Page MenuHomeWildfire Games

Balance changes for Alpha 23
ClosedPublic

Authored by temple on Feb 4 2018, 4:22 PM.

Details

Summary

This is a balance patch for a23, discussed with Grugnas and Hannibal_Barca. It includes changes from their mods (Monkey Wrench and Vox Populi) and includes or supersedes some other patches (D834, D859, D904, D995).

Here are the changes:

  • cavalry capture rate from 2 to 1 (champion 5 to 2.5).
  • black cloak 20s train time.
  • mauryan elephant stables in village phase.

rP21285

  • cavalry train time from 12s to 15s.
  • champ infantry train time from 30s to 20s, champ cavalry from 40s to 30s.

rP21284

  • spear cav 1.5x bonus against cav, 2s repeat time, total dps from 5.4 to 5.5 (champ similar).
  • spear cav speed 1.1x, sword cav speed 1.2x (champ similar).

rP21282

  • ranged cavalry health from 120 to 100 (champ 270 to 240).
  • skirm and skirm cav spread from 3.0 to 4.0.
  • archers walk from 0.9 to 1.1, skirm walk from 1.5 to 1.4.

rP21281

  • naked fanatic 20s train time, 200hp.
  • thorakites 75w, 150m cost.
  • healers 200f cost, 150xp promotion.
  • juggernaut 5 pop.
  • siege tower 40s train time, 10m min range.

rP21280

  • embassy build time 150s, limit three.
  • military colony remove max arrow limit.
  • sentry tower 10m min range, 500 health, 500 capture points.
  • army camp 60m max range.

rP21268

  • rome team bonus only citizen-soldiers.

rP21269

  • pers hero darius aura +15% speed.
  • cart hero hamilcar aura 60m range, new aura -20% enemy mercenary attack in 30m range.
  • ptol hero ptolemy i aura -35% mercenary cost.
  • athen hero themistocles aura batch modifier 0.5.
  • sele hero seleucus from 80m to 60m range.

rP21271

  • advanced and elite units +20% loot.

rP21272

  • advanced and elite units ranged spread from -5% to -10%.
  • archery tradition spread -20%.
Test Plan

See that all are actually implemented and applied.
Argue about specific numbers.

(I had an error with arc after using svn mv, so this patch deletes rather than moves the hamilcar aura, but I'll fix it when committing.)

Event Timeline

There are a very large number of changes, so older changes are hidden. Show Older Changes
mimo added inline comments.Feb 4 2018, 9:37 PM
binaries/data/mods/public/simulation/data/auras/units/heroes/athen_hero_themistocles_1.json
8

Why not "multiply: 0.7"? 0.2 with a usual value of 0.7 is 30% reduction (i guess there are no protection in the code against negative values, so we should not allow the possibility).

binaries/data/mods/public/simulation/templates/special/player/player.xml
24 ↗(On Diff #5645)

The original goal of this limit was to restrict the possible units (forcing the player to choose which embassies to build) by preventing building all 3 embassies, and i still think it was a good decision.

Iirc the design was even to be able to only choose one embassy, but as long as we have no mecanism in the code to allow several embassies of the same kind once one is chosen, it was decided to temporarily limit at 2 embassies.

Still, some reasons why you need to change that would help.

binaries/data/mods/public/simulation/templates/structures/maur_elephant_stables.xml
6–7

I'm not specially opposed, nor favorable. Some reasoning for this change would help decision.

binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

Why removing that? colonies are really cheap compared to civil-centre, so having them a bit weaker is a priori good. Once again, some reasonning on why you do that would help.

binaries/data/mods/public/simulation/templates/template_unit_cavalry.xml
10

This i'm strongly against. That's not what we should change two weeks before a ff as it certainly has lots of implications on the ai and how it conducts its attacks.
In addition, it should be linked to a global rethinking of the capture attacks (the defender GarrisonRegenRate increase when phasing is imo too big, at least from village to town).

binaries/data/mods/public/simulation/templates/template_unit_champion_cavalry.xml
10

Same as before, too late to change it now (A small decrease, from 5 to 4 for example, would be ok is really needed, but i don't like such changes just before ff).

binaries/data/mods/public/simulation/templates/units/gaul_champion_fanatic.xml
8

That's not a balanced change, but a gameplay one. These were a special kind of champion less powerful but rapidly trained which i think help to diversify the civs and thus the gameplay. Please, explain the reasons to remove that.

temple added a comment.Feb 5 2018, 4:41 AM

These were things we three agreed on, or at least didn't object to.

binaries/data/mods/public/simulation/data/auras/units/heroes/athen_hero_themistocles_1.json
8

The batch time modifier is a weird thing. I can multiply by 0.7 instead, it doesn't matter.

binaries/data/mods/public/simulation/templates/special/player/player.xml
24 ↗(On Diff #5645)

Fine.

binaries/data/mods/public/simulation/templates/structures/maur_elephant_stables.xml
6–7

Because they can make worker elephants in the village phase. Not a big deal either way.

binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

Military colonies have other weaknesses, but I don't feel strongly so 15 is fine with me. Maybe Grugnas or Hannibal_Barca wants to argue about it though.

binaries/data/mods/public/simulation/templates/template_unit_cavalry.xml
10

Okay. I think capture and territory need to be rethought too.

binaries/data/mods/public/simulation/templates/template_unit_hero_cavalry_spearman.xml
16

1.5

binaries/data/mods/public/simulation/templates/units/gaul_champion_fanatic.xml
8

Fanatics have lower armor/health in exchange for speed. Black cloaks also used to train at 15s for no apparent reason. The other champions now train at 20s so it seems weird for these two to train at half the time = 10s. I can change those to back to 15s if you'd like, I don't have a strong opinion.

As I am also involved, I'll try to explain some changes:

binaries/data/mods/public/simulation/templates/special/player/player.xml
24 ↗(On Diff #5645)

The idea was to allow the player to build 1 of each kind of embassy for a total of 3 and eventually have a tech that allows to train rank 3 mercenaries since they can train from 1 building only.
While embassies have some appeal into being a special units training building, other civs can train mercenaries from no restriction structures like barrack.
Limiting 2 embassies of the same type ,for the moment, is fine anyway.

binaries/data/mods/public/simulation/templates/structures/maur_elephant_stables.xml
6–7

Having a dedicate building to train special units replacing storehouses or farmsteads without to sacrifice civic center production queue seemed nice.
While phase 2 is too late for such a building, elephants require 20 sec training time which is high in comparison to its value and a bunch of food that can be eventually gathered from berries out of territory influence ( made harder by opponent troops in mid game ).

binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

Having military colonies and civil center with different max arrow count ( not that +5 arrows or -5 makes a big difference anyway ) can be misleading as the number of garrisonable units is the same and it is supposed to be a military structure ( is there a reason why an economy building should be better into defending ? ).
also, as temple stated, it has other weaknesses.

binaries/data/mods/public/simulation/templates/template_unit_cavalry.xml
10

The reason is that cavalry and especially champion cavalry can capture and destroy with ease bypassing armies in late game or simply overwhelm any other infantry unit in early game and still capturing houses to prevent population growth.

Hannibal_Barca added inline comments.Feb 6 2018, 3:57 PM
binaries/data/mods/public/simulation/templates/special/player/player.xml
24 ↗(On Diff #5645)

I'm fine with the limit of 2.
Although maybe faster train speed for them could be nice.

binaries/data/mods/public/simulation/templates/structures/maur_elephant_stables.xml
6–7

Yet again, the time to build one of these structures demands workpower and probably more than it takes for a single worker elephant...
Also, this offers a selection among the Town Phase structures so players can snap out of the usual "a market, a blacksmith, a tower, a temple" monotone.

binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

Since they do have half hp and they are named military colonies then why not 20 arrows.

binaries/data/mods/public/simulation/templates/template_unit_cavalry.xml
10

As far as I am concerned, the cavalry capture rate issue isn't really that pressing and it can be postponed.

HMS-Surprise requested changes to this revision.Feb 6 2018, 4:00 PM
HMS-Surprise added a subscriber: HMS-Surprise.

what about bolts and walls?

This revision now requires changes to proceed.Feb 6 2018, 4:00 PM
temple updated this revision to Diff 5682.Feb 6 2018, 5:31 PM
temple marked 9 inline comments as done.
temple edited the summary of this revision. (Show Details)

Addressed mimo's concerns, but kept 20 arrows at military colony and village phase elephant stables.

temple added a comment.Feb 6 2018, 5:32 PM

what about bolts and walls?

Bolts have been fixed since a22. What's the issue with walls?

elexis added a subscriber: elexis.Feb 6 2018, 6:03 PM

Might want to commit some of the accepted things if there is no opposition soon and update the diff with the rest.

binaries/data/mods/public/simulation/templates/structures/cart_embassy_celtic.xml
4

k

binaries/data/mods/public/simulation/templates/structures/cart_embassy_iberian.xml
4

k

binaries/data/mods/public/simulation/templates/structures/cart_embassy_italiote.xml
4

k

binaries/data/mods/public/simulation/templates/structures/rome_army_camp.xml
18

Not convinced, this thing can be built in enemy territory.

binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

Barracks are military structures too and fire 0 arrows. Didn't see a convincing argument in either way.

binaries/data/mods/public/simulation/templates/template_structure_defense_tower_sentry.xml
12

I liked the minrange, but ok to remove it for now. (I have a dead patch on phab to give all buildings a minrange until the tech is researched)

18

Interesting to make sentr towers non-capturable, why not. What actually happens now when there is enemy territory? Health decay or still capture point decay or none of that?

binaries/data/mods/public/simulation/templates/template_unit_champion_infantry.xml
9

Why reduce champ train time from 30 to 20? It was intentionally changed as the games were always a champion-rushing game. Now we have lots of citizen soldiers and many champion games, no?

binaries/data/mods/public/simulation/templates/template_unit_hero_cavalry_spearman.xml
16

I like the (1.5) multiplier idea, but one has to be careful to not make the unit OP, i.e no other unit being able to counter this one. Especially worth to conisder when reducing the repeat time so drastically and buffing the actual attack on top of it..

binaries/data/mods/public/simulation/templates/template_unit_infantry_ranged_archer.xml
34

Not sure about making archers faster than average infantry. Especially since the idea with 90% walk speed must have been that they are good at distance but eventually can't run away once the player took a wrong decision.

binaries/data/mods/public/simulation/templates/template_unit_infantry_ranged_javelinist.xml
17

Should do some fights, possibly in the atlas sim, possibly using the "gift from the gods" cheat

35

40% walking speed bonus in comparison to general infantry definitely sounds sufficient. (Even 30% would still be the maximum infantry speed):

template_unit_infantry_melee_pikeman.xml:    <WalkSpeed op="mul">0.8</WalkSpeed>
template_unit_infantry_melee_spearman.xml:    <WalkSpeed op="mul">0.95</WalkSpeed>
template_unit_infantry_ranged_archer.xml:    <WalkSpeed op="mul">0.9</WalkSpeed>
template_unit_infantry_ranged_javelinist.xml:    <WalkSpeed op="mul">1.5</WalkSpeed>
template_unit_infantry_melee_swordsman.xml:    <WalkSpeed op="mul">1.05</WalkSpeed>
template_unit_infantry_ranged_slinger.xml:    <WalkSpeed op="mul">1.2</WalkSpeed>
binaries/data/mods/public/simulation/templates/template_unit_mechanical_siege_tower.xml
9

That's probably nice.

32

probably ok, 45s?

binaries/data/mods/public/simulation/templates/template_unit_support_healer.xml
5

One buff by itself should be sufficient if they need to be buffed.
Is it actually needed? Just because temple buildings can be more efficient doesn't inherently mean that we have to make these guys more efficient too. Could nerf the temple building instead.

binaries/data/mods/public/simulation/templates/units/gaul_champion_fanatic.xml
8

I like/d their ability to rush and raid. It's also one of the historic traits of the civ. I'd be ok with making cloaks 2-5s slower if that helps with differentiation. (Could be split from the diff too)

binaries/data/mods/public/simulation/templates/units/ptol_champion_juggernaut.xml
4

Ok for me, commit this juggernaut pop change already if you want.

In D1279#52335, @temple wrote:

what about bolts and walls?

Bolts have been fixed since a22. What's the issue with walls?

What do you mean? People play "no bolts" game. It's a common rule. 150 pierce damage is too much to handle. Maybe it could have less hp or something.

The issue is walls that people don't play that much with walls. It's expensive and takes too much time to build.

I believe that players shouldn't rule the game as "no cavs, no bolts, walls, no wonders... etc there must be some balance. No one likes bolts because it's overpowered, it's hard to kill.

Vulcan added a comment.Feb 6 2018, 6:30 PM

Successful build - Chance fights ever on the side of the prudent.

Updating workspaces...
Build (release)...
Build (debug)...
Running release tests...
Running cxxtest tests (309 tests).....................................................................................................................................................................................................................................................................................................................OK!
Running debug tests...
Running cxxtest tests (309 tests).....................................................................................................................................................................................................................................................................................................................OK!
Checking XML files...
Vulcan added a comment.Feb 6 2018, 6:31 PM
Executing section Default...
Executing section Source...
Executing section JS...
temple added a comment.Feb 6 2018, 7:31 PM

What do you mean? People play "no bolts" game. It's a common rule. 150 pierce damage is too much to handle. Maybe it could have less hp or something.

They were fixed with D767.

binaries/data/mods/public/simulation/templates/template_structure_defense_tower_sentry.xml
12

(I'm adding minrange, not removing it?)

18

They inherit, so still capturable but with 500 points.

binaries/data/mods/public/simulation/templates/template_unit_champion_infantry.xml
9

In rP18025 costs were increased, stats were decreased. Here we're only changing the training time, infantry from 30s to 20s where before it was 13s, cavalry from 40s to 30s where it had been 15s.
In a22 I rarely see champs. In a21 there were some, but also keep in mind that veteran mercenaries were new for a22 too (which players can make instead of champions).

binaries/data/mods/public/simulation/templates/template_unit_infantry_ranged_archer.xml
34

Why then can slingers and skirms run away?

mimo added inline comments.Feb 6 2018, 8:17 PM
binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

I've also no strong feelings about it, but expansion at phase 2 is really much easier with colonies compared to cc such that can give you a huge advantage to access more resources, and balancing it by less defensive power looks reasonnable to me.

temple added a comment.Feb 6 2018, 9:54 PM

Doing some testing, think we should hold off on the skirmisher spread changes, the other changes might be enough.

Seemingly a misunderstanding.

elexis added a comment.Feb 7 2018, 6:18 PM
In D1279#52402, @temple wrote:

What do you mean? People play "no bolts" game. It's a common rule. 150 pierce damage is too much to handle. Maybe it could have less hp or something.

They were fixed with D767.

That's still a23, not 22. Was fixed some days after the release.

Grugnas added inline comments.Feb 8 2018, 10:18 AM
binaries/data/mods/public/simulation/templates/template_structure_civic_civil_centre_military_colony.xml
5

in some situations military colony replaces the civil center ( i.e. the player loses his starting civic center) and there can't be researched the next phase. 20 Garrisoning slots and 15 arrows seems confusing somehow.

Hard to judge without more games, but maybe reduce spear cav attack slightly, maybe reduce cav speed slightly just so that rushes are a tiny bit slower and hunting is a tiny bit less efficient. Could probably lower ranged cav health a bit more.

Can you commit the parts that we got any kind of agreement on?
It would probably be good to split the parts that weren't discussed here yet to a separate patch.

temple edited the summary of this revision. (Show Details)Feb 20 2018, 12:56 AM
temple added inline comments.Feb 20 2018, 2:28 AM
binaries/data/mods/public/simulation/templates/structures/maur_elephant_stables.xml
6–7

If the elephant stables was a village building then you couldn't use it as one of the four required town buildings in order to advance to the city phase. So for that reason I think it's better to leave it in the town phase for now.

binaries/data/mods/public/simulation/templates/structures/rome_army_camp.xml
18

It's actually less than it seems because it's counted from the center of the building rather than its edge, and it doesn't have an elevation bonus whereas towers do. At 68m slingers can still attack it without being attacked back, if they're at the corner:

Should slingers be able to kill a military colony freely? I guess we could try 60m as a compromise.

temple edited the summary of this revision. (Show Details)Feb 20 2018, 5:10 AM

I've committed the changes listed in the summary. Thanks for everyone's input. I was hesitant about the spear cav buff so I fell back to 5.5 dps instead of 6.5. (They trade about evenly with sword cav, but they cost wood instead of metal so that's a win.)

I'm still concerned about rushes, particularly with spear cav. They shouldn't be impossible to defend, so maybe we'll need to reduce cav speed at least in the village phase. And of course we'll probably need to make some other tweaks as well. I guess the best way to find out is to play a lot of games! Hopefully lobby players will migrate to svn and we can get some feedback.

In D1279#53769, @temple wrote:

I've committed the changes listed in the summary. Thanks for everyone's input. I was hesitant about the spear cav buff so I fell back to 5.5 dps instead of 6.5. (They trade about evenly with sword cav, but they cost wood instead of metal so that's a win.)
I'm still concerned about rushes, particularly with spear cav. They shouldn't be impossible to defend, so maybe we'll need to reduce cav speed at least in the village phase. And of course we'll probably need to make some other tweaks as well. I guess the best way to find out is to play a lot of games! Hopefully lobby players will migrate to svn and we can get some feedback.

This is where the 150 hp instead of 160 could help. I wouldn't mind to let any civ to start with skiremsh cavalry ( even ptolemies with mercenary skirmish cavalry ) if spear cav rushes reveal themself troublesome.

temple edited the summary of this revision. (Show Details)Feb 20 2018, 4:27 PM
gaius22 added a subscriber: gaius22.Mar 8 2018, 2:25 PM

I think cavalry, and I'm very sure about that, is still much too strong, and champs in comparison too weak.
I would suggest that cav cost at least 125 Food 75 wood and 2 Pop. I would them also make them a little bit slower and/or give them less HP.
Hunting gather rate from 5 to 2.5 (like borg- advises). For that reason cost of goat or sheep from 50 to 40.
Champ Sword and Champ Spear: more Armor and 50 Food less costs.
Reduce range of Cav Archer and Cav Skrim: makes historically more sense and makes Spam and hit and run more difficult.
Make tower a tiny bit stronger.
Slinger Range from 42 to 32 and Javelin Inf attack from 16 to 13

Siege Weapons:
I would also reduce Armor from Rams: from 99 to 40. That is more reasonable in every way. And Rams are op.
Catapults and Bolts from 50 to 20 and Catapults attacks at least 1.5 times stronger.
Make Ships more expensive.

Civ Balancing:
Rome will be very strong in next A23 right now: Best Hero (2 Armor for every Unit and Building). And Spear Cav will be much stronger and more useful against Cav and Siege Weapons.
Suggestion: make Army Camp more Expensive and nerf Hero.
Carthage: reduce Temple cost to normal 300 Stone. Carthage is weak enough.
Iberians: Reduce Building time of fortress from 750 to 500
Ptol: improve Champ inf: otherwise its a useless unit.
Make Miltitary colony 25% more expensive.

If no changes will be made, there would again only few Cav Civs playable (Persian, Sele, Ptol plus Rome).

Hope my suggestion helps. If wanted: i could code my suggestion also in the xml templates.

temple abandoned this revision.Mar 8 2018, 8:43 PM

Thanks for the suggestions, gaius22. (There's a patch for the hunting gather rate at D1351.)

temple reclaimed this revision.Mar 8 2018, 8:47 PM
Hannibal_Barca accepted this revision.Mar 8 2018, 8:51 PM

Seems like this has been done thoroughly and efficiently.

This revision is now accepted and ready to land.Mar 8 2018, 8:51 PM