Solving the Mutual Annihilation Issue

Idk, maybe he hints that it isn't an issue, that it is a non-issue.

You see ZaphodX build thousands of units in less than an hour, and he even loses to artillery and nukes if the player uses them against the army while building a "sufficient" army with defenses. He isn't completely undefeated.

Point is, there is a lot going into balance... cost, unit speed, health, damage, build speed, metal available... Maybe this problem is solved though balancing various things to where ZaphodX's production of units isn't TOO much better than a newbie's production and battles do last a little longer but not very much and units simply find their way into field battle sooner and faster.

Well all that is a given to anyone that even remotely understands the balance levers available to a game like this, the problem is that The Unit cannon and Teleporter don't directly affect engagements, they might make it so that there are more engagements in more places but that doesn't change the engagements themselves you know?

Mike

Well, kinda. Transporting units faster of any type basically makes the combat "repleatable". You know what I mean? It isn't the health and survivability of the unit then. It is the health and survivability of everything you are currently building that is being poured into that battle. You could be building bots at 5-bots-of-hp/sec and dumping them all into battle in a cycle so its a constant battle that loses lots of health but constantly refuels health as well.

It isn't EXACTLY the same thing, but if one was to balance it to work right now and then add transports then battles would be stalemates because TOO MUCH health is able to be dumped in constantly into battle. They do need to buff it, but not too far. I would honestly say 2x more health for everything would be a start. EXCEPT t2 mexes and some other t2 structures, lol.

That is still not really aligned with the topic thought, the issue has nothing to do with getting units to battle, but that once engaged the battle is both decided very quickly and tends to result in one side winning by only small margins.

I don't think imporved transport does anything to length the engagement at all because right now it's over way too fast in most circumstances, it'd be more so that you'd just have having more small engagements rather than one big ongoing one.

Mike

Oh, that also. More units need to survive engagements with damage, but right now is better than the alternative: One side loses a whole army and one side has it's whole army just all injured.

Basically, that I DO like right now, which is if 50 units go against 75, sometimes the other guy will be left with 15 left. That is "effectual" because at least the smaller army accomplished "attrition", where otherwise losing a fight means a total loss on the loser's end, it is better for the loser to at least have "done damage".

If the losing fights can still cause damage to the opponent army, and still go on for about twice as long (so you don't need split second timing to react and micro-skillz), then that would be much better. As long as it doesn't start having the problems from the "other end of the spectrum" of balance (too much health)

Well obviously closely matched forces should have close outcomes in terms of basic combat, the problem is that currently we have very little in terms of tools or techniques to shift the outcomes in our favor aside from just bringing more units. Things like new Units and the more varied compositions that stem from them help this issue, but we also need things like flanking and other such tactics to have a noticeable affect as well and I feel like right now at least there is still a lot of room for improvement in this regard.

Mike

Re the 'sweet spot', there's a time-linked note on this here: https://forums.uberent.com/threads/balance-discussion-meeting.51399/page-4#post-788576

EDIT: Direct link

I like artillery having a faster forward-traveling top speed, with slow turn and no turret rotation, to keep it viable to travel with tank armies. Formations are coming eventually. That makes repair, artillery, and lines instead of blobs, add to the game's combat encounters.

Besides that, being macro, it don't need too much more in the way of "more tweaking", besides blobs being made of any more than 2 direct combat units, repair engineers and artillery being viable, formations of lines staying in formation when traveling. Everything else this game needs is interesting balance and interaction on the grander scale, between types of tech and types of strategies. Orbital for instance could be much more interesting.

It seems like a lot of players are confused about what I meant about unit HP.

Low unit HP on large numbers of units is good because it means a large, expensive army consists of many individually squishy units. Thus, an army occupies a lot of space and can be used in a lot of different ways. When both sides have squishy units, you have to be cautious especially with large army movements because you can lose a lot of units easily, but you can also pick off enemy units easily. This is much more desirable than having fewer units that are individually tougher, which you just blob together and attack. By sheer force of HP you will guarantee that you will close to range and destroy the enemy.

However I do think it is a problem that blobs of units collide and just mutually annihilate, leaving nothing behind. For a small portion of an army acting as shock troops this is perfectly acceptable, but for an entire army to fully engage and be gone so quickly is actually an issue. An army should be a large, durable, and weighty object on the map and battles between armies should resolve slowly even if individual unit groups die quickly.

The way to slow down engagements is with soft controls like deterring charges, giving players ways to pick off enemy units for free, and giving players unit types that are less efficient in pitched battle but have other advantages. Using these units means the player is getting less DPS and HP for cost, which discourages just attack-moving by the player who builds them, and these units can also deal a lot of damage efficiently in the right circumstances, which discourages the other player from just blindly attacking.

The engagement slows down because both players are exercising caution, and being tactical about how they maneuver and use their units. In fact, much of the time while the two armies fighting each other are facing off, it is entirely possible that nobody is actually shooting. But when two squads of units collide at point blank range, that little confrontation is still over almost immediately.

Obviously the brute force solution remains on the table. You can in fact just muster up a giant army of units of try to muscle your way wherever you want. However that type of approach should have a variety of fairly severe counters that make such all-out assaults (as are currently very common in PA) bold, high-risk moves. Slower, more cautious play trying to pick off enemy units highly efficiently with minimal losses using any of a wide variety of units and tricks should be a continuous, slower process.

Relatively slow-paced battles could continue to receive reinforcements, escalating the conflict in the region. Relatively few units are dying because both players are being cautious, meaning the reinforcements increase the number of units in the area. Highly aggressive plays can cause a lot of casualties in a big hurry, and may be impossible to replenish quickly. So if you make a big assault and the enemy crushes it because you seriously miscalculated your attack, you may lose the entire region since you lost a LOT of units and the enemy lost relatively few.

Tactical maneuvers will always be better than blobs. I still use them when defending on a line (it is easier to get formation set up stationary with repair bots and such).

There is a bit of a risk for a low skill level, to be able to lose such a large investment of an army all at once without time to react. That sounds like Vietnam in the 1969s really though. The point in this game, is that a large army isn't entirely an investment. Those units really are easy come easy go. They are as expendable and replaceable as they come. 10 factories can be pumping them out 1 instantly a piece.

Really, a base is more of an investment. Even mexes, especially t2 factories, god forbid nukes, all are much higher risk investments than a couple hundred units.

Absolutely, troops should be very replaceable. However, if you can eliminate enemy units efficiently and are still producing troops at the same rate, your army gets bigger instead of merely replacing units that have been destroyed.

Furthermore, if maps are suitably large, there should actually be a significant amount of time between the unit's creation and the time when it reaches your army in the field. When an army loses units, even if you have a prodigious amount of production a long distance away you could still be defeated by the enemy army. Losing units means you have less ability to destroy enemy units, which makes it more likely you will lose more units. This is why choosing where to send reinforcements is such a hard problem (unless you only have one army, anyway).

Even though the units in the army are replaceable, an army in the field has a job to do with only the units it has. Reinforcements will take time to arrive. In that time, you might lose more units, or even be completely destroyed. It is the threat of being decisively defeated and losing the entire army in a terrible exchange that creates the incentive to retreat from a disadvantageous situation, even though you are surrendering control of metal spots and valuable territory that may be difficult to retake.

Expediting reinforcements by building forward bases and using transports becomes valuable precisely because of the large amount of time needed to traverse the map. A forward base should certainly be more valuable than a hundred units, but this is because of the huge amount of time that forward base saves your units from having to travel from the rear.

@ledarsi, the engagement slows down, but the level of micro increases (and pretty much introduces a paper-rock-scissors mechanic), and stand-offs are quite simply boring and not what a game involving annihilation should be about. It also allows players with slightly more skill than others to steamroll other players, since you haven't alleviated the issue of small mistakes costing large parts of your army, which cannot be alleviated unless you increase (to use Nanolathe's term), their effective HP. I'd go as far to say I'd prefer they don't introduce more units than make units that can "freely" kill others without doing something about unit survivability, as it only makes units' lives even shorter.

In regards to re-inforcement, I disagree with the statement "an army in the field has a job to do with only the units it has". No army in the world acts like this on a large scale. On a squad/small group scale maybe, but that reinforces why this view goes against what PA is about.

it's not related to the formation's bit it's destined to the thread in general, as it's where you should vote.

There is actually a tremendous amount of great stuff hidden in this short paragraph, but I will try to address each point briefly.

First, yes it creates the possibility of micro. However there really is no way to have interesting combat without creating the possibility to improve the effectiveness of your units by some method. If there exists any method of increasing your units' effectiveness, then controlling your units will allow a player to achieve that increased effectiveness. The approach PA should take is that despite the fact that you could micromanage your units to increase their effectiveness, most of the time you have large-scale, strategic concerns to worry about, and are too busy to care if one individual unit is being micromanaged perfectly. However PA should have interesting combat, and that necessarily creates the possibility of controlling your units.

Second, yes it causes player decision-making and skill to become more significant. In any serious game of strategy, the more "skilled" player should usually win. In Chess, an inferior opponent really has zero probability of beating a stronger opponent. However I would like to stress that there is a world of difference between strategic skill and mechanical skill. PA should minimize the need for mechanical skill (pressing buttons). However PA should most definitely make the more strategically skilled player have a tremendous advantage.

Third, I don't consider taking every single unit you have and blindly attacking into a well-defended position a "small mistake." I would consider that a colossal mistake. And it is also not a mistake that a player is likely to make twice, since the outcome is extremely direct. Furthermore, in my experience, weak RTS players never actually do this. They lose because they fail to expand and they make lots of turrets on a small base. Weak players will lose because they are weak, and attempting to design a game so that they can win anyway is foolishness.

And lastly, less efficient units that can kill only a few enemy units at a time, but can do so very efficiently, actually do take considerably more time to kill a large group of units than using a large group of high-efficiency units with excellent DPS and HP for cost. Even though they can kill a couple enemies without dying, you are paying for that extra range, alpha strike capability, splash, or whatever other advantage the unit offers. You are getting less raw DPS and HP. Using these units means it will take longer to kill the enemy, but can give you an advantage in the process. These units are also valuable targets for the enemy to eliminate, and it makes sense to protect them, conceal them, and use them judiciously.

That is interesting, I see it completely the other way around. An individual squad always has the option of calling for help, either from other units or from artillery or air support or what have you.

An army is a very large group of infantry, tanks, helicopters, planes, perhaps even ships, and so on. But an army is quite literally on its own in the theatre. The army can't call for help because there isn't any extra help to be had; its reinforcements are already being produced and transported as quickly as possible. The army has to win in the theatre on its own. Losing a large number of units is a big problem, and weakens its total strength substantially.

I love that'd anybody would assume its easier to achiever a stalemate with more contrast.

when watching a starcraft match for the first time you're like : "wtf is going on? this guy just quit because he "lost" ...apparently, and I couldn't freaken tell anybody had the upper hand in is this god-forsaken omelette! switch the channel!"

when watching a supcom match for the first time you're like : "wtf is going on? there's this huge *** hideous blob of monochrome cereal and another hideous blob of different-colored monochrome cereal but that one's really tiny, and it's getting ...eaten? by it? I guess?. yyyyay bigger blob...? yyyou win...? Oh yeah! I was right, he did win, yay me, now back to watching something that doesn't look like vomit"

well aside both of these extracts showing a clueless bastard you can clearly see more health doesn't result in stalemates. it results in making a bigger difference. ....well... and not being limited to 200 supply, which helps amplify this phenomenon but you catch my drift.

I think it is par for the course in a TA style game that the game begins with aggression and raiding. Assuming both players are reasonably capable, it then proceeds to a relatively stable state where land has been secured by both sides.

What follows is a "stalemate" where both sides have troops as well as quite a bit of secured land that will require more troops to clear than either side has. However the game also has a variety of tools for breaking stalemates in different ways, including assault units, artillery, and so on. Up to and including nukes. The structured 'stalemate' is actually quite important.

PA adds the additional possibility of having an entire planet be secured, and a defended planet stalemate can be broken by smashing an asteroid into it.

I think it needs to be pointed out that there are actually two issues here that are being conflated. To summarise, the statement that Ledarsi initially presented can be written as:

Battles in PA are currently dominated by mutually annihilating blobs of units, which destroy each-other extremely quickly.

The issues are that battles are:

1. mutual annihilation of blobs
2. over ridiculously quickly

Although linked, these are still two distinct different problems. The first, is that simply most battles leave very little left alive on both sides. Attrition tends to be very symmetrical for most engagements, meaning that the only path to victory is simply to bring more tanks to the battle than the other guy. Ledarsi has correctly pointed out that the only antidote to this is to add more variety to the unit roster in terms of potential roles. If well used, different units can leverage their unique strengths and weaknesses to create situational advantages, meaning that what has been brought to the battle is just as important as how much. This is an obvious conclusion, and one that I have absolute confidence that Uber are working on. I expect we will start to see new units trickle into the game over the next few updates, which will greatly enliven the state of play.

However, this doesn't address the second concern. Just because we have 50 gunships vs. 50 tanks instead of 50 tanks vs. 50 tanks, it doesn't automatically follow that the battle will have a slower pace. In this engagement, we would expect the gunships to come out on top, and rightly so. However, it doesn't say anything about whether this situation is reached in 2 seconds or 20 for the given number of units. The pacing of battles is not fixed by adding more unit types.

So the next question is, do we want to slow down battles? I would suggest that the answer is absolutely a yes. Even in the asymmetrical gunship vs. tanks battle I have just described, there is a very different character to a fight lasting 2 seconds and a fight lasting 20 seconds. A good player, watching their tanks wander into a gunship ambush, knows that the tanks are doomed. However, they can take actions to mitigate this defeat. The most obvious one is to withdraw the tanks into the protection of some ground-based AA. If they have fighters, they can send them in to protect the tanks, and possibly even turn the tables on the gunships. Unfortunately, they can do neither of these things if the tanks are already dead by the time they have focused on the engagement. Fundamentally, slower attrition rates reward adaptability and flexibility while the only thing that higher attrition rates do is punish a lack of foresight. This results in a high skill ceiling, and a high requirement for experience of what battles will be lost before they are even started. There is less room for adaptability and being able to think on one's feet. I think that this is an issue that needs to be fixed in order for this game to reach it's potential. To clarify, there is also an issue with battles which go on too long as well. The game becomes less interesting as the pacing becomes glacial, micromanagement becomes an increasingly valuable skill, and deathballs can become more prevalent if units become far to hard to kill. However, to quote bobuncles here:

There is no reason why a good compromise cannot be found. So lets look at the factors which can lengthen engagements:

1. Wreckages help lengthen times by creating a tough wall that units must navigate around. They block shots, thus negating the amount of damage taken. Interestingly, wreckages are dealt with differently by different units, in that more manoeuvrable units can navigate wreckage fields more easily. This also makes reclaim and resurrection more viable features. Both of these are interesting mechanics for generating emergent strategy and unit differentiation, so I would heartily recommend that wreaks are made tough enough to survive multiple shots before disintegrating.
2. Inaccuracy of shooting is also important. This effectively boosts unit durability by meaning that less shots connect with their target. At the moment, it is a little unused, and most units are extremely accurate (excluding the odd bug in the firing solution code). However, an interesting aspect of inaccuracy (when target leading is still in use) is that it biases damage towards large clusters of units. In other words, this acts as something which subtly biases against deathballs, as they are more likely to take damage then dispersed units. Hence I would recommend that a small random inaccuracy is brought in as a common mechanic for most units.
3. The main point of contention is also the most simple. Just make all units tougher. No matter which way you wrangle it, an army with 500 HP (collectively) will take longer to destroy than an army with 50 HP (assuming parity between fire-power and unit count). It is a mathematical certainty that more HP equals longer engagement times.

So the question is, which of these potential dials should we be twiddling? In the long run, the first two are the most interesting, and should ultimately be the main sources of lengthened pacing. However, they are also harder to balance, and may require new mechanics systems be implemented in order to get them running. Hence, why the last and most boring of these is the most critical at this juncture. We don't know what the optimum length of a battle is for this game, only that it is more than it is now, but less than some arbitrary value. This is why we should be taking the time to experiment with different HP multipliers, and seeing how it effects the game. Multiplying all the HP stats in the game by a fixed quantity is not a difficult change. It's the kind of thing that anyone familiar with a scripting language and regexs can do in 20 minutes. And if I could do it in 20 minutes, then I would bet that the coding geniuses at Uber could do it in a matter of seconds. This is why I'd like to see a bit more experimentation now, by twisting that one balance dial back and forth, and seeing if we can come to a broad consensus of where it should rest. This isn't a case of fine tuning, which would obviously be invalidated later as new units are added. In fact this isn't even a case of balancing. It's a case of finding out what we want the balance to be, so we know for later what the other balance factors should be adjusted around.

Mutual annihilation will be solved with more units. However, pacing will not be. Now is the time for fiddling with unit HP values, and seeing what works best, so that we know what to aim for when the more serious changes come along. I hope we can see these kind of broad experiments in balance soon.

This. So very much this.

Fantastic post, MadSci.

where have you guys been hiding? under a rock? I've been fighting these creeps alone!