Currently Online

Latest Posts

Topic: Efficiency of Scouting

the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 674
Ranking
One Elder of Players
Posted at: 2020-10-28, 11:45

Scouting can also maybe need some macrobalancing: For example the scout that costs 1 ration and covers up a relative big part of the map is something that is nice looking but not cost relevant or balanced.

In my version i put the cost of covering up the map, what the scout does to 20 ration, which solves a lot of probelms at once. For example the defensive / offensive part: At the beginning 20 ration is relatively high, so you better invest it in upgrades, making the defence stronger - with this 20 food you can feed your coalmine, get maybe 3-4 upgrades which gives: Defense a better strategy in the beginning. By this we can lower healing and have a perfect at the start defensive optimum whilst later tending from time to time a more offensive gameplay. Later 20 food is compared nothing and so players will invest in scouting. Combined with lowering the view of atlanteans towers, which are a lot higher than the towers of the others it gives wonderful gameplays tending to a defensively start at the beginning whilst the game slowly getting more interesting (and not at once, atm i can easily food rush or labyrinth rush at every map and win)

If you look at many principles for a economic or mathematic side of view there are big differences in efficiency. I know that it is not wished to talk about balance cause no one takes "responsibility" for macro, but you dont need to be a mathematics professor to get behind the good looking.

Edit (GC): Deleted ad hominem attack

Edited: 2020-10-28, 18:24

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 1607
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-10-28, 11:56

the-x wrote:

If you look at many principles for a economic or mathematic side of view there are big differences in efficiency. I know that it is not wished to talk about balance cause no one takes "responsibility" for macro, but you dont need to be a mathematics professor to get behind the good looking.

So far you never provided any mathematical evidence for any of your suggestions. So it seems that it is more difficult then you pretend.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 674
Ranking
One Elder of Players
Posted at: 2020-10-28, 12:03

I have send you all the values two weeks ago in my balanced Version ; )


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 1607
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-10-28, 12:11

the-x wrote:

I have send you all the values two weeks ago in my balanced Version ; )

You sent changes you applied, but no mathematical reason for doing so. This is a difference as you might realise.

We never received an answer on this post for example https://www.widelands.org/forum/post/35100/


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 674
Ranking
One Elder of Players
Posted at: 2020-10-28, 12:33

Recently, I played 38 in Singleplayer (not reliable) and 30 Games in Multiplayer testing:

-

old version without macrobalances:

Atlantean vs Barbarian: 10/10 Atlantean wins 0/10 Barbarian win

Atlantean vs Imperium: 10/10 Atlantean wins 0/10 Imperium win

Atlantean vs Frisian: 10/10 Atlantean wins 0/10 Frisian win

new version:

Atlantean vs Barbarian: 5/10 Atlantean wins 5/10 Barbarian win

Atlantean vs Imperium: 7/10 Atlantean wins 3/10 Imperium win

Atlantean vs Frisian: 5/10 Atlantean wins 5/10 Frisian win

-

Game Statistics in the old Version:

Barbarian always ahead in Units for a long time, Atlantean always ahead from the first combat

Game Statistics in the macrobalance Version:

Barbarians and Atlanteans now tend to start almost at the same time, if you play Labyrinth at the beginning with Atl, Atl are still a bit faster but not so much as the first 1 iron 1 coal instead of 2 iron 1 coal upgrade gets heavier weight.

-> Result:

  • All tribes tend now to a more adapted graph

  • Timing of the tribes is real interesting: You get a small increase in the beginning, whilst then depending on which strategy is played an higher increase and later an exponential increase which will decide any game, even defense bug of fortresses. Timing before was a late start at around 35-40 min then strictly linear without any changes, Barbarian always ahead of Atlantean, Atlantean always winning against Barbarian.

If you need more let me know, i added them also the topic


Top Quote
Mars
Joined: 2009-03-28, 10:26
Posts: 81
Ranking
Likes to be here
Location: Germany
Posted at: 2020-10-28, 12:34

Maybe we can and should approach the balancing discussing in a different way.

@the-x: As long as you can't provide lot's of simulations showing that your changes improve balance they will not get merged. Your changes are loo large and change too many thinks at the same time. This is not how software development works.

But I think we agree on:

  • Atlanteans are too strong
  • scouts are unbalanced

Usually the solution for large projects is to propose SMALL changes. If they get accepted, you can propose another small change.

Maybe we also need a balancing forum where we can create a separate topic for every balance related change request. This could be:

  • rework scout mechanics
  • decrease healing
  • lower Altanteans tower sight

Every change has to be independent of other changes. Otherwise it will be difficult to argue about whether they are needed or whether they improve balancing or not.

Regarding scouts:

I don't really like the current mechanics.

There is only a really short amount of time when it is possible to attack an enemy military building disclosed by a scout. Especially when you play at a higher speed or have a large kingdom to manage, it is difficult to keep track of where the scout is and whether he is about to uncover the building you are aiming for. In my opinion it would be more important to increase the uncover time of the scout (we could increase the vision range or only fade out the uncovered area after some time).

I think increasing the cost is ok, but going from 1 to 20 is too extreme, especially for new players. Maybe we can go to 2 or 3 first and see how it works.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 674
Ranking
One Elder of Players
Posted at: 2020-10-28, 12:49

Mars wrote:

I think increasing the cost is ok, but going from 1 to 20 is too extreme, especially for new players. Maybe we can go to 2 or 3 first and see how it works.

Of course, but even without a lot of calculations I see that covering this big amount of a map is rather 20 rations than 2 or 3 because:

  • 2 ration is anyway a highly circling cost even in late games if you have 4-5 scouts unattended

  • 2 ration in the beginning I will always invest to see my enemys headquaters or the tower i want to conquer. For 2 food, always.

  • 20 ration is not extreme but the value where everybody will think about Will I or wont I? In the right beginning i probably wont cause 20 is a lot and so it will go in upgrades or defensive. Try it on a small or medium map, its amazing ; ) after 1 h 15 my food production rolls, my defensive is set up im thinking hey, now is a good moment and i might be stronger in this area so 20 ration is peanuts for covering your enemy. We shouldnt look at the change or feel like giving it more or less, we should set costs exactly to efficiency.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 674
Ranking
One Elder of Players
Posted at: 2020-10-28, 13:03

Mars wrote:

Maybe we can and should approach the balancing discussing in a different way.

Of course

But I think we agree on:

  • Atlanteans are too strong
  • scouts are unbalanced

Agreed

Usually the solution for large projects is to propose SMALL changes. If they get accepted, you can propose another small change.

Maybe we also need a balancing forum where we can create a separate topic for every balance related change request. This could be:

  • rework scout mechanics

Very easy. Cost to efficieny would do it face-wink.png

  • decrease healing

A possibility might be at least that a castle is conquerable. I propose values around 15 - 27 hitpoints per second, cause they seem valuable in most playtests

  • lower Altanteans tower sight

+1 Barbarians have lower soldiers inside, lower sight whilst having higher costs

I don't really like the current mechanics.

There is only a really short amount of time when it is possible to attack an enemy military building disclosed by a scout. Especially when you play at a higher speed or have a large kingdom to manage, it is difficult to keep track of where the scout is and whether he is about to uncover the building you are aiming for. In my opinion it would be more important to increase the uncover time of the scout (we could increase the vision range or only fade out the uncovered area after some time).

So do I, the mechanics challenges you only in fast clicking whilst having an eye at economy and scout multitasking

I think increasing the cost is ok

+1

Edit (GC): Deleted ad hominem attack

Edited: 2020-10-28, 18:24

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 1607
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-10-28, 13:13

Mars wrote:

Maybe we can and should approach the balancing discussing in a different way.

But I think we agree on:

  • Atlanteans are too strong

Yes, but only a bit. We made a lot of small changes to reduce imbalances already.

  • scouts are unbalanced

I don't htink so as they are pretty much the same for all tribes.

Usually the solution for large projects is to propose SMALL changes. If they get accepted, you can propose another small change.

Maybe we also need a balancing forum where we can create a separate topic for every balance related change request. This could be:

  • rework scout mechanics
  • decrease healing
  • lower Altanteans tower sight

Every change has to be independent of other changes. Otherwise it will be difficult to argue about whether they are needed or whether they improve balancing or not.

I mostly agree however there is one part missing. Suggestions should be based on evidence, this means that the statistical evidence need to be debiased. Problem is that our human feeling provides no good evidence as it is biased by personal playstyle, choice of maps and so on. So finding good performance indicators and evaluate them would be a better attempt on this.

Regarding scouts:

I don't really like the current mechanics.

There is only a really short amount of time when it is possible to attack an enemy military building disclosed by a scout. Especially when you play at a higher speed or have a large kingdom to manage, it is difficult to keep track of where the scout is and whether he is about to uncover the building you are aiming for. In my opinion it would be more important to increase the uncover time of the scout (we could increase the vision range or only fade out the uncovered area after some time).

Therefore they are probably cheap in maintenance. To increase this you could built more of them increasing the chance of overlkapping views. This would increase costs though which is suggested here.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 674
Ranking
One Elder of Players
Posted at: 2020-10-28, 13:24

hessenfarmer wrote:

Mars wrote:

  • scouts are unbalanced

I don't htink so as they are pretty much the same for all tribes.

There are always 2

  • Balance between the 4 tribes, aim: that Atlantean win as often as Imperium

  • Balance the game itself - which rational choices a player will make in a game


Top Quote