remaining resistance balance issues
Moderator: Forum Moderators
remaining resistance balance issues
I ran the attached dr Perl script against the units in wesnoth-0.8.8 and most of the campaigns from the campaign server. It appears there are still some remaining balance issues, where there exist specific attack types which generate day-night swings of damage of more than 3 times against some movetypes (for unled, non-neutral units). These seem to relate to resistances of armoredfoot, deepsea, swimmer, undeadfly, undeadfoot, treefolk, drakefoot, drakefly and scuttlefoot movetypes. The resistances of the worst offenders, undeadspirit and spirit, were recently fixed in CVS, but these others still seem problematic.
The archive includes the verbose output as well as an edited summary with the specific attack type/movetype combinations which seem problematic. As an example, which exhibits a factor 4 difference between day and night.
The archive includes the verbose output as well as an edited summary with the specific attack type/movetype combinations which seem problematic. As an example,
Code: Select all
cold 12-3 magical: Lich
against -25 0 25 50 75 100
--------------------------------------
deepsea 2 5 8 11 14 17 <- 4.00
swimmer 2 5 8 11 14 17 <- 4.00
undeadfly 2 5 8 11 14 17 <- 4.00
undeadfoot 2 5 8 11 14 17 <- 4.00
- Attachments
-
- dr14.zip
- (28.07 KiB) Downloaded 628 times
We need a new damage type, plain and simple. Mermen absolutely should be resistant to cold, but they should not be resistant to "evil magic". We need the cold resistance for any creatures from the north, which we currently cannot create because wesnoth's damage types were made with a much smaller set of units in mind.
We need a shadow damage type, and a poison damage type. It also wouldn't hurt to have a general "magic" damage type, for units which use magic but do not have it as any elemental type. I think that level-1 mages would be good candidates for this.
We need a shadow damage type, and a poison damage type. It also wouldn't hurt to have a general "magic" damage type, for units which use magic but do not have it as any elemental type. I think that level-1 mages would be good candidates for this.
Holy damage is already in game. It's used by Paladins, Mages of Light and such.SaintDust wrote:Shadow-magic(Dark)
Light-magic(Holy)
Slow-magic(base magic)
Poision
We need to think of something better than base magic, or standard magic for the orginal nuetral magic type
I'm not sure that I see the need for a slow damage type, seeing as slow is an effect, rather than a damage dealer.
-Gafgarion
"Language is the source of misunderstandings." -Antoine de Saint-ExupéryElvish Pillager wrote:Normal Trolls use clubs, not ostriches.
Same with poison... WTF would poison magic be?
FIre is basic magic, BTW.
FIre is basic magic, BTW.
For I am Turin Turambar - Master of Doom, by doom mastered. On permanent Wesbreak. Will not respond to private messages. Sorry!
And I hate stupid people.
The World of Orbivm
And I hate stupid people.
The World of Orbivm
This isn't just about swimmers. These problems are due to weird interactions between high resistance and additive day/night damage modifiers.
Looking more closely at the data, and removing non-aligned units from the lists, there are four attack types that exhibit damage with greater than a factor of 3 variance between day and night. I think six of these attacktype-movetype combinations need to be revisited.
First, blade vs armoredfoot. Blade is a very common attack type, while armoredfoot applies to the human infantry and its derivatives. This combo shows variances between 3.17 and 4.
Second, cold attacks vs each of deepsea (2 sea monsters), swimmer (20 Mermen, Mermaids and Naga), undeadfly (bats, Dark Spirit and Vampire Lady) and undeadfoot (most undead). Variances between 3.5 and 6.
Third, fire vs drakefly and drakefoot. This only applies to Thunder Spirit from the Flight to Freedom drake campaign, showing variance 3.33. This one probably isn't worth bothering with.
Fourth, the common pierce vs treefolk (Woses).
The basic logic here is that aligned units should be more powerful during the time of day favourable to their alignment, and less powerful during the adverse time of day, BUT within limits. It makes no sense to have huge swings in damage due to alignment. Most units see a reasonable 25% difference, but units that do lots of damage, against certain movetypes with high resistance to that damage type, can see ridiculous differences between day and night. Should a duelist really do 10 damage per swing against a tree during the day and 2 per swing at night? Or 4 damage against an Infantryman during the day, but only 1 at night? I think not.
Looking more closely at the data, and removing non-aligned units from the lists, there are four attack types that exhibit damage with greater than a factor of 3 variance between day and night. I think six of these attacktype-movetype combinations need to be revisited.
First, blade vs armoredfoot. Blade is a very common attack type, while armoredfoot applies to the human infantry and its derivatives. This combo shows variances between 3.17 and 4.
Code: Select all
5: Mage,Crow,Direwolf Rider,Sergeant,Young Ogre,Rogue Mage,Cavalryman,Sea Orc,Gargoyle,Wolf Rider,Ghoul,Assassin,Huntsman,Huntsman-Peasant,Orcish Slayer,Duelist
1 3 4 5 6 7 <- 4.00
9: Wolf,Orcish Grunt,Mounted Commander,Direwolf Rider,General,Dark Spirit,Spectre,Halbardier,Shadow Lord
2 5 7 9 11 13 <- 3.50
13: Pirate Captain,Pirate,Orcish Warlord
3 7 10 13 16 19 <- 3.33
21: Saurian Margrave
5 11 16 21 26 31 <- 3.20
25: Skeletal Dragon
6 13 19 25 31 37 <- 3.17
Code: Select all
6: Rogue Mage,Spectre,Saurian Soothsayer,Ancient Lich
1 2 4 5 7 8 <- 4.00
7: Initiate,Dead_Costly_Delusionist,Delusionist,Costly_Delusionist
1 3 5 6 8 10 <- 5.00
8: Shadow Mage,Saurian Icecaster
1 3 5 7 9 11 <- 5.00
9: Dark Spirit
1 4 6 8 10 13 <- 6.00
10: Dark Adept
2 4 7 9 11 14 <- 3.50
11: Deathmaster
2 4 7 10 13 15 <- 3.50
12: Demilich,Shadow Lord,Necromancer,Lich,Demilich
2 5 8 11 14 17 <- 4.00
13: Ancient Lich
2 5 8 12 15 18 <- 4.00
15: Dark Queen
2 6 10 14 17 21 <- 5.00
Code: Select all
13: Thunder Spirit
3 7 10 13 16 19 <- 3.33
Code: Select all
6: Spearman,Soul Shooter,Death Knight,Cuttle Fish,Merman Hunter,Saurian Jarl,Bowman,Merman Fighter,Noble Commander,Merman Netcaster,Commander,Trapper-Peasant,Saurian Margrave,Trapper,Saurian Knifethrower
1 2 4 5 7 8 <- 4.00
6: Horseman Commander
2 5 8 11 14 17 <- 4.00
7: Saurian Flanker,General,Sea Hag,Merman Entangler,Bone Shooter,Merman Warrior,Merman Spearman
1 3 5 6 8 10 <- 5.00
8: Merman,Wall Guard,Orcish Crossbowman,Noble Lord,Grand Marshal,Pirate Captain
1 3 5 7 9 11 <- 5.00
9: Transport Galleon,Giant Scorpion,Merman Hoplite,Clipper,Cave Naga,Old Orcish
Shaman,Vampire Lady,Lord,King of Wesnoth,Pirate Cruiser,Pirate Galleon
1 4 6 8 10 13 <- 6.00
9: Chocobone,Horseman,Lancer Commander
3 7 12 16 21 25 <- 4.00
10: Pirate Schooner,General,Galleon,Triton,Longbowman,Merman Javelineer,Huntsman,Huntsman-Peasant
2 4 7 9 11 14 <- 3.50
11: Soul Shooter,Orcish Ruler,Master Bowman
2 4 7 10 13 15 <- 3.50
12: Lancer
4 10 16 22 28 34 <- 4.00
14: Knight
4 11 18 25 32 39 <- 4.50
14: Pikeman,Orcish Sovereign
2 6 9 13 16 20 <- 4.50
15: Dragoon,Duelist
2 6 10 14 17 21 <- 5.00
15: Paladin
5 12 19 27 34 42 <- 3.80
17: Cavalier
3 7 11 15 20 24 <- 3.67
17: Grand Knight
5 14 22 31 39 48 <- 4.40
18: Saurian Jarl,Sea Serpent
3 7 12 16 21 25 <- 4.00
30: Saurian Margrave
5 12 20 27 34 42 <- 4.00
-
- Posts: 873
- Joined: July 4th, 2004, 9:14 pm
- Location: My imagination
- Contact:
Horseman led by Captain charging at Elvish Scout in the daytime did so much damage that it could kill in one blow.Circon wrote:Somebody remind me why percentages were made additive?
Play a Silver Mage in the Wesvoid campaign.
-
- Retired Developer
- Posts: 2633
- Joined: March 22nd, 2004, 11:22 pm
- Location: An Earl's Roadstead
NOT POISON MAGIC. Gah. Just plain old poison. aka, an assassin gets hit with a poison dagger, but because of his years of working with the stuff, he has developed a partial immunity to it, causing it to affect him only half as much as it does everyone else.turin wrote:Same with poison... WTF would poison magic be?
This would also be ESPECIALLY useful for monsters, and one of the primary reasons I would like to add this is to give it to the naga. The naga should be extremely resistant to poison.
I also want to add that I do not want the hack for "non-living" to be used there, because that would bork any other legitimate use of non-living.
Could be, but it seems a bit odd in certain cases. I'm no so adamant about this one being added, but it would make things a lot more intuitive, and would make for a damage type which no one is unusually resistant or vulnerable to.turin wrote:Fire is basic magic, BTW.
Except, perhaps, for mages, and magical creatures.
I understand the general gist here, but what are bogi and boni?Darth Fool wrote:Additive percentages seem reasonable for boni, but they result in problems for bogi. Multiplicative percentages have the opposite problem, ie for boni they end up with strange
values, but for bogi they are more reasonable..
-
- Retired Developer
- Posts: 2633
- Joined: March 22nd, 2004, 11:22 pm
- Location: An Earl's Roadstead
Getting back to the topic at hand, anyone have ideas for what to do about the huge day/night swings that remain?
Based on the huge amount of changes and balancing since 0.7.1 I think it's too late to go back to multiplicative modifiers before 1.0. We are stuck with additive calculations for the moment.
The only way to restore sanity to out-of-whack additive calculations is to
Based on the huge amount of changes and balancing since 0.7.1 I think it's too late to go back to multiplicative modifiers before 1.0. We are stuck with additive calculations for the moment.
The only way to restore sanity to out-of-whack additive calculations is to
- * reduce resistances to more reasonable values, as was done with undeadspirit (60% down to 40%),
* to make the relevant units neutral, or
* to change the affected attack from a few damaging swings to more swings each doing less damage but retaining the same overall damage potential.