well, it's not "in certain cases"; it is almost for every mid-range city :
forge is sufficent. (+20%)
+ civic nationalism (+10%)
(old warrens gave 130%total for 2units : roughly 260% prod per unit.
new warren : 100+100+30: 230% /per unit)
in those cases the "new" warrens are already a bit less powerful.
(but more powerful with pacifism and apprenticeship (-20 / -10%: total -30 : +100 : 170%instead of 70*2=140)
for heroic epic (+forge) you would have had : 100(base)+100(HE)+20(forge) : 2units = (roughly) 440% prod (+340%)
now : heroic epic + forge + warrens : +220.total 320%
maybe just putting warrens at 150% would be sufficent to mitigate the heroic epic case and as it would do more than compensate in forge cities, it would average ?
and as all units get the free xp/promotions : this would balance with the slightly lower production
I don't think that we should balance the Warrens having in mind cities with the Heroic Epic. You are right about the forges and other lesser production modifiers; the new implementation reduces the efficiency of Warrens in high production cities but I believe this is partially offset by the fact that the Clan is one of the civilizations that can abuse Multiple Production more easily. By the way, units duplicated by the old implementation of the Warrens already got all of the goodies they would have gotten if they were built normally (at least in More Naval AI).
[to_xp]Gekko;13056828 said:
it should be noted that producing stuff a lot faster is more useful than building two at the same time, getting stuff out faster gives you tons of flexibility. so it's debatable whether it's a buff or a nerf. not a big issue since it can be easily balanced by raising/lowering the bonus after proper playtesting
I seem to remember that the bug with warrens not giving proper xp to doubled units was fixed a while ago but I don't usually play clan, did anyone test that lately? not sure if the behaviour was intended or a limitation honestly. Fall Further changed warrens by making the units weaker but that was overly complicated, a simple -10% strength promo should suffice for example.
That's my point of view, mostly. As you mention, any rebalance should take into account the increased flexibility, specially in the early stages of the game, in which as I mentioned the new implementation gives the Clan an edge by allowing them to create faster workers and settlers. Jonas Endain is specially powerful now in this aspect.
I'd rather not introduce new promotions; that would make the AI code a bit more confusing. I believe we should be able to rebalance the new implementation by tweaking how it modifies production only.
[to_xp]Gekko;13056828 said:
about uniques, the manual states that "warrens do not double any Hero, National, Siege or Naval units" so if the behaviour in EMM could mimic this is would be best imo
"only living" removes at least siege and naval..
and on top of that removes chances to boost undead ones (what did the old warrens do on those ?)
I believe that the old warrens doubled things like Diseased Corpses and Stygian Guards but I'm not entirely certain. However the 'living only' certainly makes more sense! ^_^
I'm not sure about how it works in vanilla Fall from Heaven 2, but in More Naval AI the Warrens only affect units that are world units, not national units, not mechanical and that are alive. The actual source code that checks which units to duplicate in the old implementation is:
Code:
if isWorldUnitClass(unit.getUnitClassType()) == False:
if isNationalUnitClass(unit.getUnitClassType()) == False:
if not unit.isMechUnit():
#if unit.getUnitCombatType() != UnitCombatTypes.NO_UNITCOMBAT:
if unit.isAlive()
Therefore, both the More Naval AI implementation and the new one in ExtraModMod affect the same units, except for world and national units. I agree that the new implementation should not work with World and National units either.
That being said, it *is* still additive rather than multiplicative as you have come to realize ... I feel as though it would be best if it did act upon the 'final' production, and perhaps had no effect on heroes and national units.
Although possible, that would complicate how production is evaluated in the DLL, and how is it displayed in the information panels in the city GUI. Even with these cons, I also believe that one of the best ways to balance the new implementation is to make it as close to the original implementation as possible by making Warrens multiply the final production instead of being added to the rest of multipliers and then used to multiply the raw production. Given the increased flexibility and faster expansion in the early game that the new implementation provides (along with Multiple Production), I don't think that in this case the bonus should be of 100%, though.
I'm going to be playing another big MP game on the 22nd and can play Clan if people would like an MP run through. Let me know if there is any other playtesting you need done, Terkhen: I can try to convince people to play those specific races.
Great! The Kuriotates have not been playtested properly in MP in either More Naval AI or ExtraModMod for years, so that would be another race that would require testing. Other than that, I can't think of any specific civilization that requires playtesting. Thank you, and have fun!
It seems like this would slightly nerf the RoK hammer multiplication trick? That would be the biggest impact. Otherwise I don't really see the issue: you lose a little bit of production, but potentially get more active turns out of the units and don't waste overflow. Traditionally FFH assumes higher-tier units are more cost effective, so producing a single high tier unit twice as fast should be preferrable to two low tier units in the same time.
rebounding on blakmane :
there is the RoK hammers... (but no more than for other units)
(with heroic epic, with 45hammers you could get 2 SoK ==> 90hammers worth).
now : HE + warrens : need 60 hammers to get 2units : 90hammers : the return is less. 300% prod instead of 400%)
Yes, it reduces its efficiency. In my opinion using Soldiers of Kilmorph to increase you production is not an intended effect of the Warrens, though; and a micromanagement based production strategy in a game devoid of any other similar production strategies always felt out of place to me (specially in a turn-based strategy game in which production micromanagement requires no skill and only increases waiting time for other players).
Because of these problems I thought about disabling the effect of Warrens for units which can be sacrified in order to generate production, but since this is not something that affects my games much and it is usually regarded as a valid strategy by most players I decided to not implement this change.
Judging from Calavente's numbers (specially if for the worst case scenario they are taking Heroic Epic into account, which will not be built in all of unit production cities anyways) it still seems like a valid strategy with the current implementation, though.
and .. buying with gold.
Before, it you bought with gold, you bought 1 and got 2.
now you'll still need to buy 2 as iirc the gold price is not reduced by the %production.
--> maybe gold price could depend on %production
(not to say it is a bad idea.. )
I wasn't aware that the Warrens also duplicated bought units. That is something that in my opinion does not make sense. Why should units which have not been raised in the Warrens benefit from their effects? It sounds as if slave traders and the Guild of the Nine did 2x1 sales to the Clan because they had rough childhoods. IMO preventing this is a positive change of the current implementation.
In conclusion, I propose two different implementations:
- The Warrens increase production of living units (except national and world units) by 120%. This would just give a slight buff to the current implementation. The extra 20% is meant to alleviate a bit the production reduction in cities with high production modifiers created by the new implementation, but without making the improvements to the early game too crazy.
- The Warrens multiply production of living units (except national and world units) by 80%. The reduction in 20% takes into account that without it this implementation would be more powerful than the old one; increased versatility, half of the cost for workers and settlers (one quarter cost settlers for Jonas Endain!), and with certain setups I reckon that it would easily allow to build three or even four weak units each turn.
I believe that in order to discuss which option is better, tweaks to them or proposing a new one, the current implementation should be playtested first. My own testing games make me prefer the first one since I barely noticed the problems mentioned in this thread and it is cleaner from a code point of view, but I wouldn't mind using the second one either. The second implementation worries me by how much it would buff production oriented gameplay for the clan, though.
![[to_xp]Gekko](https://civfanatics-data.community.forum/avatars/m/74/74677.jpg?1475248346){kind=avatar}
let's try it out with +100% production and see how it feels 
