teppo
Joined: 2012-01-30, 08:42
Posts: 423
Tribe Member
|
Posted at: 2017-12-13, 18:29
WorldSavior wrote:
Why not leaving the crops how they are and simplifying the supercomplicated tree growth model? A new building can make the game more complicated and slow
I do not quite understand. What is the benefit of reworking the tree algorithm?
The tree-growth thing is quite complicated internally. In the end all reduces to a simple "fertility-like" number: Trees grow well, poorly or not at all. The complication is there just to choose suitable-looking trees for each terrain type.
In the old days, we had blackland, greenland, and iceland (or whatever were they called). They each had their own trees that looked different. This system was made so that blackland-ish trees continue showing up mostly on blackland-ish terrain, and so on, and while also keeping tree growth lower on wasteland.
The four-tuples should be hidden from the end user, and replaced with a simple scale (from grows poorly to grows well).
If there is some real-world benefit for rethinking the trees, that is okay. This just sounds like spending a lot of time on something without reason.
Top
Quote
|
|
|
kaputtnik
Joined: 2013-02-18, 19:48
Posts: 2439
OS: Archlinux
Version: current master
One Elder of Players
Location: Germany
|
Posted at: 2017-12-13, 19:14
teppo wrote:
If there is some real-world benefit for rethinking the trees, that is okay. This just sounds like spending a lot of time on something without reason.
The main problem is that there is no representation of 'possibility to grow' on the map:
- A map maker have to struggle with the help and visually compare terrains on the map with the one in the help and then find some trees that may grow on that terrain.
- A player can't see where to place a forester for good log harvest, resulting e.g. in bug reports. Especially for beginners this could be frustrating.
As far i understand those problems couldn't be solved with the current implementation because 'possibility to grow' couldn't be calculated for a node when adjacent terrains are of different types.
Edited: 2017-12-13, 19:15
Fight simulator for Widelands:
https://wide-fighter.netlify.app/
Top
Quote
|
|
|
einstein13
Topic Opener
Joined: 2013-07-28, 23:01
Posts: 1118
One Elder of Players
Location: Poland
|
Posted at: 2017-12-13, 21:37
About geologist new job:
There are always several solutions for the same problem. For geologist we can provide combined messages / two messages per one node.
Second solution is to make REAL 2 JOBS for the same geologist. You can send him to find resources OR send him to find "fertility" for trees (& crops?).
This should solve general problem with terrain affinity for some cases.
And to answer @WorldSavior discussion:
I don't think that Widelands should be simplified. Even more: I like to make it even more complicated in the background. But also I have to think about impacts (what can go wrong) and how to solve it and make more easier to start with.
einstein13
calculations & maps packages: http://wuatek.no-ip.org/~rak/widelands/
backup website files: http://kartezjusz.ddns.net/upload/widelands/
Top
Quote
|
|
|
GunChleoc
Joined: 2013-10-07, 14:56
Posts: 3324
One Elder of Players
Location: RenderedRect
|
Posted at: 2017-12-14, 08:35
kaputtnik wrote:
As far i understand those problems couldn't be solved with the current implementation because 'possibility to grow' couldn't be calculated for a node when adjacent terrains are of different types.
Not true. It is calculated in-game for the actual tree growth, and I already pulled out a function a while ago to make reuse more easy:
probability_to_grow(const TerrainAffinity& immovable_affinity,
const FCoords& fcoords,
const Map& map,
const DescriptionMaintainer<TerrainDescription>& terrains);
This will calculate the probability to grow for any node on the map.
What we need is somebody to program the tooltips. At the moment, that person is not me, since I have a stack of critical branches waiting.
Busy indexing nil values
Top
Quote
|
|
|
WorldSavior
Joined: 2016-10-15, 03:10
Posts: 2094
OS: Linux
Version: Recent tournament version
One Elder of Players
Location: Germany
|
Posted at: 2017-12-14, 16:25
kaputtnik wrote:
The main focus of widelands is building and managing a good economy. Ware fare is just one part of it, not the main goal.
Widelands doesn't even have a peaceful mode. Almost every match is about war, especially tournament matches. Mostly the economy aims for war.
teppo wrote:
WorldSavior wrote:
Why not leaving the crops how they are and simplifying the supercomplicated tree growth model? A new building can make the game more complicated and slow
I do not quite understand. What is the benefit of reworking the tree algorithm?
It will solve a lot of problems, for example:
-
It's hard to know how well trees grow, especially for beginners
-
Trees grow poorly at coasts
-
Mixed terrains in general are very complicated for trees
-
Foresters don't always walk to the best places
-
Foresters don't always use the best tree.
The tree-growth thing is quite complicated internally. In the end all reduces to a simple "fertility-like" number: Trees grow well, poorly or not at all. The complication is there just to choose suitable-looking trees for each terrain type.
In the old days, we had blackland, greenland, and iceland (or whatever were they called). They each had their own trees that looked different. This system was made so that blackland-ish trees continue showing up mostly on blackland-ish terrain, and so on, and while also keeping tree growth lower on wasteland.
Sure? Hard Ground 4 and the Wasteland Forested Mountains have trees with 100% growth chance.
The four-tuples should be hidden from the end user, and replaced with a simple scale (from grows poorly to grows well).
If there is some real-world benefit for rethinking the trees, that is okay. This just sounds like spending a lot of time on something without reason.
That has already happened - by implementing this tree growth model 
einstein13 wrote:
About geologist new job:
There are always several solutions for the same problem. For geologist we can provide combined messages / two messages per one node.
Second solution is to make REAL 2 JOBS for the same geologist.
Geologists already have two jobs: Checking out mining areas and checking out areas which can have well-water
You can send him to find resources OR send him to find "fertility" for trees (& crops?).
This should solve general problem with terrain affinity for some cases.
And to answer @WorldSavior discussion:
I don't think that Widelands should be simplified. Even more: I like to make it even more complicated in the background. But also I have to think about impacts (what can go wrong) and how to solve it and make more easier to start with.
I think it's just wrong to change the old tribes that much. Why not putting your ideas into new tribes instead?
Wanted to save the world, then I got widetracked
Top
Quote
|
|
|
einstein13
Topic Opener
Joined: 2013-07-28, 23:01
Posts: 1118
One Elder of Players
Location: Poland
|
Posted at: 2017-12-15, 13:35
WorldSavior wrote:
kaputtnik wrote:
The main focus of widelands is building and managing a good economy. Ware fare is just one part of it, not the main goal.
Widelands doesn't even have a peaceful mode. Almost every match is about war, especially tournament matches. Mostly the economy aims for war.
No, you are wrong. There are peaceful modes. Collector, tree gnome, ...
If you don't have very experienced players, they don't fight then. They fight for points, not in a military way.
teppo wrote:
I do not quite understand. What is the benefit of reworking the tree algorithm?
It will solve a lot of problems, for example:
-
It's hard to know how well trees grow, especially for beginners
-
Trees grow poorly at coasts
-
Mixed terrains in general are very complicated for trees
In the early unified worlds in Widelands, there was no difference on which terrain which tree is growing. That was awful. So NO. Reworking whole model is not the perfect solution. Better is to show on which terrain trees are growing well.
- Foresters don't always walk to the best places
I think that it isn't necessary. They walk randomly around their houses. As it supposed to be.
- Foresters don't always use the best tree.
Yes, you're right. We have 6 best choices. That is enough to have multiple-species forests.
einstein13 wrote:
About geologist new job:
There are always several solutions for the same problem. For geologist we can provide combined messages / two messages per one node.
Second solution is to make REAL 2 JOBS for the same geologist.
Geologists already have two jobs: Checking out mining areas and checking out areas which can have well-water
For you only: to make real 3 jobs for the same geologist.
I don't know how you think here, that "REAL" is different between finding resources and finding resources... 
And to answer @WorldSavior discussion:
I don't think that Widelands should be simplified. Even more: I like to make it even more complicated in the background. But also I have to think about impacts (what can go wrong) and how to solve it and make more easier to start with.
I think it's just wrong to change the old tribes that much. Why not putting your ideas into new tribes instead?
So you think that best idea is to push "fertility-like" for crops and add some benefits into new tribes? I don't think that it will work. New tribe would play well on any map, while other tribes would play well only on very fertile maps (like elven forest). Barren maps (like The Nile) would be unplayable for them.
einstein13
calculations & maps packages: http://wuatek.no-ip.org/~rak/widelands/
backup website files: http://kartezjusz.ddns.net/upload/widelands/
Top
Quote
|
|
|
GunChleoc
Joined: 2013-10-07, 14:56
Posts: 3324
One Elder of Players
Location: RenderedRect
|
Posted at: 2017-12-15, 19:08
WorldSavior wrote:
- Foresters don't always walk to the best places
We actually have an open bug for that: https://bugs.launchpad.net/widelands/+bug/1574379
Busy indexing nil values
Top
Quote
|
|
|
WorldSavior
Joined: 2016-10-15, 03:10
Posts: 2094
OS: Linux
Version: Recent tournament version
One Elder of Players
Location: Germany
|
Posted at: 2017-12-16, 20:39
einstein13 wrote:
WorldSavior wrote:
kaputtnik wrote:
The main focus of widelands is building and managing a good economy. Ware fare is just one part of it, not the main goal.
Widelands doesn't even have a peaceful mode. Almost every match is about war, especially tournament matches. Mostly the economy aims for war.
No, you are wrong. There are peaceful modes. Collector, tree gnome, ...
If you don't have very experienced players, they don't fight then. They fight for points, not in a military way.
Come on, they are no peaceful modes.
teppo wrote:
I do not quite understand. What is the benefit of reworking the tree algorithm?
It will solve a lot of problems, for example:
-
It's hard to know how well trees grow, especially for beginners
-
Trees grow poorly at coasts
-
Mixed terrains in general are very complicated for trees
In the early unified worlds in Widelands, there was no difference on which terrain which tree is growing. That was awful. So NO. Reworking whole model is not the perfect solution. Better is to show on which terrain trees are growing well.
Well, other solutions could be possible. For example, as foresters plant right now only trees which fit well to the terrain, one could just increase the growth possibilities a lot and foresters wouldn't plant "wrong" trees. And one could make it easier in the editor to avoid "wrong" trees.
- Foresters don't always walk to the best places
I think that it isn't necessary. They walk randomly around their houses. As it supposed to be.
Why is it supposed to be like that? The foresters would be more efficient if they would only walk to the places with the best tree growth, and if they would walk first to the closest places. That would be intelligent.
- Foresters don't always use the best tree.
Yes, you're right. We have 6 best choices. That is enough to have multiple-species forests.
Who needs multiple-species forests if choosing always the best tree is much more efficient?
einstein13 wrote:
About geologist new job:
There are always several solutions for the same problem. For geologist we can provide combined messages / two messages per one node.
Second solution is to make REAL 2 JOBS for the same geologist.
Geologists already have two jobs: Checking out mining areas and checking out areas which can have well-water
For you only: to make real 3 jobs for the same geologist.
I don't know how you think here, that "REAL" is different between finding resources and finding resources...
There is a difference between water and ores/coal...
And to answer @WorldSavior discussion:
I don't think that Widelands should be simplified. Even more: I like to make it even more complicated in the background. But also I have to think about impacts (what can go wrong) and how to solve it and make more easier to start with.
I think it's just wrong to change the old tribes that much. Why not putting your ideas into new tribes instead?
So you think that best idea is to push "fertility-like" for crops and add some benefits into new tribes?
No, I don't think so. I think that plants should grow well.
I don't think that it will work. New tribe would play well on any map, while other tribes would play well only on very fertile maps (like elven forest). Barren maps (like The Nile) would be unplayable for them.
.
GunChleoc wrote:
WorldSavior wrote:
- Foresters don't always walk to the best places
We actually have an open bug for that: https://bugs.launchpad.net/widelands/+bug/1574379
Reasonable...
Edited: 2017-12-16, 20:41
Wanted to save the world, then I got widetracked
Top
Quote
|
|
|
teppo
Joined: 2012-01-30, 08:42
Posts: 423
Tribe Member
|
Posted at: 2017-12-17, 20:01
WorldSavior wrote:
Yes, you're right. We have 6 best choices. That is enough to have multiple-species forests.
Who needs multiple-species forests if choosing always the best tree is much more efficient?
The odds are weighted; the best-suited three gets planted significantly more often than the sixth-best one. The net difference in overall tree growth is not that huge.
It is quite easy to increase efficiency, in this sense. We could make all trees grow in zero time in any landscape. Would that increase the fun of playing?
I think that the foresters could favor spots where trees grow well, over less good ones within their area of work.
Top
Quote
|
|
|
GunChleoc
Joined: 2013-10-07, 14:56
Posts: 3324
One Elder of Players
Location: RenderedRect
|
Posted at: 2017-12-18, 09:22
teppo wrote:
I think that the foresters could favor spots where trees grow well, over less good ones within their area of work.
Exactly - we already discussed how to solve this a while ago, and that's the exact solution that we came up with. See the bug that I linked above. We just need somebody to implement it.
Busy indexing nil values
Top
Quote
|
