Currently Online

Latest Posts

Topic: Amazons balancing

tothxa
Avatar
Topic Opener
Joined: 2021-03-24, 12:44
Posts: 487
OS: antix / Debian
Version: some new PR I'm testing
Ranking
Tribe Member
Posted at: 2023-09-12, 19:18

Re https://www.widelands.org/forum/post/40728/

tothxa wrote:

The Amazons should probably require gold for one more training level, and should use rations for some training to force them to build charcoal kilns and food preservers.

A draft PR is up at https://github.com/widelands/widelands/pull/6150

Comments are welcome. face-smile.png


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 967
Ranking
One Elder of Players
Posted at: 2023-09-12, 23:19

This seems a bit unlogical, the problem is more that with Frisian and Barbarian Training without gold is nearly impossible, and those both are too weak

You see the goldproblem with bar and fri even more on water maps where you need harbors


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 967
Ranking
One Elder of Players
Posted at: 2023-09-12, 23:22

Amazons finally are a tribe that is quite well balanced, but we have the problem still that 2 tribes are extremly weak


Top Quote
Nordfriese
Avatar
Joined: 2017-01-17, 18:07
Posts: 2058
OS: Debian Testing
Version: Latest master
Ranking
One Elder of Players
Location: 0x55555d3a34c0
Posted at: 2023-09-13, 19:18

I didn't look at the PR yet, but I am much in favour of making Amazons either weaker or more expensive. The match I played against the-x today (Fri vs Amz) illustrated clearly that there's a balancing problem – an amazon level 7 solder (trained without gold) is approximately equal to a frisian level 10 soldier, even though the amz7 is way cheaper.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 967
Ranking
One Elder of Players
Posted at: 2023-09-13, 19:29

If you look at the fight simulator, a Frisian is stronger. If we keep lowering Amazons we only make Atlanteans and Imperium stronger and indeed more imbalanced. In practical 1 v 1 Multiplayer Games the main problem with balancing are still rookies. Try to play Barbarian v Amazon on any 2 Player Map.

In all multiplayer Widelands games, the following tribes were chosen in a 1v1 game:

  • 35 x Atlantean
  • 21 x Empire
  • 16 x Amazon
  • 1 x Frisian
  • 0 x Barbarian

Making Amazon weaker doesnt solve this problem


Top Quote
kaputtnik
Avatar
Joined: 2013-02-18, 20:48
Posts: 2554
OS: Archlinux
Version: current master
Ranking
One Elder of Players
Location: Germany
Posted at: 2023-09-14, 08:02

the-x wrote:

In all multiplayer Widelands games, the following tribes were chosen in a 1v1 game:

  • 35 x Atlantean
  • 21 x Empire
  • 16 x Amazon
  • 1 x Frisian
  • 0 x Barbarian

I remember a tournament match between WorldSavior (Barbarians) and me (Empire) where WorldSavior had beaten me. So your statistic is probably wrong and since Amazons are available only since version 1.1 and Frisians since version Build20, comparing over all matches is useless. And without comparing the winning tribe it's useless at all.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 967
Ranking
One Elder of Players
Posted at: 2023-09-14, 10:35

The statistic is right and as you see by the low numbers all matches from my experienced 1v1s were counted -> and yes, its relevant, i had so many players who played Empire in 1v1s against me, also many Atlanteans games and in the same time almost no one chosing Bar or Fri.


Top Quote
the-x
Avatar
Joined: 2019-01-19, 13:23
Posts: 967
Ranking
One Elder of Players
Posted at: 2023-09-14, 10:43

Imagine, when you build a car, you see the car as a whole. If you start balancing one part without recognizing the other, this one part might be perfectly balanced on iteself, but does it work in practical?


Top Quote
Nordfriese
Avatar
Joined: 2017-01-17, 18:07
Posts: 2058
OS: Debian Testing
Version: Latest master
Ranking
One Elder of Players
Location: 0x55555d3a34c0
Posted at: 2023-09-14, 10:56

"all multiplayer Widelands games" is not the same as "all multiplayer 1v1 games involving the-x". If you pick an arbitrary sample you can prove just about anything. The first Tournament round also had a game where one player chose Barbarians and won against Atlanteans.

However all this is completely beside the topic of this thread and does not change the point that the Amazon soldiers' strength-to-cost ratio is much higher than for other tribes. Your car metaphor actually underlines this, Amazons are well balanced in and against themselves but not so well compared to the rest of the Widelands world populace.

I think in additions to making amazons more expensive, Frisian soldiers should also be strengthened – they are still the most expensive even with these changes and should therefore also be strongest, and a level 7 amazon having equal chances against a level 10 frisian is therefore badly unbalanced. Or make the amazons weaker. However I am aware that any changes to soldier stats are likely to cause more problems than they solve, so keeping the strengths as they are and only changing the costs is probably the safer way to go.

Regarding the PR, most changes look good (not tested); except the rare tree kiln, which seems illogical to me, I don't see how using rubber would contribute to cheap charcoal, and I am not sure if balsa is also such a superior wood type for charcoal burning. Maybe the rare tree kiln could consume 3 ironwood + 2 log + 1 balsa → 2 coal or some such?


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2747
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2023-09-14, 11:28

from the values of our (my) balancing tool the desired goal of soldier balance was to make level 10 (9 for Atlanteans) balanced regarding the chances to win.
the problem with amazons (especially versus frisians) is that amazon fighting is the most luck based fighting of all tribes as their strongest abiity is evade. So in 10000 fights the values are balanced but the feeling could be the opposite as the luck may be a strong factor in a match only having 100 fights.


Top Quote