Currently Online

Latest Posts

Topic: Inconsistent balancing with mines

No0815
Avatar
Topic Opener
Joined: 2016-05-01, 13:52
Posts: 49
Ranking
Pry about Widelands
Posted at: 2016-12-18, 19:05

Half a year ago I posted a bug report on launchpad after noticing that barbarian mines appear to have rather chaotic production values (see Bug #1583714 inconsistent production values of mines). Since it didn't get any reply I'd like to take that into my hands. But before I can do anything there are some questions that need to be answered and we have to decide, potentially, what the new balance should look like.

First I'd like to know: Is there a logic in it I can't see? At least Empire and Atlanteans appear much more systematic. When I double-checked I also discovered, that Barbarians have a significant disadvantage compared to the other tribes. Not just do they have to upgrade their mines two times, they also are the only tribe that requires a finite resource to built a mine. Atlanteans on the other hand have a significant advantage since they don't have to upgrade their mines and have a much bigger work-radius. If I'm not mistaken, they can cover more than three times the size with a single mine compared to Barbarians and Empire.

I'll do some calculations first, to get a better idea of the current balancing.


Top Quote
king_of_nowhere
Avatar
Joined: 2014-09-15, 17:35
Posts: 1668
Ranking
One Elder of Players
Posted at: 2016-12-18, 20:04

see, the general idea is that barbarians are stronger in the early game, while atlanteans are stronger later, and empire is balanced. barbarian mines are incredibly cheap to operate. you can get a ration from just one meat, which barbarians have in limitless supply thanks to gamekeepers. so a barbarian player can be making new troops at a steady rate already after 20 minutes. the downside is that once they have to upgrade the mines, they become very expensive to operate. atlanteans have a hard time getting their first mine going, as all that wood is a lot in early game. atlanteans take much longer to get their mines, or in general an economy, though they can start supplying the evade promotion after roughly half an hour. To compensate for that, barbarian basic troops are very weak, but their advanced troops are strong, just more expensive.

As for stones, technically they are a finite resource, but in practice there are always much more than needed. empire needs a lot of stones, so if a map is made to be playavble for empire it has plenty of stones for the other tribes. With barbarians and atlanteans, i consider stones a worthless resource. in fact, barbarian mines are cheaper for emplying stones


Top Quote
SirVer

Joined: 2009-02-19, 14:18
Posts: 1445
Ranking
One Elder of Players
Location: Germany - Munich
Posted at: 2016-12-18, 20:58

King is absolutely correct in his assessment of the overall design idea. That is not to say that everything is perfect. I read your bug report again - and I think it fell through the cracks because it is quite hard to understand what you are complaining about exactly - I cannot parse it at least.

For balance changes it is always the best to start listing the status quo for all tribes and then put an argument forth why it should be changed.

If I'm not mistaken, they can cover more than three times the size with a single mine compared to Barbarians and Empire.

No, only two times. This are the commands for the deepest mines for all tribes:

bararians deeper coalmine: mine=coal 2 100 10 2 (range is the same for smaller mines)
empire deep coalmine: mine=coal 2 100 5 2 (range is the same for "coalmine")
atlanteans coalmine: mine=coal 4 100 5 2

The mine command should have been documented here under Productionsite Program Reference but was not. I fixed this in this branch which should be merged soon.

the first number after "coal" is the range. The others are: at which percentage of starting resources mined is this empty (100 for the deepest mines). The next one is chance to mine if the mine is exhausted in percent and the last is chance for workers to gain experience even if work failed.

You can see that atleanteans have twice the reach of the other tribes. I think this is something that warrants discussion: for example it makes sense to me that the deeper mines for the other tribes also have an extended reach that gets to 4 at the top level. I doubt that this strongly affects balance, it is more of a convenience change (so that you get all the resources in a mountain without needing to build new mines).

Edited: 2016-12-18, 21:00

Top Quote
GunChleoc
Avatar
Joined: 2013-10-07, 14:56
Posts: 3324
Ranking
One Elder of Players
Location: RenderedRect
Posted at: 2016-12-19, 10:35

I think extending the reach is a good idea - if Barbarian deep mines have a range of 3, it will also take longer until they have to upgrade again. That will make it easier for the workers to gain the necessary experience.


Busy indexing nil values

Top Quote
No0815
Avatar
Topic Opener
Joined: 2016-05-01, 13:52
Posts: 49
Ranking
Pry about Widelands
Posted at: 2016-12-19, 14:06

king_of_nowhere wrote:

To compensate for that, barbarian basic troops are very weak, but their advanced troops are strong, just more expensive.

Hm, wasn't it the other way around when I last read about it?

As for stones, technically they are a finite resource, but in practice there are always much more than needed. empire needs a lot of stones, so if a map is made to be playavble for empire it has plenty of stones for the other tribes. With barbarians and atlanteans, i consider stones a worthless resource. in fact, barbarian mines are cheaper for emplying stones

Point taken. I shouldn't have tackled the balancing between the tribes since I'm too inexperienced to judge it substantially. Especially since it wasn't my main point.

SirVer wrote:

King is absolutely correct in his assessment of the overall design idea. That is not to say that everything is perfect. I read your bug report again - and I think it fell through the cracks because it is quite hard to understand what you are complaining about exactly - I cannot parse it at least.

Well, I was hoping that one would understand what I was aiming for. While it's clear and easy in my head I have surprising trouble to put it into words.

If I'm not mistaken, they can cover more than three times the size with a single mine compared to Barbarians and Empire.

No, only two times.

Not quite. The radius is doubled. Linear change of radius means quadratic change of size. Barbarian and Empires' mines cover 19 nodes while Atlanteans cover 61. I miscounted the number the last time. Wondering what the formula might be I came up with this:

3 * r * (r + 1) + 1

the first number after "coal" is the range. The others are: at which percentage of starting resources mined is this empty (100 for the deepest mines). The next one is chance to mine if the mine is exhausted in percent and the last is chance for workers to gain experience even if work failed.

Yes, adding this would be useful. I figured out the first three values (aside the obvious "coal") after some time but always wondered what the last one might mean.

You can see that atleanteans have twice the reach of the other tribes. I think this is something that warrants discussion: for example it makes sense to me that the deeper mines for the other tribes also have an extended reach that gets to 4 at the top level. I doubt that this strongly affects balance, it is more of a convenience change (so that you get all the resources in a mountain without needing to build new mines).

Personally I would appreciate a bigger range. I often find it annoying to fit in the mines, though that's also due to perfectionism and wanting to mine as much resources as possible with as few mines as possible.

GunChleoc wrote:

if Barbarian deep mines have a range of 3, it will also take longer until they have to upgrade again. That will make it easier for the workers to gain the necessary experience.

I'm not so sure about that. Considering the issue with how master miners are distributed by the game currently it might backfire when too much experience is gained and you end up with even more useless master miners. On the other hand, it might make upgrading to deep mines more attractive again. Currently, a snack needs exactly the same basic resources as a meal but you get 5 coal from a meal vs. 4 from a snack. Plus saving one from having to produce the otherwise useless beer. That was actually the second point I wanted to make. So, let me explain my main point first before it gets even more confusing. I'll make it a new post (might need some time). Or, better yet, add it to the original bug report.


Top Quote
king_of_nowhere
Avatar
Joined: 2014-09-15, 17:35
Posts: 1668
Ranking
One Elder of Players
Posted at: 2016-12-19, 16:56

No0815 wrote:

king_of_nowhere wrote:

To compensate for that, barbarian basic troops are very weak, but their advanced troops are strong, just more expensive.

Hm, wasn't it the other way around when I last read about it?

yes, it used to be the other way around. about one year ago, i proposed a rebalance, arguing that while it's ok that the stronger tribe early game should become the weakest later, they should not become completely useless, as they used to be for having more expensive AND weaker soldiers. it was agreed, and now barbarian attack scales a bit more. this does not affect soldiers without promotions in attack, it is virtually unnoticeable if they have attck 1 or 2, but when they have attack 5, they sometimes can kill a fully promoted enemy in 4 strikes instead of 5. that was enough to bring their likelyhood to win against an atlantean from 35% to about 50%. It was a change designed specifically to only affect fully promoted soldiers.

In the same change, a cost of 2 gold was added to the atlantean labirynth, because there was a strategy that allowed an atlantean player to get a fully promoted soldier (well, lacking defence 2, but it's practically the same) after one hour of play, without building any mine, and that was game-breaking. Now atlanteans have to make mines to get fully promoted soldiers.

EDIT: Also, after thinking about it, I suppport increasing the range of deep and deeper mines. it makes things more comfortable, especially with features like forested mountain where you may want to plant mines in the middle of a forest, but it doesn't affect the balance: it is more convenient to make simple mines as much as possible because they work cheaper, so one would just have to ignore higher cost in favor of greater convenience.

Edited: 2016-12-19, 18:46

Top Quote
GunChleoc
Avatar
Joined: 2013-10-07, 14:56
Posts: 3324
Ranking
One Elder of Players
Location: RenderedRect
Posted at: 2016-12-20, 09:28

I'm not so sure about that. Considering the issue with how master miners are distributed by the game currently it might backfire when too much experience is gained and you end up with even more useless master miners.

Why do you think master miners will be useless? They can work any mine, including the basic ones. So, no harm done.

I have had a look at your bug now, an it does seem a bit weird that the relations aren't kept constant from level to level. We should also have a look at the other tribes to compare it to what they have, so we can think about which values would be good to have here.


Busy indexing nil values

Top Quote
king_of_nowhere
Avatar
Joined: 2014-09-15, 17:35
Posts: 1668
Ranking
One Elder of Players
Posted at: 2016-12-20, 16:35

GunChleoc wrote:

I'm not so sure about that. Considering the issue with how master miners are distributed by the game currently it might backfire when too much experience is gained and you end up with even more useless master miners.

Why do you think master miners will be useless? They can work any mine, including the basic ones. So, no harm done.

actually, at least in some trunk version, the game would make a new miner instead of sending a master miner. the rationale was to avoid the situation where all your master miners are in normal mines and your deep mines cannot find a worker. but the downside is that now you accumulate dozens of master miners, which would become basically useless. I'm not sure if that particular piece of code is still active or iif it had been changed.


Top Quote
No0815
Avatar
Topic Opener
Joined: 2016-05-01, 13:52
Posts: 49
Ranking
Pry about Widelands
Posted at: 2016-12-20, 18:44

Ugh, apparently I'm waaay to slow with writing.

Or, better yet, add it to the original bug report.

Done now.

GunChleoc wrote:

I'm not so sure about that. Considering the issue with how master miners are distributed by the game currently it might backfire when too much experience is gained and you end up with even more useless master miners.

Why do you think master miners will be useless? They can work any mine, including the basic ones. So, no harm done.

Well, according to your reply in the thread from may the behavior wasn't changed yet. But reading the related bug report really confused me. I don't know anymore if something has changed in the meantime, if the bug was valid at all or what the current behavior is ... face-confused.png


Top Quote
No0815
Avatar
Topic Opener
Joined: 2016-05-01, 13:52
Posts: 49
Ranking
Pry about Widelands
Posted at: 2017-01-04, 16:26

I've posted some values in the bug report as a first draft. While I agree with SirVer that these changes won't have much impact on balancing there are some more things that are bugging me which would - potentially - affect balancing.

But first: It seems there is an agreement about extending the work radius for some mines. As far as I understand we would increase the radius for barbarians deep and deeper mines to 3 and keep basic mines at 2? It would be a interesting twist since basic mines would get a small disadvantage this way as compensation for their strength because of the low food cost and upgrading will be a bit more rewarding. What about empire's mines? Should their ranges be increased, too? Otherwise they would be the tribe with the lowest mining range.


Top Quote