Light Vehicles - When and why?

[Garfunkel]

Always keep in mind that the system has to be flexible enough to cover both boarding combat or a handful of soldiers duking it out on an asteroid, and four-way war between different alien races on a super-terrestrial planet where each has millions of troops.

That's why it's unit, element, formation and not a rifleman, recon squad, armoured battalion.

As nuclearslurpee pointed out, a single medium antivehicle shot is actually 8 hours of combat - possibly across the entire planetary body. It's not a single DM12 MPAT round out of a Rheinmetall Rh-120 L/55 put on a Leopard 2A6 that then takes out one T-72 tank. While I love that level of detail when I'm playing Steel Panthers, it's not necessary for Aurora and thinking of combat in that way can mislead you on what this system is and should/could be. You cannot say that realistically a MAV gunner would pick a heavy vehicle as a target instead of infantry because that's a ridiculous statement to make for the level of detail that Aurora ground combat handles.

As long as targeting is completely random, it's also equal and simple to understand. There already is a weighed system based on the tonnage and that's as complicated as I would like it to be until we get a lot more detail ground combat model that enables multiple terrains, movement, and better direct-indirect fire system.

[Bremen]

There's actually a very important reason why AT weapons don't specifically target tanks: It would make mixed forces (such as tanks + infantry) considerably worse than single chassis forces.

The math is somewhat complicated, but here's a very simplified example (trust me that the math still applies in more complex situations). Let's say you have a force with 50% anti-personnel weapons and 50% anti-tank weapons.

If an enemy attacks with a force of 100% tanks, 50% of your weapons are hitting their desired target.

If an enemy attacks with a force of 100% infantry, 50% of your weapons are hitting their desired target.

If an enemy attacks with a force of 50% tanks and 50% infantry, then currently 50% of your weapons are hitting their desired targets. If weapons went after their best target, then 100% of your weapons would be hitting their desired targets, making this mixed attacker take much more damage.

If the supplies are a consideration, I'd suggest it would be better to just change the rules about supplies. Or simply eliminate them, since I kind of feel like supply units may be a level of detail too high for a game like Aurora.

[Froggiest1982]

If the supplies are a consideration, I'd suggest it would be better to just change the rules about supplies. Or simply eliminate them, since I kind of feel like supply units may be a level of detail too high for a game like Aurora.

I actually like it. Having to prepare also the logistic to keep supplying my invasion forces it's something that I'd expect from a game like Aurora.

[nuclearslurpee]

There's actually a very important reason why AT weapons don't specifically target tanks: It would make mixed forces (such as tanks + infantry) considerably worse than single chassis forces.

The math is somewhat complicated, but here's a very simplified example (trust me that the math still applies in more complex situations). Let's say you have a force with 50% anti-personnel weapons and 50% anti-tank weapons.

If an enemy attacks with a force of 100% tanks, 50% of your weapons are hitting their desired target.

If an enemy attacks with a force of 100% infantry, 50% of your weapons are hitting their desired target.

If an enemy attacks with a force of 50% tanks and 50% infantry, then currently 50% of your weapons are hitting their desired targets. If weapons went after their best target, then 100% of your weapons would be hitting their desired targets, making this mixed attacker take much more damage.

If the supplies are a consideration, I'd suggest it would be better to just change the rules about supplies. Or simply eliminate them, since I kind of feel like supply units may be a level of detail too high for a game like Aurora.

The rest of this is a very good point as well.

The last bit: the issue isn't so much that supplies or even the rules of supplies are a problem. The issue is more that as it stands right now, a MAV or HAV "shot" is so much more expensive than a round of six CAP "shots" that it's a huge waste of GSP to let AV weapons shoot at infantry, encouraging the micromanagement-intensive (and not exactly realistic) metagaming approach of sending in the INF and LVH to kill off enemy infantry and then sending in the heavy armor in a second wave to take out the enemy tanks.

Kind of the root of the problem is that CAP supply consumption can't really be balanced against MAV/HAV since it is already balanced against LAV (and LB). Removing supply entirely would be unfortunate since it adds a logistics element that is characteristic of Aurora, and reworking the system is difficult without losing the current simplicity of the mechanics.

I might suggest that the root of the problem is the ground combat AI needs to be upgraded. On one hand, using a better composition of forces that isn't as predictable and can vary more from one race or ground force to the next will help counteract metagaming. On the other hand, an AI that can recognize when it would be good to pull certain kinds of front-line units back (e.g. if the player is attacking with an anti-infantry first wave, pull back the infantry and let the armor or statics defend) would reduce the effectiveness some of these gamier tactics.

[Migi]

On the other hand, an AI that can recognize when it would be good to pull certain kinds of front-line units back (e.g. if the player is attacking with an anti-infantry first wave, pull back the infantry and let the armor or statics defend) would reduce the effectiveness some of these gamier tactics.

How do you propose to "pull back", that isn't currently an option AFIK?
If the AI can pull back some elements of their army then the player has to respond by moving their own units. Doesn't that add lots of micromanagement?

[xenoscepter]

On the other hand, an AI that can recognize when it would be good to pull certain kinds of front-line units back (e.g. if the player is attacking with an anti-infantry first wave, pull back the infantry and let the armor or statics defend) would reduce the effectiveness some of these gamier tactics.

How do you propose to "pull back", that isn't currently an option AFIK?
If the AI can pull back some elements of their army then the player has to respond by moving their own units. Doesn't that add lots of micromanagement?

 - A good question. probably set them to "Support" or "Rear" formation. There'd be some... issues, but relatively simple overall. You'd just need to adjust the how and how much the AI built of each unit type. The NPRs already are already capable of demonstrating an overarching leitmotif as is, having them generate templates and # of templates to a certain set of parameters wouldn't be too messy compared to other suggestions here. The type of units would need to change to have a greater dichotomy between "armor-centric", "infantry-centric" and "mixed" units, as well as a tweak to generate a larger number of, and variety of, dedicated defensive units so that when they "pull back" by switching the infantry to the "Support" or "Rear" they are not overwhelmed due to lack of defenders.

 - Certain NPRs might focus on large mixed unit formations, others might focus on a Zerg-Rush style infantry swarm, some might have a strict separation of armored-units and infantry formations, yet another might focus on defensive tactics centered around static "pillboxes" with Artillery. I like it, and think it would add plenty.

[nuclearslurpee]

How do you propose to "pull back", that isn't currently an option AFIK?

You can move a front-line unit to the rear echelon, or vice versa. This is how for example one would keep their MAV/HAV out of the fight until the enemy infantry have been killed by CAP and then send them into the battle to kill enemy armor. Less exploit-y would be rotating in reserve formations to replace low-morale formations during a battle, although this is more RP than anything since reserves don't really have a place in Aurora ground combat mechanically as there is no front-level granularity, everything is just one giant battle.

If the AI can pull back some elements of their army then the player has to respond by moving their own units. Doesn't that add lots of micromanagement?

I don't think it would necessarily add more micro than we already have. What I'm proposing is that if the AI can see that the player is for example running only CAP weapons or otherwise disproportionately targeting infantry, they could switch their infantry formations to the rear echelon and leave their armored formation in the front line to absorb all the CAP. This would force the player to bring their MAV/HAV weapons back to the front at which point the AI could send the infantry back in. However, if the player uses a balanced combined-arms strategy this would not be necessary. The idea is to reduce micro by making the strategy of sending CAP in a first wave and MAV/HAV in a second wave less effective.

Basically this is just giving the AI the ability to use a functionality already available to the player to mitigate gamey strategies. If the player is playing things "straight" it should have no effect.

[misanthropope]

i believe in such a case moving the infantry into the rear would hurt the defender quite substantially, and not help him in the slightest.

attacker gets those choice targets for his big guns from round 1, meaning the defending armor (which *will* be shooting at infantry targets) just disintegrates.  the attacking army will breakthrough with basically its entire front line, basically every combat round, for unreciprocated violence against the defending infantry.  unentrenched attacking infantry will receive no damage instead of the customary butchery, and the defender's "natural" rear and support will succumb very much more quickly, due to accelerated attacker tempo *and* the vastly decreased overall defending army effective size.

if you need the defender to be less vulnerable to cap, have the defender substitute statics for infantry (well, for *everything*).  but egad, i don't want to play in any universe where that happens routinely.

[nuclearslurpee]

i believe in such a case moving the infantry into the rear would hurt the defender quite substantially, and not help him in the slightest.

attacker gets those choice targets for his big guns from round 1, meaning the defending armor (which *will* be shooting at infantry targets) just disintegrates.  the attacking army will breakthrough with basically its entire front line, basically every combat round, for unreciprocated violence against the defending infantry.  unentrenched attacking infantry will receive no damage instead of the customary butchery, and the defender's "natural" rear and support will succumb very much more quickly, due to accelerated attacker tempo *and* the vastly decreased overall defending army effective size.

if you need the defender to be less vulnerable to cap, have the defender substitute statics for infantry (well, for *everything*).  but egad, i don't want to play in any universe where that happens routinely.

I may need to clarify this:

The situation I'm describing is one where the player has elected to deploy their anti-infantry weapons forward while leaving their anti-vehicle weapons in the rear, based on the logic that CAP requires far fewer supplies than MAV/HAV and thus it would be wasteful for MAV/HAV shots to be hitting the enemy infantry. The player then waits until most or all of the enemy infantry are dead before sending their anti-vehicle units forward to clean up the armor with choice-targeted shots.

My suggestion here is to give the AI the capability of pulling their infantry back to the rear area (the same capability a player can use if they want to) to counteract such a tactic, thus the player's anti-infantry weapons will be left to plink ineffectually against the AI's armored vehicles, until the player brings up their own anti-vehicle formations at which point the AI can bring up their infantry again and we're back to normal combined arms tactics where force composition rather than micro decides the battle.

The AI would not retreat their infantry if the enemy force had a concentration of anti-vehicle weapons on the front line, because as you say that would be utter suicide. All this does is gives the AI some ability to respond to the player's tactics to counteract otherwise gamey and unrealistic tactics that are only possible due to the AI's inability to react. Of course such tactics will always exist against an AI in any game, but it would be progress.

[misanthropope]

are you proposing that the AI gets to react to my deployment from a position of perfect information, _between_ the time i set my lines up and the time battle resolves?  that's some pretty heavy cheating for a relatively small advantage.  and if not, you're just walking into the meatgrinder; if the AI responds to lack of AV in round n by pulling back the infantry in round n+1, you are _guaranteeing_ that a delayed commitment of tanks will land that crushing hit.

even if you're willing to just bite the bullet and cheat like crazy, the logic has to be able to handle cases like me leaving _most_ of my big guns at the back, and if it is to produce a benefit for the defender, it has to be able to juggle a lot of different factors, because the benefit (such as it is) is highly situational. 

[nuclearslurpee]

are you proposing that the AI gets to react to my deployment from a position of perfect information, _between_ the time i set my lines up and the time battle resolves?  that's some pretty heavy cheating for a relatively small advantage.  and if not, you're just walking into the meatgrinder; if the AI responds to lack of AV in round n by pulling back the infantry in round n+1, you are _guaranteeing_ that a delayed commitment of tanks will land that crushing hit.

even if you're willing to just bite the bullet and cheat like crazy, the logic has to be able to handle cases like me leaving _most_ of my big guns at the back, and if it is to produce a benefit for the defender, it has to be able to juggle a lot of different factors, because the benefit (such as it is) is highly situational.

In theory, no cheating required. The AI can read the same battle logs we do and react accordingly, i.e. based on round N make adjustments prior to round N+1.

I will admit that the implementation is more complicated than I initially thought if you want to avoid the AI getting baited into even more exploitative tactics. Mainly the player needs to not be able to predict what will happen with absolute certainty, else they can anticipate the AI's maneuvers. That shouldn't be impossible though, after all the naval combat AI has similar ability to learn the player's tactics and adapt to some extent, not as much as a human would but still reasonably so. For example, if the AI learns that the player likes to make a delayed commitment of anti-armor forces they can try to anticipate that and retreat their own armor to get the jump on the player rather than being purely reactive.

[Borealis4x]

Can't we just bias units armed with anti-x weapons to try and target x? It can happen seamlessly in the backround and would result in better battle outcomes with AT actually doing its role even when there are infantry in the field.

[nuclearslurpee]

Can we? Yes.

Should we, and if so how much? This has been discussed for half of the thread now...

[Droll]

Consensus should be reached by competitive aurora tournament. Winner takes all and Steve refs.

You can work out the rules amongst yourselves I'm sure .

[TheTalkingMeowth]

You can work out the rules amongst yourselves I'm sure .

Lock 3 aurora players in a room for an hour and you'll have 4 opinions about the game by the end. Wait longer to acquire additional opinions.