]

From my point of view: I've played widelands a long time (since build 16 i think) more or less, but i never thought about the relationship between trees and terrains.

@king_of_knowhere: I have a map with some Snow terrains from Winter. With your new values trees do not grow. With old values they grow as aspected. Could you please test (and adjust) the values to snow terrains? Feel free to e-mail me if you have new values.

Top

Quote