Light Vehicles - When and why?

[TheTalkingMeowth]

If you look in the training table, you'll see the "Inf Only" flag is Yes for every terrain training except desert. If you have them selected, they just won't be applied to vehicles.

[nuclearslurpee]

If you look in the training table, you'll see the "Inf Only" flag is Yes for every terrain training except desert. If you have them selected, they just won't be applied to vehicles.

To clarify, the "training table" looks like this (pulled from the DB):

Code: [Select]

ID  Capability Name                 Tech ID     Infantry Only   Cost Multiplier     HP Multiplier
0   None                            65830       0               1.0                 0.0
1   Boarding Combat                 65822       1               2.5                 0.0
2   High Gravity Combat             65823       0               1.5                 0.0
3   Low Gravity Combat              65824       0               1.5                 0.0
4   Extreme Temperature Combat      65825       0               1.5                 0.0
5   Mountain Warfare                65827       1               1.25                0.0
6   Jungle Warfare                  65826       1               1.25                0.0
7   Desert Warfare                  65828       0               1.25                0.0
8   Extreme Pressure Combat         65829       0               2.0                 0.0
9   Rift Valley Warfare             65898       1               1.25                0.0
10  Basic Genetic Enhancement       67771       1               1.5                 1.25
11  Improved Genetic Enhancement    67772       1               2.0                 1.6
12  Advanced Genetic Enhancement    67773       1               2.5                 2.0

As you can see there are a number of capabilities which can be applied to vehicles (and statics) including several environmental capabilities (high/low gravity, extreme temperature, extreme pressure). However, the only terrain-based capability that is not locked to infantry only is the desert warfare capability - Jungle, Mountain, and Rift Valley capabilities are unique to infantry along with the obvious boarding and HP-enhancement stuff.

[Polestar]

Light vehicles are actually not ideal for bombardment units, unless you for some reason (why?!) keep your bombardment units in the front line. While light vehicles have high evasion, this stat only matters when a unit is in front line attack position - for all other positions including front line defense the evasion stat is a nonfactor.

I did indeed expect that artillery would be in support or rear echelon formations, and spoke about support formations specifically.  Remember that support formations get 25% the chance per unit size to be targeted by an enemy front-line formation as do front line formations (ignoring exploitations). This means that, although most of the time your bombardment formations avoid getting attacked, when they are attacked they get hit at full strength. The whole of an enemy front-line formation targets the whole of your bombardment formation in such cases, and surviveability suddenly become critical.

I've put armored and unarmored infantry-type bombardment in expeditionary forces several times ... and seen them take quite a lot of casualties. Put the light and medium bombardment in light vehicles, and losses are a LOT lower. Remember that light vehicles have 3x the HPs of regular infantry, which means that at even tech crew-served AP weapons only get a kill one time every nine penetrations.

Consider an enemy armored element, of tech equal to your own, with a gun capable of killing a light vehicle in every hit and also with an AP weapon. Now compare your armored infantry (2x armour) to your light vehicles. The gun takes out one armoured infantry per hit, and the AP weapon takes out one armoured infantry per 4 hits (2x armour versus 1x pen, 1x HPs versus 1x damage). Multiply this by the 0.6 base hit chance against attacking infantry, and you get a expected kill rate of 0.6 * (1 + 6/4 effective) = 1.5. Perform the same analysis for light vehicles, and you get 0.4 hit chance * (1 + 6/(4*9)) = 0.467 expected kills. The light vehicles are, in this case, more than 3x as survivable. Against typical enemy infantry formations, they perform even better. However, against enemy medium and heavy artillery light vehicle artillery would fare less well. If you expected a lot of hostile artillery compared to your own attacking forces (a situation I've never be in yet) then I would not use light vehicles.

[shock]

As for the Inf vs Light veh:

The meat shield effect is only worth while if your trying to buy time.   If your trying to win the battle then the inf don't seem like an effective choice.   This could simply be from my lack of experience but it seems like the ai goes inf heavy armies.   If true the survivability of light veh and static should be more cost effective as a front line.   If the enemies do have a more heavy weapon focused armies, then i could see a inf heavy front line being useful.   

Inf are a bit less tonage but i couldn't really find a down size to being a large tonnage unit in combat.   Sure they are targeted easier, but that doesn't matter for your main front line unit which goal is to kinda pull fire from your more expansive units.   So its just a matter of needing a bit more transports.

The training points are interesting.   Having an extra 2x to hit over everything could be enough to make to make inf useful and its only 1. 25x cost.    It also probably a must have for invading mountain/rift and jungle combined worlds

One thing i didn't see mentioned yet is inf have a 50% nerf in their ability to cause breakthroughs compared to vehicles. 

[nuclearslurpee]

I did indeed expect that artillery would be in support or rear echelon formations, and spoke about support formations specifically.  Remember that support formations get 25% the chance per unit size to be targeted by an enemy front-line formation as do front line formations (ignoring exploitations). This means that, although most of the time your bombardment formations avoid getting attacked, when they are attacked they get hit at full strength. The whole of an enemy front-line formation targets the whole of your bombardment formation in such cases, and surviveability suddenly become critical.

I've put armored and unarmored infantry-type bombardment in expeditionary forces several times ... and seen them take quite a lot of casualties. Put the light and medium bombardment in light vehicles, and losses are a LOT lower. Remember that light vehicles have 3x the HPs of regular infantry, which means that at even tech crew-served AP weapons only get a kill one time every nine penetrations.

Consider an enemy armored element, of tech equal to your own, with a gun capable of killing a light vehicle in every hit and also with an AP weapon. Now compare your armored infantry (2x armour) to your light vehicles. The gun takes out one armoured infantry per hit, and the AP weapon takes out one armoured infantry per 4 hits (2x armour versus 1x pen, 1x HPs versus 1x damage). Multiply this by the 0.6 base hit chance against attacking infantry, and you get a expected kill rate of 0.6 * (1 + 6/4 effective) = 1.5. Perform the same analysis for light vehicles, and you get 0.4 hit chance * (1 + 6/(4*9)) = 0.467 expected kills. The light vehicles are, in this case, more than 3x as survivable. Against typical enemy infantry formations, they perform even better. However, against enemy medium and heavy artillery light vehicle artillery would fare less well. If you expected a lot of hostile artillery compared to your own attacking forces (a situation I've never be in yet) then I would not use light vehicles.

For support-echelon bombardment units, my point is that light vehicles are sub-optimal not because they compare poorly against infantry but because a heavier vehicle type is better. Medium and heavy vehicles have more armor, more HP, and more weapon hardpoints (e.g. VEH+MB/MB brings more firepower per ton than LVH+MB). Since evasion is useless in support and rear echelons the armor and HP is much better to have.

As a side note, if you're deploying armored infantry as an attacking formation you really want to give them the HP mod as well as the armor, the highest level of which gives 2x HP for infantry. This bumps up the infantry survivability which is critical for an offensive formation that needs to maximize combat power per ton, rather than per BP as a defensive formation would. Of course that's a later-game research but it does eventually change the balance. Considering the case of equal tech enemy armor, say VEH+MAV/CAP. The MAV will always overmatch both INF and LVH at equal techs, thus dealing 0.12 kills per round to INF and 0.08 kills per round to the LVH (factoring evasion and 0.2 base hit chance) while the CAP will deal damage as:

• INF with 2 armor, 1 HP: 1.5 kill rate * 0.6 evasion * 0.2 base hit chance = 0.18 kills per round
• INF with 2 armor, 2 HP: 0.375 kill rate * 0.6 evasion * 0.2 base hit chance = 0.045 kills per round

[li]LVH with 2 armor, 3 HP: 0.166... kill rate * 0.4 evasion * 0.2 base hit chance = 0.0133... kills per round
[/li][/list]
Total kills per round:

• INF with 2 armor, 1 HP: 0.12 + 0.18 = 0.3 kills per round
• INF with 2 armor, 2 HP: 0.12 + 0.045 = 0.165 kills per round
• LVH with 2 armor, 3 HP: 0.08 + 0.0133... = 0.0933... kills per round

In this case, the INF with 2 armor and 2 HP takes about 75% more casualties (in terms of individual elements), but due to tonnage efficiency you can deploy twice as many INF+CAP and come out ahead anyways. for LAV the infantry and vehicle are close to breaking even, for LB or HCAP the vehicle is more tonnage-efficient. Based on this, it makes sense early on to deploy INF+PW (or PWL, PWI) and use LVH for any heavier weapons in an offensive formation, which is a pretty standard kind of Mech Inf formation, but infantry can close the gap once you get sufficient armor and HP mods (even the 1.6x HP mod should make INF+CAP competitive).

Of course, the classic use of infantry remains Zerging a bunch of space marines with some kind of PW to soak up all the AV fire while your tanks kill things. LVH will never be able to replace infantry in this role.

As for the Inf vs Light veh:

The meat shield effect is only worth while if your trying to buy time.   If your trying to win the battle then the inf don't seem like an effective choice.   This could simply be from my lack of experience but it seems like the ai goes inf heavy armies.   If true the survivability of light veh and static should be more cost effective as a front line.   If the enemies do have a more heavy weapon focused armies, then i could see a inf heavy front line being useful.   

Mechanically, the main way the meatshield effect works is by absorbing AV fire that would otherwise OHKO an armored element. A standard MBT design might be VEH+MAV/CAP and weighs 62 tons. An enemy MAV can knock this out in one shot at equal tech. However, if you deploy half as many MBTs and escort them with an equal tonnage of INF+PW (5 tons), then about half of MAV shots in each round will be wasted on a cheap infantry unit and the overall rate at which you lose tonnage is reduced, keeping your MBTs in the fight and wasting enemy ammo.

Obviously CAP is the counter to this and enemy infantry are pretty efficient in this role as well, however both of these will waste shots against the duranium wall of your tank force which would otherwise kill off an infantry.

Essentially the net effect is that combined arms forces reduce the efficiency of enemy firepower while preserving your own. You lose some raw firepower in the initial stages of battle but your force overall has a lot more stamina to win a large battle.

Inf are a bit less tonage but i couldn't really find a down size to being a large tonnage unit in combat.   Sure they are targeted easier, but that doesn't matter for your main front line unit which goal is to kinda pull fire from your more expansive units.   So its just a matter of needing a bit more transports.

On the offensive you always want to maximize combat power per ton. It's not a matter of "needing more transports" - if you have more transports, you still want to fill those transports with the best optimized force that you can, per ton.

One thing i didn't see mentioned yet is inf have a 50% nerf in their ability to cause breakthroughs compared to vehicles. 

This is a good point.

[misanthropope]

your mileage may vary, but my experience is having many more CAP than anti-tank guns pointed in my direction.   just pulling down the most recent ground action in steve's campaign, i see CAP outnumbering all other weapons by a factor of seven.  under such circumstances light vees will soak far better than infantry. it is absolutely true that the all-CAP infantry has more firepower and no bastard ever won a war by dying for his country.  but in cases where youre trying to win an advantaged fight efficiently, the ability of vees to make breakthrough attacks and their superior morale accumulation will count in their favor.

transport costs add an overhead of maybe .06 BP per ton- steve posted a design that wasn't *exactly* about cost-containment, with a cost of .1 per ton.  That certainly isnt cheap but it doesn't dwarf the cost of producing the units themselves.  including transport costs of .08 per ton, your uber-infantry costs 2.16 and the vees you're comparing them to cost 2.88 each.  over the course of multiple battles, having to replace troops more often than transports (one hopes!) makes unit cost still more important.  GFTF time is a scarce resource as well, emphasizing unit cost efficiency yet again.

[liveware (OP)]

Lots of good discussion points here, thanks for your input everyone!

[Jorgen_CAB]

Mechanically the game works a bit oddly and you can easily game the system, but I find that to be unsatisfactory.

But...

Step 1.
When you assault bring down as much CAP weapons as you can to eliminate the enemy infantry as quickly as possible. Don't even use anti-vehicle weapons at all... they will initially mostly hit enemy infantry and the supply cost of heavy guns are atrocious.

Infantry are the best at combat enemy infantry in most situations, especially in bad terrain and in defensive line. You can back it up in easier terrain with some armoured formations using all CAP weapons.

Step 2.
Once most of the enemy infantry is gone you bring in the anti-vehicle weapons in whatever form you have it, if needed.

The issue is that all combat is completely random so you can easily know the best most optimal way to beat a specific force. But you always need to eliminate the lighter forces before the heavier the way the mechanic works and cost of supply units to what enemy a certain weapon target. Infantry is always good as they soak enemy fire for a cheap price. If you can't effectively eliminate enemy infantry you are destined to fail no matter what.

For armoured units I find medium units to the most efficient and versatile in the game. They can carry two weapons and are space efficient and can carry heavy guns to combat heavier vehicles and also decently good at light vehicles and infantry as breakthrough units.

[misanthropope]

applying the big guns with a little discretion has a bigger benefit than saving supply.  HAV (let alone bombardment) are hell on property values. 

i gotta say though, that, if you don't "game" ground combat, where exactly does the satisfaction come from?  i can understand having an OOB fetish, but my kink is maps not lists.  froggiest spends 40 hours lovingly crafting an army, i spend one, but then we both hit "resolve" and move directly on to building garrison and replacements.  not exactly "i need a cigarette" material there.

[shock]

Arn't those the reasons why games with this type of combat tend to have target weighting so heavy weapons have a higher chance to target armor, and anti-personal have a higher chance to target personal? 

[Droll]

Arn't those the reasons why games with this type of combat tend to have target weighting so heavy weapons have a higher chance to target armor, and anti-personal have a higher chance to target personal?

I honestly think this would be a good idea at least from a realism standpoint. Yes, there are situations where AT weapons will be fired on infantry but generally speaking if you have a tank and infantryman that you can fire your ATGM at you will probably choose to engage the tank with it.

[nuclearslurpee]

Arn't those the reasons why games with this type of combat tend to have target weighting so heavy weapons have a higher chance to target armor, and anti-personal have a higher chance to target personal?

I honestly think this would be a good idea at least from a realism standpoint. Yes, there are situations where AT weapons will be fired on infantry but generally speaking if you have a tank and infantryman that you can fire your ATGM at you will probably choose to engage the tank with it.

Couple of thoughts:

First, from a "realism standpoint" it's important to remember that we're talking about formations rather than units. While an individual AT gunner for instance would obviously point the gun at the tank 200 meters away than at the infantryman 20 meters to the left of the tank, at the formation level a commander often does not get to choose the target his battalion or regiment engages - the defender must shoot at the enemy in front of him, while the attacker would like to choose his target but may lack sufficient intel to do so. Considering that one "shot" represents damage dealt in an 8-hour round of combat the latter model is best to consider. Additionally, I would also note that the abstraction of the weapon types might obscure specific cases where firing at the "wrong" target would make sense, e.g. tanks in WW2 would often fire HE rounds at infantry, heavy weapons, or light vehicles out of their "Medium Anti-Vehicle" weapons.

Second, if this were implemented some kind of weighting would need to be considered as units shouldn't always shoot at the best target, otherwise many unit types would be borderline useless in combat. Such a weighting needs to be "continuum" in nature as a simple Infantry/Armor dichotomy does not exist in C# Aurora, which I suspect is a major reason Steve chose not to include it thus far. Obviously this isn't an argument against, but it does need to be a consideration.

Considering both of these, a reasonable idea might be to add a "Battlefield Intel" commander skill level, which should be fairly common and gives that commander's formation an X% chance to target the unit in the opposing formation that would give the greatest expected tonnage kill rate instead of random targeting. This would also add an interesting dimension to commander choice, enough so that this bonus should be relatively common to ensure that intel-skilled commanders are present at all levels of the battlefield (armor, infantry, artillery, etc.).

[shock]

Assuming
-aurora ground troops are meant to represent a single entity, and not a squad (which seems to be the case, although conquering planets with a few hundrad thousand tons is kinda comical)
-formations tend to be 5k tons or less to fit in drop pods (never tried loading larger formations in transports so maybe a bad assumption)

Then at a formation vs formation level the entities would all be relativity physically close to each other on the battlefield and thus in range of each other.  The local commanders should/would be trying to maneuver their bigger weapons to into positions to open fire at priority targets.  I don't think anyone thought it would be perfect targeting.  Seeing as weighting and size already exists it would just be applying a multipliers to unit sizes when picking targets for that weapon type.  Now granted displaying targeting weighting for weapons in game could be a bit of a UI mess.

The counter argument is the game already encourages mono-type formations, instead of combined arms. This is though better fortification for static/inf, better target training for inf, and better breakthroughs for vehicles.  Adding weighted targeting would be the nail in the coffin as it would make combined arm formations take more casualties.

[Borealis4x]

Arn't those the reasons why games with this type of combat tend to have target weighting so heavy weapons have a higher chance to target armor, and anti-personal have a higher chance to target personal?

I honestly think this would be a good idea at least from a realism standpoint. Yes, there are situations where AT weapons will be fired on infantry but generally speaking if you have a tank and infantryman that you can fire your ATGM at you will probably choose to engage the tank with it.

Couple of thoughts:

First, from a "realism standpoint" it's important to remember that we're talking about formations rather than units. While an individual AT gunner for instance would obviously point the gun at the tank 200 meters away than at the infantryman 20 meters to the left of the tank, at the formation level a commander often does not get to choose the target his battalion or regiment engages - the defender must shoot at the enemy in front of him, while the attacker would like to choose his target but may lack sufficient intel to do so. Considering that one "shot" represents damage dealt in an 8-hour round of combat the latter model is best to consider. Additionally, I would also note that the abstraction of the weapon types might obscure specific cases where firing at the "wrong" target would make sense, e.g. tanks in WW2 would often fire HE rounds at infantry, heavy weapons, or light vehicles out of their "Medium Anti-Vehicle" weapons.

Second, if this were implemented some kind of weighting would need to be considered as units shouldn't always shoot at the best target, otherwise many unit types would be borderline useless in combat. Such a weighting needs to be "continuum" in nature as a simple Infantry/Armor dichotomy does not exist in C# Aurora, which I suspect is a major reason Steve chose not to include it thus far. Obviously this isn't an argument against, but it does need to be a consideration.

Considering both of these, a reasonable idea might be to add a "Battlefield Intel" commander skill level, which should be fairly common and gives that commander's formation an X% chance to target the unit in the opposing formation that would give the greatest expected tonnage kill rate instead of random targeting. This would also add an interesting dimension to commander choice, enough so that this bonus should be relatively common to ensure that intel-skilled commanders are present at all levels of the battlefield (armor, infantry, artillery, etc.).

Well, one would assume that the formation has carried out some recon and tries its best to get its anti-armored units in a position to neutralize armored units. They won't always, obviously, but their target selection should have more than just a random chance to fire at the type of enemy they are meant to face. I assume that anti-tank units would actively seek out appropriate targets and would ignore infantry unless they had no choice.

A recon mechanic that increases the chances that weapons hit their proper targets is an interesting idea. Recon score could be determined by the amount of units in an army that have recon equipment or training. The more recon units, the more accurate your troops, although there are drastically diminishing returns going beyond a 1:4 tonnage of recon to non-recon troops.

[misanthropope]

this Should Be (TM) the way the system operates, but how much return is there really for the coding effort?  just wait a couple combat rounds to deploy your anti-tank weapons, like Jorgen prescribed.

if you can muscle past the objection of "with thousand ton sensors in orbit, what is a ground-based recon unit going to FIND, anyway?", reconnaissance as a general capability seems to me like an excellent idea for adding dimensionality to the ground combat system.  there are so many different mechanics that could be associated with it that there must be one or two that are both aesthetically pleasing and easy to code.