Bombardment with siege units, from discord

[wieder]

WahazarToday at 6:23 PM
I had testing bombardment flag for catapults, artillery and howitzer playing 150t game based on LT44 ruleset.
Optimal parameters were:
catapult: FP 1, BR 1
cannon: FP 2, BR 1
artillery: FP 2, BR 2
howitzer: FP 2, BR 3
(FP: fire power, BR: bombardment rate), attack, defence untouched
Optimal parameters means: similar number of units needed to conquer city if normal reuleset used
Additionally I used IgWall flag. Without this, you need more attacking units to conquer city than normal ruleset.
Bombardment flag is nice feature, at least knights/cavalry are really needed.

[Corbeau]

Ok, I'll put mine here, too, then.

No IgWall, no x2 attack vs. city.

I'm doing a recalculation of bombard units using the tool I created. What follows are the new values, along with projected attack damage. Note: the lower Bombard Rate, the greater RNG effect.

Damage projection:
(corrections made 2018/07/28)

Archers: Attack: 1, Bombard Rate: 5
vs. Green Phalanx in a city on hill with walls (Defence: 11.25): 0.4
vs. Green Phalanx on grassland (Defence: 2): 1.7


Catapult: Attack: 1, Bombard Rate: 9
vs. Green Phalanx in a city on hill with walls (Defence: 11.25): 0.7
vs. Green Phalanx on grassland (Defence: 2): 3

Cannon: A: 1, BR: 12
vs. Green Musketeers (10 HP) in a city on hill with walls (D: 16.875): 0.7
vs. Green Musketeers on grassland (D: 4): 5


Artillery: A: 3, BR: 12
vs. Green Riflemen (20 HP) in a city on hill with walls (D: 22.5): 1.4
vs. Green Riflemen on grassland (D: 4): 5

Howitzer: A: 5, BR: 12
vs. Green Riflemen (20 HP) in a city on hill with walls (D: 22.5): 2.2
vs. Green Riflemen on grassland (D: 4): 6.6

[fran]

vs. Green Phalanx in a city on hill with walls (Defence: 9): 0.6

defence
pha 2
hill x 1.5 = 3
city x 1.5 = 4.5
walls x 2.0 = 9
fortified x 1.5 = 13.5

afaik mil unit in city gets fortification bonus, which is not
the same as city bonus. hope to get corrected if I'm wrong.

btw, what means 0.6 damage. 0.6 hitpoints?
Is there a calculator for bombard attack somewhere?

[Corbeau]

Units in city don't get additional fortification bonus, so that would be 9.

And not a calculator as such, but a helping tool: http://forum.longturn.org/viewtopic.php?id=998

[fran]

Units in city don't get additional fortification bonus, so that would be 9.

How did cgalik call that? Right ... share and learn

doc/freeciv-2.5.11/README.ruleset_civ2civ3 :

FORTIFIED +50% (=)
(only Land and Big Land units can fortify; inside cities, they are
always considered fortified)

Total Defense = (UNIT DEFENSE) * (100+TERRAIN)/100 * (100+RIVER)/100
* (100+CITY+BASE)/100 * (100+FORTIFIED)/100 * (100+VETERAN)/100
(Same as classic rules).

EXAMPLE: Riflemen fortified in a fortress, on mountains, with a river,
will get a total defense of 30 against land attacks:

Riflemen: defense 4
Fortified (+50%): 4 + 2 = 6
In Fortress (+100%): 6 + 6 = 12
On Mountains (+100%): 12 + 12 = 24
With River (+25%): 24 + 6 = 30.

EXAMPLE: Riflemen in a size 9 city, with walls, on hills,
will get a total defense of 27 against land attacks:

Fortified (+50%): 4 + 2 = 6
On Hills (+50%): 6 + 3 = 9
In City with Walls (+200%): 9 + 18 = 27.


After all, we have to adjust our learning curve so we can beat both natural death and starvation by breakdown of ecosystems due to global warming.

[fran]

This is the part I don't understand:

size9 city

In City with Walls (+200%): 9 + 18 = 27.

Why 200% and not 450%?
1.5 city x 1.5 size9 x 2 walls = 4.5

So everything is multiplicative, just size9 and walls is treated as additive.
In other calculations city_base and walls are treated as multiplicative.

[Corbeau]

That's 200% in relation to the previouis value.

Also, 200% is ruleset-specific. I think that is changed to 100% in the version we're playing.

Also, "In City with Walls" is a single item meaning "walls around a city add X%", in this case 200%.

[fran]

I suggest reading doc/freeciv-2.5.11/README.ruleset_civ2civ3

The line I pinpointed is about a city with city_base, size9_bonus and walls.
The alternative is to treat it additive 50+50+100=200 like above file maintains
or multiplicative 1.5*1.5*2=4.5 which I think is right.

To decide that somebody like pneu would need to pinpoint where this is in the sources.

ok ok ok, if
Green Phalanx in a city on hill with walls (Defence: 9)
is correct it's likely because
basic city defense bonus and walls are indeed additive because

Total Defense = (UNIT DEFENSE) * (100+TERRAIN)/100 * (100+RIVER)/100
* (100+CITY+BASE)/100 * (100+FORTIFIED)/100 * (100+VETERAN)/100

does not mention walls at all. So city stuff is additive, only above are factors.
If so, my original calculation was wrong.

Still, effects.ruleset has

[effect_city_walls_0]
type = "Defend_Bonus"
value = 100
reqs =
{ "type", "name", "range"
"Building", "City Walls", "City"
}
nreqs =
{ "type", "name", "range"
; "UnitClass", "Land", "Local"
; "UnitClass", "Small Land", "Local"
; "UnitClass", "Big Land", "Local"
; "UnitClass", "Merchant", "Local"
"UnitClass", "Sea", "Local"
"UnitClass", "Trireme", "Local"
"UnitClass", "Helicopter", "Local"
"UnitClass", "Air", "Local"
"UnitClass", "Missile", "Local"
}


and if it would be additive and city + walls shall be 100 it would need to be 50.

I'm lost.

[Corbeau]

I'm not sure what you don't understand. +200% is equal as *3 and that is the value that is being shown there. We are not playing with +200%/*3 for city walls, we are playing with +100%/*2.

Why is it written in one way and not the other? I don't know. Because whoever is reading it should be able to calculate the final defence value using both methods, maybe.

[Corbeau]

Ok, I re-read your post, let me try to explain more generally.

It is never additive. If there are two bonuses and each is +100%, that's not 100% + 100% = 200%. That's +100% once and the other 100% is calculated with the result.

So, if, for example, base value is 2, then +100% gives 4. But the second +100% is with relation to 4 and it then gives 8.

[Wahazar]

"similar number of units needed" is not really similar, because in the normal ruleset a lot of those units would be lost and needed to be rebuilt again, while here they survive and can be reused.

Of course, so that is a reason, why IgWall flag need to be discussed.

Proposed stats, together with IgWall, yield "similar number" of siege units needed to almost annihilate city defence in one turn. Of course bombardment does no harm to used artillery (if no counter-attack occurs), and make simultaneous damage, on the other hand, bombardment cannot kill defenders, thus additional units are needed to finish job (infantry, cavalry etc). Overall tactics is much more interesting than standard "catapult spamming".

Without IgWall flag, balance is shifted in favour of city defenders, which also can use artillery against siege units, and such counter-attack makes more damage (and it is fair, because defenders need to shoot in many directions - it reminds me Colonel Puller famous quote).

In my opinion, IgWall should apply to howitzers and artillery only.

BTW, if bombarding units is attacking non-native tile, normal duel occurs.

[Corbeau]

Balance will shift only as much as you permit it to shift by setting the numbers. I think that artillery shouldn't be specialised for attacking cities, but for general purpose attacks, softening the defence lines and so on. Attacking a city - that is, a well fortified position - is in reality left to specially trained units. Giving the artillery IgWall flag is "getting away with it the easy way". In reality, long-range units are not at all effective against well fortified positions such as cities. You need to pull in a lot of infantry to conquer a city.

[fran]

It is never additive.

If it's never additive this

EXAMPLE: Riflemen in a size 9 city, with walls, on hills,
will get a total defense of 27 against land attacks:

Fortified (+50%): 4 + 2 = 6
On Hills (+50%): 6 + 3 = 9
In City with Walls (+200%): 9 + 18 = 27.

is wrong, because city size 9 with walls has 450%.

[Corbeau]

We are not playing with the ruleset that is used in this example. We are playing with the ruleset where city with walls gives +100% (or *2).

[fran]

Caedo answered on discord so I copy this here:

"All defense bonuses gained from city improvements etc. are done via effects and found in the specific ruleset's effects.ruleset file. Since different effects of the same type (in this case Defend_Bonus) are added, this kind of defense bonuses is additive
So instead of e.g. a x1.5 defense multiplier, something would be a +50% defense bonus, i.e. "plus fifty percent of the base defense""

Back to the beginning,

Green Phalanx in a city on hill with walls

has
base 2
hills x 1.5 = 3
fortified x 1.5 = 4.5
city/walls x 2.5 =11.25

[Corbeau]

Corrected the second post.

[Wahazar]

Balance will shift only as much as you permit it to shift by setting the numbers. I think that artillery shouldn't be specialised for attacking cities, but for general purpose attacks, softening the defence lines and so on. Attacking a city - that is, a well fortified position - is in reality left to specially trained units. Giving the artillery IgWall flag is "getting away with it the easy way". In reality, long-range units are not at all effective against well fortified positions such as cities. You need to pull in a lot of infantry to conquer a city.

Balance will shift only as much as you permit it to shift by setting the numbers.

You cannot shift balance between bombardment of unit in walled cty vs bombardment unit outside city, using numbers. IgWall flag is for that.
Modern artillery without IgWall flag looks rather silly against ancient city walls.
On the other hand - if current ruleset include upkeep payment for city walls, lack of IgWall flag for any artillery is justified.

I think that artillery shouldn't be specialised for attacking cities, but for general purpose attacks, softening the defence lines and so on. Attacking a city - that is, a well fortified position - is in reality left to specially trained units. Giving the artillery IgWall flag is "getting away with it the easy way". In reality, long-range units are not at all effective against well fortified positions such as cities. You need to pull in a lot of infantry to conquer a city.

Artillery with IgWall flag is not specialised against cities - it is unversal unit. Without IgWall flag, is just worse for attacking cities than field units. And all these infantry job is an intrinsic effect of Bombardment flag - you need infantry to kill units softened by bombardment.

BTW, I noticed that people complain about archers - seems that they had no clue how to use that unit - archers were for active defend against attacking units, not for attacking fortified phalanx in a city.

[Corbeau]

You keep failing to acknowledge a known military history fact: artillery is ineffective against cities, fortified or not. It is used for harassment, but rarely as a defence softener. Check Stalingrad and Arnhem in WWII and Vukovar in 1991. It was close-range infantry fighting that made the difference there. A city doesn't have to have walls and trenches to defend from artillery, it has its buildings, basements and streets. Firing stuff from 1 km away simply doesn't get the job done.

Artillery is far more powerful in the field. Giving long-range units IgWall and x2 attack against cities was unrealistic crowd-pleasing. If that's what you want, go ahead, but you're not making things better.

[Wahazar]

You keep failing to acknowledge a known military history fact: ancient/medieval city walls are ineffective against modern artillery

IgWall is not increasing bombardment against city, city bonuses are as usual, it only remove City Wall bonus. Where did you get "x2 attack against cities"? Firepower is against all units, including field ones.
Correct me if I'm wrong, have bad network thus cannot check sources. And what about aircrafts?

PS. is it even possible to make long-range bombardment? Such unit always move in direction of attacking object and can only fire from 1 tile distance.

[Corbeau]

Right, I mixed CityBuster with IgWall.

But here we have a failure of effects. In modern warfare, artillery is LESS effective against cities than infantry, walls or no walls, while MORE effective against units in the field than infantry. Setting IgWall makes it the opposite, it makes artillery more effective than infantry when attacking cities.

What about aircrafts?

What do you mean long-range bombardment? Across more than one tile? As far as I know, no. I did suggest creating a "artillery shell" unit which would do roughly that, but it was a bit exotic.

[Caedo]

I believe in the future (far off into the future), there will be the possibility to perform bombardments at range; however, that's still far away.

[Wahazar]

But here we have a failure of effects. In modern warfare, artillery is LESS effective against cities than infantry, walls or no walls, while MORE effective against units in the field than infantry. Setting IgWall makes it the opposite, it makes artillery more effective than infantry when attacking cities.
.

You cannot compare effectiveness, because bombardment unit cannot fight like infantry does.
I don't understand your objections - bombardment flag is just what you want to achieve: make infantry obligatory to clear wounded enemy units and conquer city.
During my test game I used at least 5 bombardment units (usually more because some were lost) and lot of infantry and cavalry units (because AI had lot of units inside and many moves was needed to finish them before tour end).
Using IgWall bring more dynamics to the game, that is, why I proposed it for artillery/howitzers, not for early units.
And if you want to be historically accurate, Warsaw Uprising give up mainly due to air and artillery bombardments. Same for Aleppo.

[Corbeau]

You get more dynamics if you give modern infantry (Marines and Paratroopers) IgWall and maybe CityBuster.

As for those examples... Do I really have to reply? Seriously?

[Wahazar]

As for those examples... Do I really have to reply? Seriously?

I just followed your historical examples. In my personal opinion, reality should never go before playability, if we are talking about game. It is just a game.