Armor Mini-Mod

Here’s a crude mini-mod demoing Lua-based armor:

Only Russia, and only 4 units ATM. Costs set by formula. I cut out pretty much everything not having to do directly with armor: logistics, flags, etc. are all gone. Just spam tanks from your HQ. CRAIG works with it though.

More information to come after I get past this current spate of papers/midterm.

Stat conversion:

HP = weight in kg
maxVelocity = speed in kph / 13.5 (this makes 8 elmos = 1 m)
turnRate = road speed in m/s / turn radius in m * 65536 / (2 * pi) / 2 / 30
= 48.29 * road speed in kph / turn radius in m (maybe use twice as much and rough speed instead?)
acceleration = maxVelocity / 60
brakeRate = maxVelocity / 15
range = range in m (1 elmo = 1 m)
weaponVelocity = speed in m/s * 2 (compromise between weapon scale and model scale)
damage = shell weight in g
ap_penetration = penetration in mm at point-blank
ap_dropoff = dropoff in penetration per 1000m (approximate as linear)
armor = armor thickness in mm (haven’t accounted for sloping yet)

Same maxSlope for all tanks, but some are slowed down more by slopes? i.e., slopeMod = k * (road speed / rough speed - 1), where k is between about 6 and 30, probably toward the latter.

Well, if you want to get indepth on turn rates/slopes, let’s start :slight_smile:
Some tanks were remarkably different in that regard.

  • T-34 had to slow down to turn (else it risked losing tracks) = large turn radius. It could not turn in place by reversing one track relative to the other - transmission design did not allow for that (most other medium tanks COULD do that).
  • T-34 tended to slip on steep slopes - tracks had suboptimal ability to ‘grab’ ground (compared to German tanks for ex.).
  • Soviet heavy tanks (KV, IS) and all vehicles based on them had much better transmission design and could turn w/o slowdown and in place. Soviet study of captured German (Tiger) and lend-leased Allied (Churchill) heavy tanks revealed those could perform the same.
  • German tanks (and also Shermans), having considerably narrower tracks than T-34, performed much better on slopes (they did perform worse in snow/mud/soft ground, which is hard to represent in spring).
    In general, slope climbing ability of tanks can be determined from their ‘ground pressure’ value (the pressure tracks create on the ground). That can be found rather easily on the net, since it’s one of the basic values used to describe tank’s passability. Other such values include maximum allowed slope (but not the speed on that slope), water depth that can be forded with/without deep fording equipment (if the tank can use that), trench width that can be crossed (with tail if a tank has one, but none of WW2 tanks had it afaik), vertical wall height that can be crossed (usually no more than a meter), diameter of tree trunk that can be crushed (to determine which kind of forest is passable to this tank).

Hmm, interesting. So does higher ground pressure = faster on slopes (do they grip the ground better)? Also, do you know how the turning radius values were calculated? Did most tanks have to slow down to turn?

Also, come to think of it, our tanks don’t have any crushStrength ATM…

Next steps: Brits, Germans, and US, probably in that order. Accounting for slope, perhaps better pricing equation (still rather rough).

In theory a tracked vehicle can turn in place around its center (0 radius). However that requires the transmission to be able to run tracks in reversed direction (one forward, one reverse). Some tanks (T-34) could not do that, they could only stop one track and run the other. So they turned around the stopped track, with turn radius being equal to tank width.
If the tank is moving, then the turn radius can only be determined experimentally. T-34 for example had to slow down to 5-8 kmph in order to turn, so the radius was fairly constant regardless of speed (attempt to turn at higher speed leads to losing one of the tracks and nearly an hour of hard work putting it back). That’s the consequence of T-34 being a maximally simplified design adapted for mass-production.
Heavy tanks were not so simplified, they enjoyed more complex drive systems that allowed to perform the mentioned track reversal. German equipment was traditionally more complex than Soviet and so also allowed for that.
Higher ground pressure does give better passability in mountainous terrain. If the ground is hard, high pressure provides better grip and leads to higher max slope. It fails on soft ground however, Soviet crews were complaining about lend-leased Matildas mk.II and their poor ability to cross mud or plowed fields for example (where T-34 had little problems). Germans had the same problem with their tanks getting stuck in mud. Sadly our maps don’t have terrain variations of that kind (it would require more movement types than K/V/S/H present in Spring map format, bad terrain would then slow down some vehicles but not others).

As you can see, the tank has only a bit higher ground pressure than a walking man, and considerably lower one than an armored car.

That’s the tank climbing on a vertical wall. As you can see, some designs are better for those than others.

That’s the crossing of a trench. As you can see, the trench can only be a little bit wider than half tank’s length, any more and the vehicle will ‘dive’ into it.

That’s the effect of turning radius. First vehicle has large turning radius, it has to stop-go back-stop-go forward to make the turn. Last veh (T-34) has low turn radius (it slows down to turn) and passes the corner easily.

(pics taken from this Russian article)

(now I start to think giving engineers an ability to dig anti-tank trenches might be a good ides… :slight_smile: )

Very interesting, although also very complicated. It will be a challenge to figure out what can be represented and how to do it.

Brits are now in; turn rates based on max velocity for now. Still need to account for armor sloping and find a better equation for turnRate, slopeMod.

Also there’s some problems with the counter system, originally I had intended Medium Tank > Armored Cars/Light Tank > Tank Destroyer > Heavy Tank > Medium Tank, but tank destroyers tend to be more heavily armored than I expected, and there’s all sorts of weirdness with armored cars/light tanks (e.g., Daimler’s gun actually has better penetration than Cromwell’s gun). Maybe it will work better in the main game where there are more units.

What? Looking at the info, that’s only true at extreme ranges (1500m+) or against face-hardened armour (used only on early Panthers and on the 30mm applique plate on Panzer IV, not exactly a major concern)…

Did you look at the spreedsheets i put up btw? The current weapons are designed to actually match the real penetration curves, via the dynamic damage range tag (though iirc they do all use the linear mode so your method with custom params can be made to match up the same way)

You seem to have done away with the seperate HE weapons in your test mod… is that intended to be permanent?

Huh, you’re right. I don’t know why I thought that… I must be going crazy…

That’s the general idea.

No; the mod is basically a bare-bones AP-versus-armor test/demonstration. HE, logistics, flags, etc. have all been removed, but obviously I’m not saying we should take those out of the main game.

Armor sloping is now taken into account, although now GBR CRAIG seems to beat RUS CRAIG most of the time… maybe I should think harder about the pricing. Or maybe it’s just that T-70s penetration is too weaksauce compared to Daimler?

Edit: Looks like cost miscalculation relative to formula, will fix and see if it helps.

The mini-mod is done for the time being. All sides are in with four vehicles each. Give it a try if you’re interested.

I’m starting to think damage = shell weight in g might not be the best idea. Maybe add some base amount of damage to all weapons? Perhaps when I start working on this again.

Perhaps take explosive content into account?

Though british forces regarded HE fillers in AP rounds as pretty pointless.

Anyway, I saw this

in the Git log. Seperate hull/turret damage plox?

FLOZi: Aieee…hoping to maintain some minor semblance of comprehensibility…detailed modeling is all well and good, but if nobody can ever understand why a round did as much damage as it did, we’d be better off sticking with what we have now :stuck_out_tongue:

I knew you’d say that. :cry:

Are you sure that system allows to find out which piece was hit? I though it’s just a more-or-less just a collision volume system.

Not at all sure

It is now: