Complex Units & Combat

There is a trend in modern RTS games to simplify the building blocks of the game. Each unit having one function, one weapon which behaves in a simple way, etc., for the sake of having simple basic elements. For example, the main combat ground units in SupCom 2 had a single weapon that deals a constant dps to one target within range. This does simplify the game. However, this makes small battles really uninteresting. It also makes the use of these units straightforward, and limits the ways in which they can be leveraged by intelligent players to gain an advantage on the large scale. Simple units lead to simple mechanics and interactions, for example in SupCom 2, main combat units have a hard line at their range limit where they do nothing beyond that distance, and within that range they are all equally effective. Maneuver with these units is trivial- make all your guys fire, minimize how many of them are firing.

A better philosophy for the structural complexity of the game is to have complex building blocks, like the difference between Play-Doh and Legos. More complex atomic pieces may introduce slightly more surface complexity, however they create exponentially greater depth as the number of pieces grows.

Weapon Complexity

So what would be an example of a more complex unit? One approach is to simply give a unit multiple different weapons, even if it is a simple unit. For example, suppose we gave a normal assault bot, in addition to a normal weapon, the ability to do high damage at point-blank range. Even if this unit is engaged with its normal weapon, there is still a decision- should this unit close to range to use its other weapon? Enemies fighting this unit, but without close range attacks of their own, will want to stay at standoff range, which may create a dynamic where one player wants to surround or charge in, whereas the other wants to skirmish on the move. If both sides are simple chumps with guns then as long as they are firing, they can drive anywhere and it doesn't affect the battle, made especially silly if their movement speeds are similar as well.

Another possible approach is to use more complex rules regarding a unit's combat effectiveness, such as ammo for specific types of weapons. An air superiority fighter carrying a limited number of powerful missiles in addition to a gun weapon is significantly more interesting than the same fighter which can fire unlimited missiles, as when it uses its big missiles its effectiveness drops. This dynamic can be leveraged to make smaller and cheaper units competitive, as powerful weapons are wasted on cheap targets- this missile-carrying fighter wants to engage valuable targets, and would actually be relatively ineffective against a group of cheap units.

Movement Complexity

More complex movement rules can also differentiate units. Allowing walkers to traverse rougher terrain than treads, which can traverse rougher terrain than wheels, having some units with amphibious capability, etc. are all examples of movement differentiation. This is something most players are familiar with, so I won't dwell on it long.

Still, complex movement has a lot more potential to be interesting on large, procedurally generated maps than it ever did in TA, due to the possibility of painting large areas of a planet's surface with terrain, rather than limiting it to the steepness of the surface. A huge desert might make wheeled vehicles very slow in a large region of the planet, for example. It should be made clear to the player what is happening, however, as hiding important information like sand+wheels=bad will lead to frustration.

Conclusion

There are lots of other ways to make units more complex. The basic idea is that interesting low-level interaction greatly increases the level of differentiation massively as the scale increases. There is a tendency for large scale games to default to very simple low-level interaction, relying on scale to multiply the complexity of the board's state. Hopefully PA tends towards making small battles intricate, and have the scale let those interesting effects have sweeping effects on a grand scale.

Completely agree with this idea. If I remember correctly some larger units and experimentals in Supcom had multiple ranged weapons. Implementing multiple weapon ranges/modes as early as tier 1 sounds really fun.

My dream addition to this idea would be close combat support—complete with hand-to-hand animations. Imagine robots ripping each other apart limb by limb. (Uber, if you're reading—please find it in your heart to try this out)

Not just ranges- this could also be differentiated by any other feature of a weapon. For example, a machine gun in conjunction with a line-fire rocket with a long cooldown. Or a vulcan cannon that can target ground and air in conjunction with a guided surface to air missile.

Could even have multiple weapons with different fire arcs, such as cannons on the left and right side that can both aim forward, but not to the rear of the unit. Other features like minimum ranges, fly-by-wire, limited turret turn radius or turret turn velocity, or a variety of other ways to make units and combat more intricate. Combat should be interesting even if only a few units are fighting.

This idea sounds an awful lot like "players should have to micro all their units all the time to get decent effectiveness". Having to dance your units around continuously because multiple weapons mean the range you want them to engage at is not the range the AI wants to engage at is bad. Side-only fire arcs are even worse, since it means you manually have to tell units which way to face and have to tell them to stop before they'll fight effectively. And mêlée attacks? Seriously? Not only are decent mêlée animations expensive and wasted on a unit that doesn't primarily attack in mêlée, the idea that units get as close to each other as they do is already a conceit, so nuclear death robots wanting to move up and hit each other with swords is just lame. Perhaps Dawn of War is more the game for you? Spammable units should be simple; complexity should be left to expensive units.

Bloody Americans...

Melee attacks are probably not worthwhile, especially melee animations. However having assault units deal huge damage in close quarters makes a lot of sense, for example the Zeus in TA. The assault unit example above is that we could have an assault unit that has a weaker weapon with more range in addition to its assault weapon.

Having a more complex unit like this example does not necessarily create more micro. This unit can function as a main combat unit, and can also be ordered to assault a position. This example assault unit will automatically use whichever weapons are in range (meaning both of them from close quarters).

Same goes for various other features that might make small fights more intricate. Even features as detailed as turret rotation rate and firing angles affect a unit's performance in complex ways, but do not force micromanagement, as they are easy for a unit to control on its own. A good example of this type of complexity would be things like Kbots raising their weapons, creating a small delay before they fire, or tanks in TA turning their turrets at a limited rate, and firing with different accuracy while moving. Wreckage blocking projectiles was a huge contributing factor to combat complexity in TA, and is probably not feasible to implement in PA on a large scale, so we have to get that sort of combat complexity and exploitable unpredictability from somewhere else.

Very complex units might become more effective with micro, but that's not really what I am suggesting. I agree that we don't want to require very low-level micro to be effective. However we also do not want SupCom 2 style blobs just shooting at one another until one of them is dead, with nothing interesting happening.

With so many units in battles it will be complicate to micro units so they use the correct weapon and stays at the correct range every time.

Also nothing about ammo and refuel, if you want bigger abilities or attacks being limited make them with cooldowns or use of energy.

But with the movement from difference terrains and type of movement (wheels, tracks, legs, amphibious, water, etc) its a must. Make it not so complicated but enough to affect the way you thinks attacks and defences.

It necessarily creates more micro. It is inherent in the paradigm. If you have a unit with two weapons, one of which outranges the many enemies, and one of which does not, the AI will ever use it optimally. Some times you'll want it to engage at maximum range in order to take no damage from return fire, and others you'll want to close to close range in order to maximise your DPS. The AI cannot make this decision for you.

I think by now, "TA good, SC2 crap" goes without saying. It'd be like reminding the devs to make sure their game features colour graphics.

There is perhaps a perspective of scale to be had here that with a large scale war, you simply don't care that your units aren't fighting perfectly optimally in certain circumstances. As the scale gets bigger, the significance of each individual drops. You don't actually need to eke every last erg out of every single unit when you have hundreds of them. If you are making good strategic moves, battles should not be decided by such a small margin that tiny maneuvers by individual units will affect the outcome (and even then, there would likely be few survivors on the winning side if it is that close).

Slow rockets that fire in a straight line are a prime example of a weapon which benefits immensely from manual control. According to you, the inclusion of such a weapon should force the player to micromanage their units.

However the scale of the game solves the micromanagement problem for you. If I have 100 units with this weapon, firing at 100 enemies, I simply don't care. It is not worthwhile for me to micromanage this unit, even if theoretically I could increase its effectiveness by doing so. I have more important things to do (unless I don't, because I really, really need these units to win this fight, and they are badly outgunned. Then I will micro them, and I can increase their effectiveness if I really must). Generally speaking, the scale of the game means the player will always have actions and decisions that can actually decide the war.

However, this rocket weapon behaves very differently in a fundamental way than a normalized direct dps weapon, as no longer is the outcome of the battle as predictable based on numbers, and no longer does my dps function precisely map to the number of my units and the number of enemy units, resulting in more or less perfectly predictable battles. Furthermore, I can leverage this unit's strengths by falling back with it, forming the group into a battle line, or doing various other things with this unit that don't make sense to do with a basic blob of HP that deals constant reliable DPS.

Not sure why units need 2 weapons. Just have a complex army. I would like to see T1 units have very distinct weapons and roles, like TA (and more so ZK).

SupCom2 is the worst example of dumbing down T1 units. I've been playing that recently and T1 battles are just boring (plus very little diversity between factions at T1).

To be honest, I know it might not happen, but I really hope the developers will just look at ZK as inspiration for the way T1 combat can be made very dynamic and exciting without excessive micro (due to the unit AI doing most of the work). ZK is also a great example of very wide unit diversity and inter-game diversity within a single faction.

The scout/raider/skirmisher/riot/assault/artillery/support/kamikaze ground forces role distinction paradigm primarily defines dynamics within ZK gameplay and a similar role separation/interplay would help PA.

I don't know what you're talking about leddy, since even SupCom2, crappy as it was, featured microing to dodge artillery, which is exactly what you describe. And isn't something I complained about. I'm calling you out on wanting units with stupid weapon loadouts for which you force players to compensate.

By the way, you say micro becomes less significant in large groups, but then you go on to give reasons which don't actually support micro becoming less significant in large groups. If dodging one rocket with one tank is worth it, then dodging 100 rockets with 100 tanks is equally worth it. If moving 1 tank into optimal range gives a 50 point increase in DPS, then moving 100 tanks into optimal range also gives a 5,000 point increase in DPS, which is just as significant. That kind of micro does not become obsolete in the late game because you stop caring, but because it stops working. If you try to micro 100 tanks to dodge arty, it just doesn't work. Maybe some units at the edge will dodge, but all it really means is the shells will fall on a different part of the formation.

Well yes, it doesn't work as well. To add another example, the line fire rocket is dodgeable when there's only a few. This is basically the Rocko's weapon in ZK, and you CAN dodge it with fast enough units.

But when you are facing 20 rockos, you're not doding one rocket. It's a whole wall of rockets.


Different DPS at different ranges does not add that much to micro however. If you think you need to use the high DPS short range weapon, you select the whole formation and give ONE order to close to optimal range. If you want to kite, you tell it to kite.
Yes, this requires unit AI, but at the scale PA operates at, you'll Need unit AI anyway.

I suppose the goal from a game-design perspective is for some (large) proportion of battles to be interesting. Complexity in individual units is one approach; through army composition is another approach. The effectiveness of each approach depends on the distribution of the size of battles. I believe for PA they're trying to push this distribution towards 'big', which would leave small skirmishes rare -- and perhaps interesting in and of itself for that reason.

Despite the focus on larger battles, I do not fear the developers will overly simplify the individual units: looking at TA as its inspiration, complexity through weapons and movement already exists and hopefully PA won't need to drop this theme. In TA, there was a variety of weapon and movement characteristics; and the larger, costlier units often had multiple weapons. Even two units with the same kind of weapon or movement usually had their firing rates, speeds, ranges, or weapon damage slightly tweaked to make things complex. The fact that projectiles will be simulated and that terrain will be relatively complex for the RTS genre is already going to help keep battles of every size interesting.

Complex units need not create a micromanagement problem; but, hypothesizing that it would, there is something instinctively unsatisfactory about 'solving' the micromanagement problem by making it 'not worthwhile' for the player to pursue. The player would still care but they simply cannot justify spending their effort towards microing. This internal conflict between what the player would like to do and what they able to practically accomplish creates the very frustration which I feel is the micromanagement problem in the first place.

I am hoping PA's approach is to make micromanagement less necessary so that there is less incentive to use a small heavily-managed army versus a large less-managed army. Hopefully this allows the scale of the game to be larger because players, feeling that units can take care of themselves, can now focus on increasing army size -- and not because they feel forced to use larger armies due to micromanagement becoming too undesirable or impractical. Better UI or better AI can achieve this and the ideal is probably some combination of both, but that discussion is out of scope for this thread.

I can agree with OP to a point. I don't want to see PA become a large-scale RA3 or SC2 where almost every single unit has some special vital secondary ability that needs to be micro'd. I hated that in those games and loved SupCom (not 2, that one was terrible) and FA because the combat was simple and each unit had its own abilities that didn't need micromanagement.

Zero K is the primary motivation behind the OP. The units in ZK are dynamic and have complex interactions, but don't really require much micro. You can micro them to increase their performance marginally, but the AI and simply using the right units in the right numbers, in the right situation, will obviate the need for micro.

Multiple weapons was just the first example I used as it seemed the simplest to communicate. The Rocko in ZK (very similar to the same in TA) is another perfect example of a unit with more complex interaction than a tank in SupCom 2.

Basically, I agree with linecircle. The reason you don't micromanage individual units is because the gains are small and don't justify expending valuable player time and attention, unless the circumstances really call for it. Big, strategic moves will always be more significant on the large scale, and if the player is making good larger moves then micromanagement will be unnecessary. Still, the possibility of mounting a valiant defense using forces that are outnumbered and outgunned is interesting to have available, and can make exciting an otherwise boring battle where you simply A-move an obviously superior army.

Complex unit differentiation like TA, or even to a greater degree than TA, is not the same as requiring extensive micromanagement. It does make units more interesting to use, both in small skirmishes, and also has cascading effects as the numbers become larger.

To make my point another way- one unit versus one unit should be at least marginally interesting. If it isn't, then divide those units into ten units at one tenth the cost, so at the very least there can be localized superiority, composition, and distribution of forces dynamics.

I think micromanagement improving battle outcome is essential for making the game interesting to competitive players.

Now, this doesn't have to be like zerg surround micro in BW where it's an elegant, mad dance as much about technical skill as it is about strategic thinking; however, this kind of dynamic where each unit requires careful thought to be used well makes for a much richer game.

Of course, with grand scale there are limitations to this scheme, but a siege of an entrenched enemy position should be as much about managing production as it is about positioning. The dynamics don't need to be extraordinarily complicated; heavily armed but weakly armored artillery units which sacrifice mobility for firepower by "setting up" are a simple example of making the game more interesting. The commander must now devote time to deciding if, where, and when to set up his artillery; choose wrongly, and they are wiped out by enemy maneuvers. Just a little bit of nuance is necessary to make the game enjoyable.

I like interesting units but I don't think multiple weapons are the best way to make them interesting. Units are more elegant and easier to control with a single weapon. Give weapons interesting mechanics instead of slapping on an extra weapon to create complex mechanics.

Give units slow turns or some artillery has to deploys, and the micro from the battle will be correct positioning and moving, flanking, etc

As others have said there are way more options than just "Moar Dakka" to create complex unit interactions, in fact I'd say just adding on more weapons is actually one of the worst options unless used correctly.

Mike

Also, just using the applicable weapons at this instant does not solve the problem of managing the units. Should the unit be approaching to close range or kiting the long range to avoid damage? And having different commands for how to attack starts getting into a lot of commands in a battle I think I shouldn't need. If there was typed damage or more actual unit types impacted by different weapons this could work I think (i.e. tank with cannon and mg to fight off infantry swarms) but PA wont have that, a weapon is a weapon and a unit is a unit, that's it.

I think making the use of weapons more complex might be interesting... like my tank could in theory try to shoot at a plane, IF there are not ground units, very poorly... but that's as far as I'm willing to go.

Typed damage is a terrible, terrible idea. I could write an essay about it, but it is obvious to 99% of the people here so I won't waste the space in the thread.

People are really fixating on the multiple weapons example. Inaccurate single weapons (which can miss and hit something else), patterned attack weapons like linear firing slow rockets, units with small regenerating shields, or rapid HP regeneration and low HP...

There are literally countless ways to do this.