Armor Systems

Well you have to remember at least 25% more about every unit as well as increasing the importance of all the other information about units. Basically its an increase in complexity. to skip to the end a bit, i consider it an unnecessary one.

Why should a tank be immune to infinite weak units? that sounds like bad design.

Furthermore you seem to have a hatred of large numbers, a unit having 5000 hp is nothing to be scared of. If i'm sinking 1200 metal into a super heavy tank i'd feel gibbed if it only had 700 hp.

as you say, avoiding rock paper scissors is a matter of balance. All the example you present though are very rock paper scissors scenarios though. light units physically incapable of harming tough units.

At the end of the day, as you say even in without armour things could become RPS, my question is that if the game is already capable of all these things, why introduce the extra complexity? The tactics are already present, as is the game effects (even if they are slightly different in appearance and you gain a potentially misrepresentative unit health.

It's like 16 bit vs 32 bit colour. You still have red, yellow, blue, orange, etc, but with 32 bit you have all those gradients in between that just aren't possible in 16 bit. But it cost you extra memory...

With an armor system you can get just a bit more definition in the differences between unit roles, and for those in favor of armor, we believe it's worth the performance cost.

Those gradients are possible without armour i can accept that it is easier to balance such things with +1 scales of granularity.

We have now reached a matter of personal prefrence pawz *cheers*, there is no long a reason for me to continue this conversation.

unless of course someone brings up directional armour. Or someone suggests that armour should be so strong as to create a rock paper scissors situation.

Quite the opposite. I want nothing more than for PA to have HUGE numbers of units and HUGE maps.

However I do have a problem with a huge number of units which must fight together. If a player has 500 light units they should want to make a large number of small groups, not a single blob with the entire number. A tank being immune (or having a minimum damage threshold) to infinite weak units prompts the player to split up their weak units. The huge group of weak units has the option of splitting up and spreading out. This is impossible for a single more expensive unit.

I think we agree that the actual HP number is irrelevant because it depends entirely on the weapons used against it, and how they interact with the unit itself. Whether a unit has 5000 HP or 700 HP is just an abstract number until we have weapon damage and other relevant factors to apply. Armor is one of those factors. If that unit has 700 HP and a large amount of armor, it becomes effectively very durable. The HP number is not important compared to the effect that the unit is durable in practice.

It is important to remember that in actual gameplay, players don't give a damn about HP. It could be 5, 50, or 50,000. The important factor is the vital hit number; the number of hits before the unit dies. A unit with 100 HP being hit by a 25 damage weapon is dead in four shots. Four shots matters much more to the player than the numbers for HP and damage. A slightly more involved calculus for to compute the vital hit number is perfectly acceptable. Against a unit with 100 HP and 5 armor being hit by a 25 damage weapon is dead in five shots, because each shot reduces its HP by 25-5=20, and 100/20 = 5. Flat armor adds another element to the vital hit calculation that creates more variation among unit interactions.

That said, the vital hit number should generally be kept to a low and discrete number for important cases. Nobody has any idea how many hits a Krogoth takes to kill with a Rocko. Heck, I don't even remember how many hits a Krog takes from an Annihilator, but I do recall it was quite a few. It makes the Krogoth a fuzzy quantity. This is bad. It would make more sense to have lower HP with armor, so a weapon like the Annihilator can have a discrete vital hit number against the Krogoth. If this requires downsizing the Krogoth to make it work, so much the better. Good principles pull in similar directions. More units, with each individual unit's behavior being simple and discrete, is good. This is something Starcraft does exceptionally well, and which TA would have greatly benefited from while maintaining its own game philosophy.

What is a hitpoint? It doesn't mean anything outside of the context of what other units have and do in damage. 5000 could be big or it could be small.

Thanks for the tautology, champ.

The concern was obviously over having units sacrifice their HP to gain armor rating. This is seen very clearly in Dawn of War II. Tanks have a few hundred HP compared to the thousands of HP on infantry units. However, armor types make the vehicles more durable and powerful. It's kind of strange to see.

What you say does NO such thing.

Units get split up when death balls stop winning. This has to be accomplished through many angles, and can not be solved by a simple armor system.

Absolutely right. I never said armor alone would kill the deathball completely, only that certain good implementations would tend in that direction. A bad flat armor implementation might even encourage deathballs.

Still, using HP only, a blob of small weaponry units can kill units with a large quantity of HP. Flat armor can be used to change this dynamic in many ways.

I completely agree though that no armor system is going to eradicate the possibility of deathballs. PA's gameplay needs to be designed from the ground up to avoid deathball gameplay. It bears repeating as many times as necessary.

And PA wants to make a game of unrivaled scale, so it goes triple for PA over and above other contemporary RTS titles. Huge numbers of units is pointless if you keep them all together in a single blob. And huge maps are pointless if only the location of your one blob (or one main base) really matters.

But armor doesn't do ANYTHING to address death balls. Like. At all.

All armor can do is create harder counters.

I'm not really sure about death balls potential anyway.

In starcraft the deathball works because its a reasonably compact formation of all your damage.

It works a bit with restorers in FA.

but with ground units and armies 200 strong... you cannot condense the damage well enough to get a responsive deathball. Similarly because of the prevalence of splash attacks in these games if you did, every piece of artillery that lands on you will be at 10x effect.

To go back to starcraft, i'm no expert but i'm pretty sure deathballs aren't the best tactic against deployed siege tanks.

Oh ye gods don't even get me started on the waste of space that is the SCII siege tank. 3 supply, 125 gas, and it deals 35 damage? The Brood War tank was 2 supply, 100 gas, and did 70 damage. And they completely took out the Lurker. Because Browder is pants-on-head retarded and thinks positional play is "boring." The SC2 team wants constant motion and kills everywhere like the ridiculousness that is ZvZ early game. They actually LIKE that ****. And we lost the Lurker, one of the best units in any RTS ever, in exchange for some kamikaze crap. And the Siege Tank is a shadow of its former self, and the Goliath is straight up gone, replaced with this ugly, clunky, giant, expensive, and useless mecha called the Thor.

You are incorrect about this. Having armor allows lighter weapons to be more efficient DPS sources, because they will be less effective against armor. And these lighter, more efficient weapons on smaller, cheaper high DPS squishy units are best when used separately. Units that are short range or squishy or high DPS for cost are less efficient to use in huge groups than in smaller groups.

Secondly, armor allows heavier weapons to kill heavy units in fewer hits. This means a deathball of heavy units is easier to chip away at, and easier to inflict individual casualties. A deathball of high HP units is hard to even hurt because each individual member can take so many hits, and thus so much DPS-adjusted time, to kill. Grouping these high HP units stacks their DPS and kills enemies damaging them more quickly. The deathball runs away when it has enough DPS that the enemy hasn't enough time to kill one member of the ball.

With armor you can have tough units exhibit some of the properties of "squishy" units. A big deathball of heavy tanks can take casualties against relatively few high-damage hits. This eliminates one of the advantages of being a deathball- the ability to have enough stacked DPS it seriously mitigates your resources or strength lost through casualties.

The Goliath has been replaced with the Viking. The Thor serves a completely different purpose from the Goliath.

Also, I think the best way to deal with the concept of death-balling has nothing to do with their DPS or HP, the best way to get rid of it is by making it slow. (Or the distances bigger). When your location matters, you can't rely on keeping all your stuff in one place, because every other place on the map will suffer and there'll be nothing you can do.

Oh, not this silly argument again. The Goliath was unlocked by the armory. The Thor is unlocked by the armory. The Goliath was made by a factory with a machine shop. The Thor is made by a factory with a tech lab. The Goliath was a walker that attacked ground targets at 7 range and air targets at 9 range. The Thor is a walker that attacks ground targets at 7 range and air targets at 9 range. The Goliath got huge, expensive, and weak- that's what happened.

The Viking is a replacement for the Wraith and Valkyrie. For starters, it flies. It can be attacked by air to air units. It is not a Goliath. Its ground transformation is pitifully weak, and is more akin to the Wraith's 6 damage ground attack as a weak secondary functionality to its air to air missiles. Speaking of which- WHY transformers? And why on EARTH did they add a second transformation unit in Heart of the Swarm? Gimmicky nonsense to appeal to stupid kids who like the Michael Bay movies.

Now, once upon a time in the pre-alpha, the Viking was intended to be a Factory unit which could attack ground and air. That would be a legitimate Goliath replacement. However the current Viking is not even close. It's a flying Starport unit that attacks air units, not a ground unit that attacks air units. You can't go Factory mech and then immediately switch to producing Vikings like you could Goliaths when the enemy shows up with carriers (or Tempests or Brood Lords in SCII). The Thor is the new Goliath, and it is a 300 mineral, 200 gas, 6 supply waste of time.

This is under the assumption you have many mission-critical areas to defend. Mexes alone won't force a player to commit large numbers of units across the map to defend them. It's cheaper to let the enemy kill them and rebuild than to spend hundreds of times the mex's value defending it.

If the map is largely empty space and it confers you no actual combat effectiveness advantage then you really have no incentive to spread your forces. You say "the rest of the map will suffer" and the savvy player goes "totally OK with that." There needs to be a reason to spread economy and industry to force players to split defensive forces. It will always be more militarily efficient to consolidate in one place and heavily fortify it. It is more efficient (if you have the APM) to use troops to clear an area (if it is even necessary if the opponent is doing the same thing) and send another constructor to rebuild mexes, and then move on. Repeat forever. You don't actually have to defend or even have a persistent presence everywhere on the map.

In my opinion this is the real reason why SupCom adjacency bonuses were stupid. It is already advantageous to build economic structures close to each other because they are easier and more efficient to defend. What should be encouraged is spreading your economy and industry across the entire map. This would create a tension between what is economically efficient and what is militarily efficient.


Back on topic, just because units are slow doesn't make them ineffective as a deathball. It makes the deathball less effective, sure, but if it is a unit best served by a deathball then it still makes sense to ball them together. You then run the risk of having a slow, strong deathball. Which I suppose is a significant improvement over an omnipresent deathball like SupCom ASF's, but not by much.

I would also like to point out that "strong and slow" is an even worse-abused design concept than "strong, but can't attack air." Actual limitations, liabilities, and weaknesses are much better. But weak players often find limitations and liabilities "annoying" and would rather just build something that is easy to use and decent against pretty much everything. There's nothing wrong with it, but most RTS games take it as the One True Strength Limitation and use it on everything, which is ridiculous.

You know you came across as someone who understood unit roles, so this total failure to understand the purpose of these units is interesting. The Goliath and the Thor are nothing alike. Other than the basic "looks alike" and "comes from the same building" the two don't match in their intended purpose in any way. The Goliath and Viking at least have some matches in how they are used, even if they aren't that clear since the whole unit roster got a pretty serious overhaul.

As to "why transformers", because it's cool, obviously. It's a shame the transformation is hardly ever used, because conceptually it's pretty sweet.

Starcraft II is derailing this thread, but I suppose unit roles are related to armor.

I think what you mean is that Blizzard intends the Viking to be the Terran counter to units like Carriers, whereas the Goliath was the intended counter to Carriers in Brood War. In their little "help" section it lists similar unit types that the two are intended to counter.

However the ROLE of the Goliath was as mech anti-air. In order for classic mech, Vulture-Tank, to function, you need a lot of factories. If you have a lot of factories and the enemy builds air units, you need a unit you can build that attacks air. That is the Goliath. It allows you to have factories and fight air units, because neither of the factory's other units in Brood War can attack air.

In Starcraft 2, the factory's alleged anti-air is the Thor. Whatever you say about what unit is intended to counter what, if you build a lot of factories then Vikings are simply not on the table. You would need to build many starports to switch over to a large amount of Viking production. The Thor, however, is an incredibly bad anti-air unit. Even ignoring how weak the tank is, this means the factory has no anti-air, and factory mech is non-viable. To some extent Blizzard even introduced the Widow Mine in Heart of the Swarm with the intention of giving the factory better anti-air. Very weird. Now, it is worth noting, that the Thor is so bad as an anti-air unit that it is often the correct decision to actually spend a ton of extra money to build 3-6 Starports just to make Vikings instead. And that does happen. But if you can possibly get away with it, most players use more Barracks for Marines instead, since it can be made usefully earlier, bolsters their army with Marines earlier, and isn't dead infrastructure later. Still, you can't mech because the Factory cannot get that job done- it needs the Barracks or Starport to cover its incomplete roles.

In short, I am sure if you took Brood War and either removed or nerfed the heck out of the Goliath, we would see Brood War Terrans building more Barracks or Starports for Marines or Wraiths when they need AA, and we would see a LOT less TvP mech because the carrier switch would immediately end the game.

It is you, not I, who has the meaning of unit roles mistaken. The role of the goliath was to be mech anti-air. The Viking does not do that job (although it is certainly more effective anti-air than the Thor). You might as well say the Goliath is unnecessary because of the Marine, and that they do the same job, as many people do. But then you find yourself with 10 barracks and 2 factories going "derp, I am meching." And even if you want to say the role is actually to counter big fliers like Carriers or BC's, there is still one massive difference you are overlooking. The Viking is itself a flying unit, and the Goliath was a ground unit. This is such a huge difference it justifies having both units because they exist in entirely different realms, and can be damaged by entirely different weapons.

The Viking is just like the Wraith. A Starport supplemental anti-air unit. You could use Wraiths to counter Carriers, Battlecruisers, or Guardians (lol) in Brood War also. You just didn't see it often because having that many Starports was not as useful as having more Factories for main unit production. And you could build Factories because the Tank was actually strong, and you had the Goliath for anti-air if needed.

Regardless the reason why i brought up starcraft was to say that those long range splash weapons on the siege tank are the equivalent of 50% of the guns in sup com or PA.

I've played a decent amount of online play and excepting restorers i've yet to see anything that would classify as a death ball because a blob of units is just so vulnerable to artillery and harassment. They are always either in motion towards a target or hiding, very rarely do they group up tight for maximum damage dealing. Then anything with the dps to call a true death ball is several hundred units and to big for any of those guys to apply their dps to the even remotely the same target.

I think its pretty safe to say that they are armies and not death balls

as for using weak damage dealers in smaller groups, wouldn't that just let them be killed very easily? If you had them protected with your own tanks maybe but putting everything in the one army tends to make these armour differences melt away because you aren't controlling them (a damage bonus here and penalty there i mean).

No. The only point of armor is to create a new layer of RPS. At the end of the day, all it does is create a layer for more COUNTERS. Armor has absolutely NOTHING to do with death ball play.

Army splitting depends on a strong need for scouting and good game design. Territorial control and harassment are the most important reasons an army gets split up. If harassment can't deal good damage, there's no reason to split up, so the best option is to plow down the middle.

calling it a new layer of RPS might be a bit hard, however the basic premise of unit counters is true.

Surely it's all relative? I was thinking of letting you guys thrash it out between yourselves but I suppose I might as well get involved.

A death ball by definition is an unstoppable concentration of a player's force in as compact an area as possible. Depending on a units range or it's ability to overlap the same space, they can become a viable death ball strategy. As rightly stated, once the weapon range is exceeded by the depth or width of the formation then the death ball effectiveness is reduced as the whole point is to concentrate fire power. This is where Restorers are really powerful, not to mention that they weren't vulnerable to air either as that was their main role, where as Percies and Bricks as a comparison, obviously need AA support. Restorers are also pretty damn fast compared to Percies and Bricks, which can't cover terrain particularly fast. This is the other big strength of the Restorer death ball: It can move large distances relatively quickly - negating the penalty for concentrating your force into a single area.

In my eyes, Percivals and Bricks were powerful units, but they aren't really pedigree death ball material. Ground units will always be restricted by the fact that they can only occupy a certain area before their range limits the number of units that can fire on any single target simultaneously. Therefore you have more of a sharks teeth effect, where as a unit falls, another steps forward to take it's place. It's either that or you split the death ball. Restorers on the other hand are pure, single unit, death ball spam. Extremely effective and very simple to do. This is what we don't want in PA.

The biggest problem with restorers is that they broke the RPS rules that all the other units follow. If you are going to make a game with those rules, all units should adhere to them. Effectively in the Restorer we have a single unit that encompasses (for arguments sake) rock and paper in one package, meaning it is not really vulnerable to any other unit in the game. They were too effective at air to surface attacks when their main role was an AA gunship. There are plenty of units that have secondary weapons, like the Broadsword for example. But you wouldn't even know it had AA weapons on board they are that useless. They might fend off some T1 air but that's about it and therein lies the heart of the matter.

The fact is that you cannot stop players using a strategy if it is viable. Even in a game like supcom, in which the maps are huge even when compared to TA, players would still concentrate force and focus it on relatively small areas. It's part of playing the game and it's not a bad thing to bring all your forces to bare on where you think the enemy is the weakest. It's actually the smart thing to do in most cases. But it should have risk attached and with Restorers there wasn't that element of risk.

In fact I would go as far as to say that the whole game was centred around concentrating more force into a smaller area. That's effectively what going up the tech levels does. Units occupy less space when we use HP and fire power as the comparison. It also has other effects, like increasing the power of your army while maintaining your unit cap and so on, but that's another part of the story.

What you have to ask yourself is; why is that strategy viable and do we have a problem with balance? In the case of restorers, yes we do. But in most cases there is significant risk involved in concentrating your force in one area. It could win or lose you the game. In terms of unbeatable combinations of units, I'm not concerned. I don't think there's any reason to suggest that there will be any occurrence of this in PA. The guys working on it are far too experienced to make a school boy error like that.

My argument for armour really hinges on the fact that compared to the blunt tool of using HP to make tough units tough, armour is an elegant solution that actually simplifies things if you do it right. If a player can assess a units' armour value and assess his weapon's armour penetrating value then he can see how effective his unit will be. If you see an armour value of 10 and a penetration value of 5, you know you are only doing 50% damage. The higher your weapon penetration, the more damage you will do.

Obviously the closer the figures the more efficient you are going to be at cost:effect. It might even be that a penetration of 4 against an armour rating of 5 might still be cost effective. If the figures are displayed in game when a unit is selected, then this isn't even a calculation. It's just a case of looking at numbers and assessing if the weapon will be effective enough. As for armour, it's just looking from the other side of the fence. And that's just one way of implementing armour. There's a few suggestions in this topic about how to do it and they all have their merits.

Deathballs have little relevance to the armor discussion, other than the peripheral fact that armor would be a factor in balance.

Badly balanced units could create deathballs, armor or not.



Armor just gives us the tools to more finely tune that balance.

Personally I'm in favor of a flat type of armor, and then on the flip side, a flat type of armor piercing. There should be weapons that do low(er) damage but punch through armor.

At a very simple level you could just start off with 5 levels of armor, and 5 levels of AP, and then users very quickly would learn and understand that 'light machine gun' does not do much damage against 'heavy armor'.

It may be useful to lay out some scenarios that would be possible with an armor system, but not possible without it - if the naysayers can come up with good alternatives for each, perhaps that might spur further useful discussion?

Lunch is over for me otherwise I'd start

I agree with almost all of your post.

This is only simple when you have a 1v1 unit fight. Large blobs of varied units and unit counts make any kind of eyeball estimate of strength extremely difficult when units don't do consistant damage to one another. It is strictly more complicated than HP only.