[Game Design] Doubting the way armor works is well thought out

[Xilmi]

As far as my experiemtns show Armor reduces damage in the following way:

DmgPostArmor=DmgBeforeArmor*(1.0-min(DmgReduceCap,0.08*max(0.0,Armor-ArmorPenatration)))

I'm not quite sure about what the exact value of DmgReduceCap is, as it becomes difficult to tell when the numbers that pop up have look like this: 0.1, 0.2.
It appears to be somewhere around 0.72 and 0.9.
I just know the cap is there because units with 0 armor-pen still were damaging units with 14 armor, which would otherwise be immune.

The problem I see with this formula is that armor becomes more valueable the closer you get to the cap. Unless this is a requirement by GamesWorkshop to emulate Tabletop-Combat-Rules, I would recommend to use an armor-formula that works like in WarCraft 3, which is as follows:

DmgPostArmor=DmgBeforeArmor*(1.0-0.08)^max(0.0,Armor-ArmorPenetration)

This formula makes caps unnecessary as their's diminishing returns to the reduction but it also smoothes out armor progression. One point of armor will always add the same amount of effective-health as does the last.

Examples of the impact on damage-taken from weapons with 0 armor-penetration:

Guardsmen (4 Armor):
Before: 68%
After: 71%

Space-Marine (8 Armor):
Before: 36%
After: 51%

Captain with +3 Armor from Items (14 Armor):
Before: Remainder after cap (whatever exactly that is, 10% perhaps)
After: 31%

So the difference is really small for lower armor-values but higher-armor-values are less extreme in how they affect the survivability of a unit. The forumla can also easily be adjusted without causing many issues. For example if the DR for the 1st point would be increased from 8% to 10% to compensate that it is slightly weaker, it still wouldn't run in any problems or the need to require a hard-cap.