Latest Posts

Topic: taking a millitary building

Nordfriese
Avatar
Joined: 2017-01-17, 18:07
Posts: 1929
OS: Debian Testing
Version: Latest master
Ranking
One Elder of Players
Location: 0x55555d3a34c0
Posted at: 2021-02-08, 17:52

BoeseKaiser wrote:

I would say that for me the attacking player should only be able to take a building when their force are at least a bit stronger than the opponent. (modulo potential chance factors) Of course, that requires a good definition of "stronger" and that's quite hard to find. I would work with something like "if an army would win in most simulations of randomised fight-order, it is a stronger one" with probably a correcting factor to reward big number of soldiers (for the definition of strenght, I'm not talking about the implementation of the actual fights ! )

It will never truly work like that because of strategies. Let's say your enemy has 20 supersoldiers, and you have 2 supersoldiers and 20 half-trained soldiers. The enemy is way stronger, isn't he? Now he sends all his soldiers to take one of your military buildings. The half-trained guys swarm out to intercept them and keep them busy for a brief while. Meanwhile, you send your pair of supersoldiers to attack one of the enemy's buildings, which is empty now except for one defender and easily conquered. To round it all off, you can dismantle the milsite under attack when the attackers have pushed past the defenders and prevent them taking anything. All in all, the outcome of the battle is a success for the weaker player (or at least, very bad for the stronger one). The only way to prevent this would be to empty the game of all military strategy which is definitely not what we want.

Edited: 2021-02-08, 17:55

Top Quote
BoeseKaiser
Avatar
Topic Opener
Joined: 2019-02-21, 11:03
Posts: 41
Ranking
Pry about Widelands
Posted at: 2021-02-08, 18:23

Nordfriese wrote:

BoeseKaiser wrote:

I would say that for me the attacking player should only be able to take a building when their force are at least a bit stronger than the opponent. (modulo potential chance factors) Of course, that requires a good definition of "stronger" and that's quite hard to find. I would work with something like "if an army would win in most simulations of randomised fight-order, it is a stronger one" with probably a correcting factor to reward big number of soldiers (for the definition of strenght, I'm not talking about the implementation of the actual fights ! )

It will never truly work like that because of strategies. Let's say your enemy has 20 supersoldiers, and you have 2 supersoldiers and 20 half-trained soldiers. The enemy is way stronger, isn't he? Now he sends all his soldiers to take one of your military buildings. The half-trained guys swarm out to intercept them and keep them busy for a brief while. Meanwhile, you send your pair of supersoldiers to attack one of the enemy's buildings, which is empty now except for one defender and easily conquered. To round it all off, you can dismantle the milsite under attack when the attackers have pushed past the defenders and prevent them taking anything. All in all, the outcome of the battle is a success for the weaker player (or at least, very bad for the stronger one). The only way to prevent this would be to empty the game of all military strategy which is definitely not what we want.

I don't see how this remove all strategy to the game. no one forces you to send all your soldiers to attack. I rarely do. Plus, in your scenario, the weaker player loses a lot of half trained, which is much more ressources tahn a dismanteled or even destroyed millitary building. So this scenario is a win for the stronger player. Plus, the counter attack is already a thing, right? I don't see how trying to ensure that a stronger force can defend a site removes strategy in the game.

The only effect of this kind that i can think of is that swarming your opponent with a lot of weak soldiers would be a less efficient strategy. And that is only if we ensure that no attacking soldier can take a building if there is any defending unit left outside. I would find that overkill, that's not what i'm advocating. swarming would still work if we just ensure that all the defenders outside that are still able to fight gets one of the attacking soldiers. so the attacking player actually has to occupy all the defending units before reaching the millitary building.


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2021-02-08, 21:14

BoeseKaiser wrote:

I would say that for me the attacking player should only be able to take a building when their force are at least a bit stronger than the opponent. (modulo potential chance factors) Of course, that requires a good definition of "stronger" and that's quite hard to find. I would work with something like "if an army would win in most simulations of randomised fight-order, it is a stronger one" with probably a correcting factor to reward big number of soldiers (for the definition of strenght, I'm not talking about the implementation of the actual fights ! )

-1 from me, cause this concept bears a lot of problems:
1. For me the biggest problem here is the definition of "army". I.e. which soldiers would count for the calculation of an "army".
2. It is intended that fights are luck based to some extent. Therefore our fighting calculations have multiple instances of the random function. So calculating a means of randomness would contradict this.
3. there are many permutations of halftrained soldiers so calculating if an "army" would eventually win, would consume a lot of power. (calculating the probability table for balancing military strength takes already a few minutes for the 5 x 5 matrix if we go for a good statistically base of 10 000 Which still has 2% uncertainty in it .
4. This would render a lot of tactics like mock attacks and other rather useless.
5. this might end in endless games where the balance is kept all the time.

Furthermore from my perspective it is the players fault if he does not dismantle the building in the right moment. It is his fault if he has not staffed it well. It is his fault if he diddn't counterattack early enough to make the soldiers meet in the middle and letting them come near your homeland.
Finally changing the complete game mechanics in this particular sector due to some experience that can be avoided by experienced players does not sound right for me.


Top Quote
BoeseKaiser
Avatar
Topic Opener
Joined: 2019-02-21, 11:03
Posts: 41
Ranking
Pry about Widelands
Posted at: 2021-02-08, 22:10

hessenfarmer wrote:

  1. For me the biggest problem here is the definition of "army". I.e. which soldiers would count for the calculation of an "army".

the one that were sent to attack and to defend (so only the one involved in a fight)

  1. It is intended that fights are luck based to some extent. Therefore our fighting calculations have multiple instances of the random function. So calculating a means of randomness would contradict this.
  2. there are many permutations of halftrained soldiers so calculating if an "army" would eventually win, would consume a lot of power. (calculating the probability table for balancing military strength takes already a few minutes for the 5 x 5 matrix if we go for a good statistically base of 10 000 Which still has 2% uncertainty in it .

As I said, this definition was not intended as an idea of algorithm, but rather a evaluation of what could be seen as a reasonable outcome for a given fight. As a simplified example : "should an attacking force of 2 heroes and 8 soldiers with 3 upgrades be able to take (force dismantle) a building defended by 3 heroes and 10 level 3 soldiers" Of course there would still be a part of randomness, I don't want to take that out of the equation.

My point is about soldiers going out of the buildings to defend but not geting a chance to engage. This can be solved, as you said, by counter attacking early enough. But if this is supposed to be the solution, why send soldiers out to defend that are never gonna have a chance to reach the opponent? this still seems illogical to me.

  1. This would render a lot of tactics like mock attacks and other rather useless.
  2. this might end in endless games where the balance is kept all the time.

I think this refers to you thinking that my definition was an algorithm proposition. I agree with you, this as algorithm to decide the outcome of a fight would be really bad both in complexity and because it would make no sense whatsoever to actually do the fight like that.

Furthermore from my perspective it is the players fault if he does not dismantle the building in the right moment. It is his fault if he has not staffed it well. It is his fault if he diddn't counterattack early enough to make the soldiers meet in the middle and letting them come near your homeland.

I did say that i can see a couple of ways to reduce this effect, but wondered if this was a good mechanism. In the cases i'm describing though, i'm talking about a defenser that has a well staffed building. And the dismanteling is also irrelevant : Even if I dismantle the building in time, I still have to give up my position even though i had more than enough to defend.

Finally changing the complete game mechanics in this particular sector due to some experience that can be avoided by experienced players does not sound right for me.

I was also not asking for a change in game but rather asked if this was intended. I insisted because I think what I was saying was not well understood. I still think that there are cases where experienced players could end up losing buildings (by that I mean having to dismantle them) to rather smaller forces due to this effect. Making sure that every soldier that go out to defend get to fight an opponent if there is one avilable doesn't sound like a complete change in game mechanism to me face-smile.png

This is also not a rant because I lost a game or so. I get that it can sound like that, but this is really something that I find annoying independently of which side of it i find myself.

(Edit : formatting)

Edited: 2021-02-08, 22:14

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2021-02-09, 07:11

BoeseKaiser wrote:

hessenfarmer wrote:

  1. For me the biggest problem here is the definition of "army". I.e. which soldiers would count for the calculation of an "army".

the one that were sent to attack and to defend (so only the one involved in a fight)

Even with this definition it is hard to evaluate. Cause there might be (and often are) simultaneous attacks on different buildings just for the purpose of tricking defenders out.

  1. It is intended that fights are luck based to some extent. Therefore our fighting calculations have multiple instances of the random function. So calculating a means of randomness would contradict this.
  2. there are many permutations of halftrained soldiers so calculating if an "army" would eventually win, would consume a lot of power. (calculating the probability table for balancing military strength takes already a few minutes for the 5 x 5 matrix if we go for a good statistically base of 10 000 Which still has 2% uncertainty in it .

As I said, this definition was not intended as an idea of algorithm, but rather a evaluation of what could be seen as a reasonable outcome for a given fight. As a simplified example : "should an attacking force of 2 heroes and 8 soldiers with 3 upgrades be able to take (force dismantle) a building defended by 3 heroes and 10 level 3 soldiers" Of course there would still be a part of randomness, I don't want to take that out of the equation.

My point is about soldiers going out of the buildings to defend but not geting a chance to engage. This can be solved, as you said, by counter attacking early enough. But if this is supposed to be the solution, why send soldiers out to defend that are never gonna have a chance to reach the opponent? this still seems illogical to me.

If this is your real point, I could agree that the algorithm of sending soldiers out can be improved. and they might get in earlier rather then standing in front of their houses watching the enemy retreating.

  1. This would render a lot of tactics like mock attacks and other rather useless.
  2. this might end in endless games where the balance is kept all the time.

I think this refers to you thinking that my definition was an algorithm proposition. I agree with you, this as algorithm to decide the outcome of a fight would be really bad both in complexity and because it would make no sense whatsoever to actually do the fight like that.

Ok sorry for the misunderstanding.

Furthermore from my perspective it is the players fault if he does not dismantle the building in the right moment. It is his fault if he has not staffed it well. It is his fault if he diddn't counterattack early enough to make the soldiers meet in the middle and letting them come near your homeland.

I did say that i can see a couple of ways to reduce this effect, but wondered if this was a good mechanism. In the cases i'm describing though, i'm talking about a defenser that has a well staffed building. And the dismanteling is also irrelevant : Even if I dismantle the building in time, I still have to give up my position even though i had more than enough to defend.

You would only loose your position if you only had that one building and the opponent already built an own MilSite near the border. So seeing an opponent build such milsite should make you build some as well.

Finally changing the complete game mechanics in this particular sector due to some experience that can be avoided by experienced players does not sound right for me.

I was also not asking for a change in game but rather asked if this was intended. I insisted because I think what I was saying was not well understood. I still think that there are cases where experienced players could end up losing buildings (by that I mean having to dismantle them) to rather smaller forces due to this effect.

Well, I would say, if this is due to good attackers tactics this should be inbtentional. Good tactics include the numbers and timing of attacking waves to make use of the current mechanism. Expert gameplay would try to use every weakness in our algorithms to their advantage.

Making sure that every soldier that go out to defend get to fight an opponent if there is one avilable doesn't sound like a complete change in game mechanism to me face-smile.png

Ok As stated above this algorithm can be (and probably should be) improved, while this would not avoid the mentioned situation that you still might get tricked out to loose one special building.


Top Quote
BoeseKaiser
Avatar
Topic Opener
Joined: 2019-02-21, 11:03
Posts: 41
Ranking
Pry about Widelands
Posted at: 2021-02-09, 10:23

Furthermore from my perspective it is the players fault if he does not dismantle the building in the right moment. It is his fault if he has not staffed it well. It is his fault if he diddn't counterattack early enough to make the soldiers meet in the middle and letting them come near your homeland.

I did say that i can see a couple of ways to reduce this effect, but wondered if this was a good mechanism. In the cases i'm describing though, i'm talking about a defenser that has a well staffed building. And the dismanteling is also irrelevant : Even if I dismantle the building in time, I still have to give up my position even though i had more than enough to defend.

You would only loose your position if you only had that one building and the opponent already built an own MilSite near the border. So seeing an opponent build such milsite should make you build some as well.

Still, you lose a building, this weakens your position. (the case i'm talking about can only happen if there are multiple defending buildings anyways) Plus, if you are in a front scenario, you tell me to build more, but that donesn't bring me much if I already have a lot of buildings but that I cann't use them reliably for defense. Of course I can build one, let it empty, and occupy it when i lost the first one. this works pretty well, but should i really have to do that when I could have defended through sheer soldier strenght and numbers?

Finally changing the complete game mechanics in this particular sector due to some experience that can be avoided by experienced players does not sound right for me.

I was also not asking for a change in game but rather asked if this was intended. I insisted because I think what I was saying was not well understood. I still think that there are cases where experienced players could end up losing buildings (by that I mean having to dismantle them) to rather smaller forces due to this effect.

Well, I would say, if this is due to good attackers tactics this should be inbtentional. Good tactics include the numbers and timing of attacking waves to make use of the current mechanism. Expert gameplay would try to use every weakness in our algorithms to their advantage.

this happened a couple of times with no fighting tacticks, just me or my opponent sending a force to attack and forcing dismantle on a building just like that even when the defender has more.

Making sure that every soldier that go out to defend get to fight an opponent if there is one avilable doesn't sound like a complete change in game mechanism to me face-smile.png

Ok As stated above this algorithm can be (and probably should be) improved, while this would not avoid the mentioned situation that you still might get tricked out to loose one special building.

I have no problems getting tricked if I can relyably understand how and why face-wink.png

Edited: 2021-02-09, 10:31

Top Quote
DragonAtma
Avatar
Joined: 2014-09-14, 01:54
Posts: 351
Ranking
Tribe Member
Posted at: 2021-02-09, 13:18

Keep in mind that a captured building creates a two-spot no-mans-land around it (enemy border, then your border) and burns down any buildings in that two-spot area (unless one of them is an occupied military building, in which case the 'captured' one burns down instead).

So don't put a warehouse right next to a military building. ;)


Top Quote