8/8 Infantry - The cheapest Soft-Attack Division

  • We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Even shitty little minors can afford to outfit at least 1 or 2 tank divisions. Having some armour on the go really isn't that expensive so long as you adjust the templates and the amount of them trained accordingly to the industry you have.
 
  • 2Like
Reactions:
Even shitty little minors can afford to outfit at least 1 or 2 tank divisions. Having some armour on the go really isn't that expensive so long as you adjust the templates and the amount of them trained accordingly to the industry you have.

The funny thing is that one of the best options is heavy tanks for breakthrough. You don't need as many of them to maintain armor higher than the typical enemy piercing ability. Saves a lot of fuel too.
 
8/8 with MOT and LSPART is one of my favorite early game divisions for soft attack and encirclements while still having decent breakthrough and hardness but they fall off hard once infantry tech starts being good enough to pierce the small armor given by light SPG 2.
 
Here the Text from the Wiki:

Higher armor than the enemy's piercing will also increase the number of unit attacks in combat and change the organization damage dice roll per attack from 1d4 to 1d6 as the unit can move around the battlefield more freely without getting pinned or damaged.
From: https://hoi4.paradoxwikis.com/Land_units#Combat_stats

I think dicerolls are defined in Hoi4 as:
1d2 = {1,2} equally distributed
1d4 = {1,2,3,4} equally distributed
1d6 = {1,2,3,4,5,6} equally distributed

Then the average of 1d4 is 2.5
The average of 1d6 is 3.5
3.5 is 40% more than 2.5

That means having more Armor than the other has Piercing, would lead to +40% damage.
 
  • 3
  • 2
Reactions:
I summarized the results in a table. Also added it to first posting.


Table of Soft Attack per 1000 IC of all Templates discussed in this thread:

TemplateSoft AttackIC CostSoft Attack per 1000 IC
8/8 1936 no doctrine266,01575169
8/8 1939 SFP 5R392,71789220
8/8 1939 SFP 5L397,21789222
14/4 1939 SFP 5R322,81561207
7/2 1939 SFP 5R184,3867213
4/4 1939 SFP 5R219,3981224
8 Mot/8 Mot Art 1939 SFP 5R392,73789104
8 Mot/8 Mot Rocket Art 1939 SFP 5R416,73857108
8/8 Rocket Art 1943 SFP RR676,62232303
8/8 Rocket Art 1946 SFP RR727,92232326
8 Mot/8 Mot Rocket Art 1946 SFP RR789,54012197
8 Medium 6 Mot 4 MSPG 1939 MWF 5R402,2779952
 
Last edited:
  • 4
Reactions:
Well, it must not work that way because there's this in the game

123.png
 
  • 2
Reactions:
Well, it must not work that way because there's this in the game

View attachment 580926

Intersting...
Here in the wiki: https://hoi4.paradoxwikis.com/Land_battle#Damage_dealing

It says:
When armored units are in combat against targets with insufficient piercing, the organization dice size is increased to 6, representing the ability of the armored unit to move more freely under fire, obtain better positioning and thus deal more damage.
And:
For 25 organization dice rolls, (1 + 6) / 2 = 3.5 per roll is expected.

"(1 + 6) / 2 = 3.5" is a {1,2,3,4,5,6} equally distributed dice.

So either the wiki or this tooltip is not 100% correct.
 
  • 1
Reactions:
I think dicerolls are defined in Hoi4 as:
1d2 = {1,2} equally distributed
1d4 = {1,2,3,4} equally distributed
1d6 = {1,2,3,4,5,6} equally distributed

Then the average of 1d4 is 2.5
The average of 1d6 is 3.5
3.5 is 40% more than 2.5

That means having more Armor than the other has Piercing, would lead to +40% damage.

In the stochastics the probality calculation always includes the "not-happened-case" ( didn't found the right term for it )

In "our" example 1d2 means 0,1,2 !!!!

1d4 = 0,1,2,3,4 -> average = 2.0
1d6 = 0,1,2,3,4,5,6 -> average = 3.0

The result 50% more.

In don't know what has been developed, but this could perhaps be an explanation cause it meets what @Kryndude stated above.
 
  • 1
Reactions:
In "our" example 1d2 means 0,1,2 !!!!

1d4 = 0,1,2,3,4 -> average = 2.0
1d6 = 0,1,2,3,4,5,6 -> average = 3.0

That cannot be correct.
In what world exists a dice2 with 3 possibilities, a dice4 with 5 possibilities and a dice6 with 7 possibilities?
If it includes 0 it should be 0,1,2,3,4,5 -> average 2.5 but wiki says 3.5 and a dice has 1 to 6 not 0 to 5.



In the stochastics the probality calculation always includes the "not-happened-case" ( didn't found the right term for it )

Also in stochastics rolling a d6 dice has 6 outcomes.
Ok, someone could include some fantsy 7th outcome: "dice breaks during rolling" ;)
 
That cannot be correct.
In what world exists a dice2 with 3 possibilities, a dice4 with 5 possibilities and a dice6 with 7 possibilities?
If it includes 0 it should be 0,1,2,3,4,5 -> average 2.5 but wiki says 3.5 and a dice has 1 to 6 not 0 to 5.
...

In common sense you are absolutly right. When we read d6, we are referring to reality and imagine a dice with 6 sides = 6 outcomes. That meets our "experience". Perhaps we never saw d7, but we conclude = 7 outcomes.

My point is, that we both do not know, what a d6 is for the game-engine. For software, a "dice" is a placeholder for possibilities. We don't know the code used for calculating possibilities in this case.
So I stated, that if the code includes the possibility of "nothing-happened", the outcome-values are equal to the tooltips.

My intention was to show, that there are more ways to calculate, than our own logic and experience will "allow" us to think of.
 
So I stated, that if the code includes the possibility of "nothing-happened", the outcome-values are equal to the tooltips.

Thanks for the clarifying post! You had me googling after the first one, lol.

After the second post I can see you explained yourself in the first post, but for some reason it did not click with me like your second one. Weird how a rephrase can help so much.
 
  • 1Like
Reactions:
Not the first time a tooltip is wrong.
I double checked and the wiki is correct; the tooltip is wrong. The tooltip calculates LAND_COMBAT_ORG_ARMOR_ON_SOFT_DICE_SIZE / LAND_COMBAT_ORG_DICE_SIZE - 1 = 50%.
The code doing the actual damage calculation does 1 + <random number> mod <die size>, resulting in outcomes as @Simon_9732495 explained (2.5 and 3.5 average respectively).
 
  • 5
  • 2Like
Reactions: