Non-Deterministic Combat

Ever tried to fight using rhino's in FA on a map that is hilly like sentry point?
They shoot into the ground most of the time, thats just the same.

I am not sure you are using the term non-deterministic correct. AT least to my understanding non-deterministic basically means random and that's not what you are asking for.

Keep things entirely skill based. Luck can just be the unintended effects of poor players getting it right.

I think the biggest reason to why fights in SupCom felt so determined from the start was because you couldn't really micro large unit-groups.
Orders were always delayed with half a second and when you ordered large groups of land units to move, only 3-6 of them would get orders incrementally with a delay.
Also if you tried to use max range with your units they would just play bumpercars at the edge and once the target got out of range they would have to accelerate and be out of range before they could catch up again.

If you look at SupCom there is still alot of random elements that are out of the players control or is very hard to control.
-Overkill. Shots will be wasted because some of the projectiles are in the air when the target dies.
-Missing the target. This can be caused by units bumping into each other, accelerating and decelerating. In large groups it is hard to perform dodging maneuvers as units get their orders at different times and if you spam commands no orders will reach the ones at the end until you stop spamming orders.
-Target priorities. Units tend to follow their target priorities. So if you want to destroy the enemy commander in FA you might have to destroy all other units near it before you can get the commander. If you target the enemy commander when out of range you can't maximize your firepower because units won't be in range of the enemy commander all the time but will stop at the edge and very close to eachother, susceptible to Overcharge and splash damage.
-Movement queue. It is somewhat random which land units get the move commands first. Maybe you could learn it but I don't know.
-Pathing. Learning how the pathing works is important like knowing how to block the pathing of the enemy commander. However if you wanted to execute it perfectly you would basically have to know how the pathing works, how often the pathing is checked for a moving unit and in which direction a unit will go if it is obstructed by an obstacle.

Anyway. I want interesting weapon mechanics and most of the suggestions I have seen here sounds good.

^exactly. Also, I don't think that bigger armies beating smaller armies all te time isn't really a problem that needs to be fixed... The only time a smaller army should beat a larger one is when (like orange knight said) the controller of the smaller army has superior tactics, better unit composition and smarter micro.

There are two issues being improperly fused here.

The first is the escalating efficiency of increasingly large armies, which is a problem. A very large army will sustain fewer casualties fighting an enemy force than a smaller army, resulting in highly incentivized deathballs.

Larger armies should defeat smaller armies, but they should do so less efficiently than smaller armies which still win. Massive overkill should be severely punished, not rewarded.


The second issue is the deterministic nature of combat. Doing stupid things like having a random miss chance or having critical hits is not going to change a thing. Battles will still be equally bland, deterministic affairs based on type and quantity. It just changes a unit's dps, it doesn't introduce more depth into the low-level combat.

Instead, how about having more interesting and complex weapon interactions. For example, inaccurate direct fire weapons that perform better from close range. Burst and reload instead of constant fire. Linear-firing rockets, shotguns that fire in cones, shaped splash damage, be imaginative.

These things don't have random effects, but they do make any particular fight between two mixes less obvious to the player, and makes a particular situation's positioning, terrain, and other factors more important.

The positioning and care to your troops should be represented. Like if you have the intelligence to position yourself in an ambush. Where you have a larger platform to fire from than your enemy. Like a line infront of a column.

Even if your units are more spread out, you should get more kills, because units that are near eachother will get hit by bullets intended to their neighbour. This is basic Infantry tactics 101.

I mean, if we allow basic tactics such as Flanking, Contact surface and Range to determind the outcome of the battle. Not necessairly the players ability to cock out units at a higher rate since he is such a good economy player.

I mean. This should not be a clean computer simulation, it should have actual logical choices to determind battles. Rather than just a mathematical algorithm that depicts if I will win or lose.

I mean if all I had to do, was build units and send them to my enemy, I could play a game on my browser right now. If all PA has is good looks and big asteroids, what power does the player have other than just making lots and lots of units. I want to strategically place my units out, so they can respond to enemy attacks or ongoing skirmishes. And I want to make plans on how to crush or penetrate parts of my enemies defenses.

This will make the game alot more exciting, and gives way more depth and enjoyment to the game.

On of the things TA did very well, that a lot of games just ignore, is turning radius and tracking speed. This can play a large part in non deterministic combat. Many RTSs (Starcraft 2 in particular) don't have good flanking mechanics because instant turning, instant tracking, instant hits pretty much make them a moot point. in SC2 the term surround is often used for the zerg units rather than flanking. In TA flanking was a real thing, units took time to track and turn. In SC2 it's just about getting your units close enough to hit before they get killed.

The ball-o-death had distinct disadvantages in TA. Yes you could swarm, but proper group placement was better. Cover fire was useful, spotting for artillery was a skill.

Play style in SupCom seemed to follow a lot more of that ball-o-death style (not always). I think a lot of it is the laser style weapons. Weapons in TA seemed to have more travel time than SupCom weapons, so things like mobility and ground cover helped a lot.

Another big plus for TA was non hard counters. The initial Anti-Aircraft in TA was better against air than ground units but those missiles still worked against ground units. Many grounds Units could at least try to target air units as well, so a single gunship didn't decimate an entire army because you forgot to make a couple AA units. At the same time many gunships will wreck havoc on an army with no AA. StarCraft and many similar games suffer from the "oh look I have a single flying unit, watch as you army is destroyed because you AA tech is hard to get".

Similarly there are the air and anti-air units that apparently can't point their weapons at the ground ever. Terrans anti air missile turrets that couldn't shoot ground targets always perplexed me. I've seen some of this in SupCom (and yes TA too, but to a lesser extent), where anti-air units just can't even shoot grounds units, and ground units can't even try to hit air units, making harassment with single or low numbers of units disproportionally effective.

I'm not sure if this is what Ledarsi is trying to convey, but I think it plays a large role in what he is talking about. Better strategy, rather than just bigger armies or hard counters.

Actually, no they shouldn't be punished. That's the wrong way to approach the problem entirely. By using a large army against a small army, you are already 'less efficient' simply because you're using more mass than is required to get the job done. And apart from overkill being dependent on weapon damage vs unit health, overkill should not be a defining factor for a large army. Units should be smart enough to spread out their shots and destroy the enemy as effectively as possible.

There are much better ways to combat the deathball.

These ideas sound interesting, but keep in mind that it's very easy to slip past interesting into irritating, which is what happens when you don't have *enough* control over your units. Additionally, the more types of combat / weaponry you introduce, the smarter you're going to need your unit AI, because if your units can't perform half decently without a helping hand, you're going to massively increase the amount of micro required to effectively fight a battle.

So for example, if you give an attack order to a shotgun toting mech marine, it should be smart enough to zig zag its way past the dumb firing rocket attack of its enemy, and the enemy rocket bots should be smart enough to back up & spread out when they see him coming.

In my above comments I was referring to the number of units in a locale as "overkill," not that too many shots are being used to destroy a single target. Deploying 300 tanks to take out a small, lightly defended outpost is overkill. A much smaller force would do the same job much more efficiently. This is how you punish excessive force concentration and thus discourage deathballs.

Skilled play should involve matching your assets to where they are needed as accurately as possible. Sending too few means you lose one battle or patch of dirt. Sending too many may weaken you elsewhere, or result in more casualties than were necessary to accomplish that group's objective.

I do agree that the AI needs to be capable enough that micromanagement of individual units is not mandatory to be effective. However the automatic combat AI doesn't have to be very clever for this to be the case.

I am seriously not understanding this overkill thing. The only reason I would deploy 300 tanks to kill a small outpost would be to kill it very quickly. And how would different weapon mechanics punish someone for overkill??

There are already disadvantages to "death blobs", no need to add some artificial disadvantages too. In SupCom/FA it was very easy to win against people with the deathblob mentality, because of the size of the maps. Basically, he cant be everywhere at once with his blob, while i can take down his attempts of expanding with more spread-out units.

SupCom2 however ruined everything, as many maps are basically made out of chokepoints, and too small at the same time.

If the scale of the game is big enough, having multiple groups of units is a natural advantage in itself.

Technicly shouldn't have chokepoints also counted against death blobs?

Single fire, high yield, high damage heavy weapons (splash optional). It's complete overkill on a single little target.

No. Chokepoints count against short ranged units, with limited movement options, on the attack. Any Seton's Clutch player knows that a chokepoint is retarded if you get a pile of high range, high damage units to cover it.

You mean the defender? Becuse then you can force the enemy ino a position where they are eaiser to deal with and their numbers mean less.

Unless you mean the attacker, in whtich case that's just good staratigy to cover your forces. Becuse if you don't attacking will be hard.

No no no **** no any RNG. Units need to be perfectly deterministic. But what game needs is a lot of micromanagement potential so players and their actions can make the difference in the fights, thus creating non-deterministic fights where everything depends about players and their skills.

AoE attacks, dodging them, "spells" that disrupts movement/attacking of units and so on.

ogwi, have you played Total Annihilation, Supreme Commander, Zero K, or any TA-style game? Micro is definitely a very low priority item in any of those games. Instead the games are more reliant on higher level strategy. As I said in response to a previous post of yours, don't come into this expecting another Starcraft.

Myeah. Micro is a bad idea. Most of the battles in this game are fought with the player only popping in twice a minute to see if their troops are still more or less there.

:lol: You make it sound so unattractive.

No, that's exactly what the game doesn't need. Nobody wants intense micro when there is such a large playing field. If you want to have all that play Starcraft 2 where the playing field is really small.

It is a problem with these games, but one I don't know how to solve. One of the reasons I love spectating huge Zero-K games, is because that's the only way you'll actually have time to enjoy the mayhem and destruction.

If you're actually playing you're constantly balancing watching things explode with trying to win the game, which tends to involve a lot of base building.

Although I certainly hope the multi-windowing will help in this regard. I'd love to be able to set one of my windows to "show the most exciting battle" or something in that trend, so that it will automatically focus on battles that are happening. Both because it's cool to just watch them happen, and because it's important to know when they're going down.