Diverse Unit Roles

Alright, I looked up unit diversity and came up with a couple threads similar to this but not exactly expressing what I want to say.

Basically, I'm thinking that units should be far, far more diverse from each other than in Supreme Commander was.

Now, a lot of battles came down to building a big ol' blob of units that just blew everything up, didn't matter what the enemy was doing. Now the reason for this is that nearly all the units just came down to big explody damage dealers. So basically if your blob consisted of a bunch of tanks it wouldn't be too different from a blob with a bunch of bots.

So how about we differentiate unit roles enough that you really have to be careful and think of the right balance of units. Here's an example: Supreme Commander 2. Now, most people hate Supreme Commander 2 because of its lack of units, its research tree, and of course the fact that it has more experimentals than ordinary units. I for one think that last point is actually something reasonably OK. See, the reason that Sup Com 2 had so many experimentals is that they all actually did something DIFFERENT than eachother. Here's a comparison:

Supreme Commander: Forged Alliance Aeon experimentals:

Galactic Colossus:

Walks around, blows stuff up.

Tempest:

Floats around, blows stuff up. Can build a few boats.

Czar:

Flies around, blows stuff up. Build planes.

Paragon:

Makes resources really fast.

Supreme Commander 2 w/ Infinite War Aeon experimentals:

Universal Colossus:

Walks around, blows stuff up.

Sooprizer:

Flies around, blows stuff up.

Urchinow:

Absorbs hits, blows stuff up.

Darkenoid:

Flies around, blows stuff up.

Wilfindja:

Floats around, Blows stuff up.

Airnomo:

Blows up air units.

Loyalty Gun:

Rapidly captures enemy units.

Space Temple:

Teleports Units long distances

Pulinsmash:

Disables enemy units, blows them up.

As you can see, there is a really large pattern of BLOWING STUFF UP. I tried to be as objective as possible so I didn't make it seem like Sup Com 2 experimentals are all fancy, there are still a bunch of blow stuff up units. But the Miscellaneous Experimentals is what I'm really looking at. The Space Temple teleports units instantaneously to a position, no other unit can do that. The Loyalty gun captures units extremely rapidly, engineers can barely compare in that ability. Pulinsmash disable units slowly and instantly destroys them rather than just shoot big laser things that makes enemy units go boom boom.

So please, what I'm asking is we have more units, yes just REGULAR units like those misc. experimentals. A unit that makes emp blasts that disable units but have almost no damage, a unit that shoots gravity bombs to hinder enemy movement. Even already established units such as artillery, bots and tanks could have extreme differences, rather than basic things such as cost, health and dps.

You should probably look at Stacraft, that would have been a much better example as there are hardly any overlaps in unit roles. For each race, units have a specific role that is more or less unique to them, no other unit in that race has the same role. I think PA should do something like this, that's a lot better than making a lot of different units where unit roles are overlapping everywhere.

Exactly my point. I had actually been thinking about how Starcraft does this so well but didn't want to mention it because of how widely different these RTS styles are. But yes, the way you have one unit to do one thing and not twelve is an exellent game element.

You seem very focused on the experimentals. In my opinion the smaller units make or break the game, while the super-expensive endgame-only units are really just toys that are intended to look flashy and sell boxes.

It seems to me that you want there to be gimmicky differences between units. Things like the Magnetron and Alantis that are big, and very different units. However those units really don't change much in terms of gameplay, and even if they do appear, they don't create lots of different gameplay from a single unit- they do what their designers made them do.

Among small units very subtle differences actually become quite significant when many units of that type are acting in concert. And you have options for mixing groups of them, and lots of options for how to use them. "Unit diversity" means a variety of roles; units that play differently and do different jobs. Even if they are both tanks, if one of them is a cavalry tank that is effective at firing on the move, and the other just has heavy armor, those units will play quite differently. It would be incorrect to say "oh they're both just tanks" and ask for more 'diversity.'

Mhm, I disagree. Give me Zero-K's thousands of units over Starcraft's half a dozen.

I don't know much of the Sup Com 2 units.
But the FA experiments have pretty defined roles as do the the main units.

One of the best examples of this from SC:FA:

Aeon T2 Heavy Tank:
Heavy tank. Equipped with a single cannon and a shield generator.

Aeon T2 Assault Tank:
Fast, lightly armored tank. Armed with dual, rapid-fire autoguns.

What an incredibly toxic view point. There's nothing wrong with units taking up non traditional roles. It creates an opportunity for unique dynamics and to flip typical counters on their head. Of course, it has to be designed from that separate angle, and can't just be another average unit that blows stuff up.

Non-traditional is great. But "moar different" doesn't create unit diversity. What you get from that approach is SupCom 2- variously themed gimmicks. Unit diversity is reached by having interestingly different roles to fill, not from "wouldn't it be cool if we had a T-Rex unit" or pretty much anything else. Any reasoning along the lines of "wouldn't it be cool if" is doomed to failure- especially for gameplay-significant features. "Wouldn't it be cool if we had snipers, or battleships, or flying aircraft carriers" and so on. All you get is a mish-mash of superficially different units, and the resulting gameplay is very likely to be boring, along the lines of 'spam unit X' because it happens to emerge as dominant.

Strong gameplay design should inform and determine unit properties, not the other way around. Step 1- decide what kind of gameplay you want. Passive? Defensive? Aggressive? Army-centric? Harassment? Discrete isolated bases? Relatively stable battle lines? Breakthroughs? The idea is to decide what players will find fun. Step 2- what unit roles do you need/want to produce that kind of gameplay? If you want there to be stable battle lines with breakthroughs, then you're going to need to give players tools to form such a line, and different tools to break them such that a player might have one and lack the other. What relative importance does each type have? What quantity? Lots of general-purpose units with a few specialists, or does a large number of interdependent specialists work? And then, lastly, step 3- design units' properties to fit their assigned roles.

this.

Unit classes are how I hope PA tackles unit interactions. OP is concerned that killing stuff isn't an interesting unit interaction - that's probably true in Supreme Commander 2. In ZK most of your units (there are lots of exceptions though) can kill but how they interact is what makes them interesting. Raiders are good at swarming Skirmishers but not so good against Riot Control. That's counter to SupCom 2 where you simply want the best ground attack unit and the best air to ground unit.

Don't get me wrong, a lot of Supcom2 experimentals were rather remarkable. They were poorly thought out and functionally identical to the vast array of other options on the field. That doesn't detract from the high end expis that WERE good. But you seem to be defining a gimmick as:

which is completely wrong. Suggestions such as "t-rex's and lasers and electricity oh my" are NOT game ideas that change the RTS genre, thus they can't even qualify as a gimmick. They are aesthetic choices that are used to represent the underlying rules of the game. Sure a T-rex may look cool, but under the hood it's an assault weapon with short range power. There are thousands of ways to make a heavy assault weapon, and the choice for how it appears has no direct link with its function in game. It was no surprise that the numerous heavy assault bots in Supcom2, while visually unique and even with a few unique toys, did not differ much in function or scope. At the end of the day were all still heavy assault bots.

A proper gimmick is a rule changer. They could be stun weapons, debilitators, or defenders. They could be movement differences like hover, amphibious, walking or ocean crawling. They could be one of the dozens of AoE shapes, percentile damage (D-gun was 100%), or a spectacular death blast. A gimmick alone certainly does not create a unit, but they do create opportunities for new niches between units. Both Starcraft I and II showed that very well.

Too many special tricks leads to a mirco focused game over macro. This game is a macro focused game so anything that makes me focus too much on how my units fight rather than where and when is a bad direction for this game.

That said, I'm all for units filling unique specialized roles, so long as it requires little direct attention from the player to utilize it to maximum effectiveness.

So assuming the unit is automated how far is too far in unit specialization.

I think we can clearly state that we won't have many toggles. Definition: Toggle: a button you push to activate an ability such a the D-Gun or the engineer capture command.

I think we can also clearly state that we won't have any armor or weapon types that do special damage to create rock paper scissors scenarios. I do think that using the right units in a situation will be meaningful however but I think this will be due to the actual weapon trajectories/velocities and unit turn/velocities.

What if we had a siege tank like in Star Craft, but it automatically sieged for you when it stopped moving (or an enemy target came into range) and unsiege when given a movement command?

So, would the above unit be a no-no or would the automation allowing you to attack move it into the enemy base be sufficient for the unit to fit the PA criteria?

I guess what I am asking is are we against micro abilities with toggles or are we against sophisticated mechanics regardless of whether the unit needs microing?

yeah so... RPS is actually how you balance RTS. if you mean contrived armors and damage (it appears you do) that's different. that is to say X>Y>Z>X or Artillery>Point Defense>Tank>Artillery or Firebat>Zergling>Hydra>Firebat are all RPS, the difference is just how contrived the rules are.
http://www.sirlin.net/articles/rock-pap ... games.html

Roles should be about unit functionality, not what the unit is good against. You don't get artillery because the enemy made X or Y, you get artillery because you want to be able to shell the enemy from a distance. The definition of RPS is that your selections' value is entirely dependent on your opponent's selections. Ideally what units you have and what your opponent has will not predetermine the victor, but rather will change the way both sides play.

that's really what modern war is though and even if it weren't a video game is contrived and that's what it is in a video game. in TA artillery is designed to defeat static defense, it may have use in other ways, in the same way zerg guardians can be used to skirmish, but both artillery and zerg guardians > static defense.

no that should come down to battle management. in order for battles to be less binary you need unit variety and that's where balancing comes in. even though a unit counters another it doesn't have to be an "X Defense", it can have secondary uses, that's just proper unit design. X>Y>Z>X doesn't have to mean Y is useless versus X, it just means X beats Y - that's what RPS means. understandably RPS is used to signify contrived mechanics, but let's be serious - video games are contrived.

I probably wasn't clear enough in my original post. I meant that those miscellaneous experimentals with extremely unique roles such as the space temple and loyalty gun would be converted into smaller tier 1 or 2 units, along with other unique unit role ideas aside from making something get destroyed. I didn't mean we should have tons and tons of experimentals and just a few regular units.

It is my assertion that a unit built around a feature like "teleporter" or "capture gun" is a gimmick, not a "diverse unit."

In order to create a unit you should first come up with what the unit's role is- what jobs it needs to do. Then, you come up with properties and tools the unit needs to do that job. You might even end up with a unit that has a seemingly gimmicky feature which it uses in the execution of its role. Lock-on missiles are cool, but they have a gameplay function of delaying when the projectile is fired, changing unit interactions at different ranges and speeds- i.e. if both sides are firing, one side's lock-on delay is a big problem. Just one example of infinitely many possible implementations.

But the player isn't going "ooh, now I want to build that unit that has the teleport!" This is one of the reasons why SupCom 2's design is just so awful. Units like the C-Rex and Loyalty Gun are gimmicks governing unit design to wow players for a moment, not a tool used to solve an in-game problem. Even if they're powerful units, even if they give you an advantage when constructed, there's just no substance to those decisions.

In a strategy game the player is analyzing the game state as a problem to be solved, and must decide which tools they want to pay for in order to best solve that problem. The game designer is doing the process in reverse, first deciding what kind of problems are fun to solve, and then figuring out what situation and tools they need to offer the player to create those interesting problems.

Alright, I can give you the Cybernasaurus Rex being just another big walking gun, but I fail to see how a unit that instantaneously moves entire armies across the map without possibility for them to be harmed in the process is "gimmicky".

I really don't think you want to hang your hat on the Space Temple as a concept. But I will concede that if your intended role was to create... what the Space Temple was... then it's not so much a gimmick as a total failure to design a unit, since it makes the game extremely boring when used.

The Space Temple is just an incredibly boring design. At a stroke it destroys what little positional play SupCom 2 had, since you can move any army anywhere instantly. This functionality was what those devs considered "cool" and they designed the unit around this feature, which makes it extremely expensive.

Because I believe the Space Temple was created because "teleporters are cool" I consider it a gimmick-designed super unit. If you wanted to make an interesting teleportation structure, I bet it could be done, but it would be nothing like the Space Temple. My approach would be to make it far more limited, with interesting constraints about how it can be used, and in addition make a few options for counter-play available to the other player. Figure out an interesting teleporter sub-game, and then design the unit to fit the intended new dimension of "teleportation war" gameplay.

A two-way teleporter which requires a constructed gate on both ends captures the logistical features of the Space Temple better than it does, and its total cost can be much lower due to its lack of offensive potential. In addition, perhaps a one-way teleporter allows a player to teleport anywhere within a limited distance of any other player-controlled teleporter, or much cheaper teleport beacon structure (with mobile variants available). It would cost a large lump sum of energy to activate such that it would be a significant investment in the late game, and require case-by-case decisions about whether to use it or not. The other player can scout and destroy teleporters and beacons, and perhaps might also need a structure which makes teleportation into a certain area impossible, or perhaps highly inaccurate. Or, perhaps, makes any unit teleporting into that area simply disappear, which would mean the teleport-user needs to scout to ensure their teleport won't splinch their troops. This type of teleport gameplay as a starting point would be much more interesting than the Space Temple, and now we can start figuring out its teleport mechanic, how much it should cost, its range limitations, etc. etc.